device and decoding methods to extend frequency bands, and computer-readable storage media

专利摘要:
DECODING DEVICES AND METHODS An apparatus and method of extending frequency bands, an apparatus and method of encoding, an apparatus and method of decoding, and a program, in which such frequency band is extended, and thereby , reproducing music signals with the highest achieved sound quality. A bandpass filter (13) divides an input signal into a plurality of subband signals. A characteristic value calculation circuit (14) uses said plurality of subband signals, as divided, and / or the input signal to calculate a characteristic value. A high frequency subband power estimation circuit (15) calculates, based on the calculated characteristic value, the high frequency subband power estimation values. A high frequency band signal generation circuit (16) generates a high frequency band signal component based on the plurality of subband signals, as divided by the bandpass filter (13), and the values of high frequency subband power estimation calculated by said subband power estimation circuit of (...).
公开号:BR112012007389B1
申请号:R112012007389-3
申请日:2010-09-29
公开日:2020-12-22
发明作者:Yuki Yamamoto；Toru Chinen；Hiroyuki Honma；Yuhki Mitsufuji
申请人:Sony Corporation；
IPC主号:

专利说明:

Technical Field
[001] The present invention concerns a device and also a method for extending frequency bands, a device and a coding method, a device and a decoding method, as well as a program and, even more specifically, the present The invention concerns a device and also a method for extending said frequency bands, a device and an encoding method, a device and a decoding method, as well as a program, whereby the music signals can be played with higher sound quality due to the extension of such frequency bands. Fundamentals of Technique
[002] In more recent years, music distribution services that distribute music data over the Internet, or similar, have become widely employed. With these music distribution services, encoded data that is obtained by encoding music signals is distributed as music data. As a method of encoding music signals, an encoding method that suppresses the capacity of an encoded data file and lowers the bit rate in such a way as to reduce the amount of time taken in the "download" event (DOWNLOAD) have become the main objective.
[003] Such methods of encoding music signals are largely divided into encoding methods, such as MP3 (Audio Layer 3 from “MPEG” Group (Moving Image Experts Group)) (International Standard ISO / IEC 11172-3) and so on, and encoding methods, such as HE-AAC (High Efficiency of MPEG4 AAC) (International Standard ISO / IEC 14496-3) and so on.
[004] With the respective encoding method represented by MP3, components of high frequency bands of the music signal (hereinafter called high frequencies) of approximately 15 kHz or greater that are difficult to be detected by the human ear are eliminated, and the signal components of the remaining low frequency band (hereinafter called low frequencies) are encoded. This type of encoding method will hereinafter be called the high frequency elimination encoding method. With this high frequency elimination coding method, the file capacity of the encrypted data can be suppressed. However, high frequency sounds, while minimally, can be detected by the human ear, so if the sound is generated and emitted from a music signal after decoding which is obtained by decoding the encoded data, deterioration of the quality of the sound quality can occur, such as losing the realistic feeling that the sound has, or the sound becoming muffled.
[005] Conversely, with the encoding method represented by HE-AAC, resource information is extracted from the high frequency components from the components of the high frequency signal, and which are encoded together with the low frequency signal component. This type of encoding method will henceforth be called the high frequency resource encoding method. With the high frequency resource encoding method, only resource information from the components of the high frequency signal is encoded as information relating to the components of the high frequency signal, and thereby, encoding efficiency can be improved while suppressing deterioration of the sound quality.
[006] In decoding the encoded data that was encoded with the high frequency resource coding method, components of the low frequency signal, and resource information are decoded, and components of the high frequency signal are generated from the components the low frequency signal and the resource information after decoding. Accordingly, components of the high frequency signal generated from the components of the low frequency signal, the technique for extending a frequency band of the components of the low frequency signal will hereinafter be called a band expansion technique.
[007] As an example of applying the bandwidth technique, there is post-processing after decoding the data encoded with the high frequency deletion encoding method described above. In this case, the post-processing of the frequency band of the components of the low frequency signal is extended, generating the components of the high frequency signal, lost by coding, from the components of the low frequency signal after decoding (see PTL 1). Note that the method for extending the frequency band in PTL 1 hereinafter will be called the method for extending the PTL 1 band.
[008] With the PTL 1 bandwidth method, a device estimates a high frequency power spectrum (hereinafter called a high frequency envelope, as appropriate) from the power spectrum of the input signal, with the low frequency signal components after decoding as the input signal, and generates high frequency signal components having a high frequency frequency envelope from the low frequency signal components.
[009] Fig. 1 shows an example of the low frequency power spectrum after decoding the input signal and the estimated high frequency envelope.
[0010] In Fig. 1, the vertical axis represents power on a logarithmic scale, and the horizontal axis represents frequency.
[0011] A device determines the end of the low frequency band of the components of the high frequency signal (hereinafter called the start extension band) from the type of encoding format related to the input signal and information such as sample rate , bit rate, and so on (hereafter called secondary information). The device then divides the input signal serving as the components of the low frequency signal into multiple subband signals. The device finds multiple subband signals after splitting, ie an average for each group for a time direction of the power of each of the multiple subband signals on the low frequency side (hereinafter simply called the low frequency side) from the start extension band (hereinafter called group power). As shown in Fig. 1, the device uses the average of the respective group powers of multiple subband signals on the low frequency side as the power, and uses a point where the frequency is a frequency at the bottom edge of the start band extension as the point of origin. The device estimates a linear line at a predetermined slope passing through the point of origin as a frequency envelope on the side with the highest frequency from the start extension band (hereinafter simply called the high frequency side). Note that the positions for the power direction of the origin point can be adjusted by the user. The device generates each of the multiple subband signals on the high frequency side from multiple subband signals on the low frequency side in order to become frequency envelopes on the high frequency side as estimated. The device adds the multiple subband signals generated on the high frequency side to be the components of the high frequency signal, and additionally, adds the components of the low frequency signal and outputs them. Therefore, the music signal after extending the frequency band becomes very close to the original music signal. Consequently, music signals with higher sound quality can be played.
[0012] The band extension method of PTL1 described above has the advantages of being able to extend the frequency bands for music signals after decoding the encoded data of the same, with such encoded data having several high elimination encoding method frequencies and various bit rates. Citation List Patent Literature
[0013] PTL 1: Untested Japanese Patent Application Publication No. 2008-139844 Summary of the invention Technical problem
[0014] However, the band extension method of PTL1 can be improved with respect to the point at which the frequency envelope of the estimated high frequency side is a linear line having a predetermined slope, ie with respect to the point that the shape of a frequency envelope is fixed.
[0015] This is to say, the power spectrum of the music signal has several forms, and depending on the type of music signal, not a few cases will widely vary from the frequency envelope on the high frequency side estimated with the band extension method of PTL1.
[0016] Fig. 2 shows an example of the original power spectrum of a “attack” music signal (“attack” music signal) that accompanies a sudden change in time, such as when a drum is beaten loudly. time, for example.
[0017] Note that Fig. 2 also shows the signal components of the low frequency signal side of the “attack” music signals as input signals, from the band extension method of PTL 1, and the envelope of frequency of the high frequency side estimated from the input signal thereof, together.
[0018] As shown in Fig. 2, the original power spectrum on the high frequency side of the “attack” music signal is approximately flat.
[0019] Conversely, the frequency envelope of the estimated high frequency side has a predetermined negative slope, and even if this is adjusted at the point of origin to a power close to the original power spectrum, the difference from the original power spectrum increases as the frequency increases.
[0020] Therefore, with the band extension method of PTL 1, the frequency envelope on the estimated high frequency side cannot realize the original frequency envelope on the high frequency side with a high degree of accuracy. Consequently, if sound is generated and emitted from the music signal after extending the frequency band, clarity of sound may be lost when compared to the original sound, from a listening perspective.
[0021] In addition, with a high frequency characteristic encoding method such as HE-AAC or the like as described above, the frequency envelope on the high frequency side is used as characteristic information for the components of the high frequency signal to be encoded, but the decoding side is required to reproduce the original frequency envelope on the frequency side in a precisely accurate manner
[0022] The present invention was made taking such situations into consideration, and allows music signals to be performed with high sound quality due to the extension of the frequency bands. Solution to the problem
[0023] A frequency band extension device according to a first aspect of the present invention includes: signal splitting means configured to divide an input signal into multiple subband signals; characteristic value calculation means configured to calculate the characteristic value expressing a characteristic of the input signal using at least one of the multiple subband signals divided by the signal splitting means, and the input signal; high frequency subband power estimation means configured to calculate an estimated value of a high frequency subband power which is the power of a subband signal having a higher frequency band than the entry based on the characteristic value calculated by means of characteristic value calculation; and high frequency signal component generation means configured to generate a high frequency signal component based on the multiple subband signals divided by the signal splitting means, and the estimated value of the high subband power frequency calculated by means of high frequency subband power estimation; with the frequency band of the input signal being extended using the high frequency signal component generated by the high frequency signal component generation means.
[0024] The means of calculating the characteristic value can calculate a low frequency subband power which is a power of the multiple subband signals according to the characteristic value.
[0025] The means of calculating the characteristic value can calculate a time variation of a low frequency subband power which is a power of the multiple subband signals according to the characteristic value.
[0026] The means of calculating the characteristic value can calculate the difference between the maximum and minimum powers in a predetermined frequency band, of the input signal, according to the characteristic value.
[0027] The means of calculating the characteristic value can calculate a temporal variation of the difference between the maximum and minimum value of the power in a predetermined frequency band, of the input signal, according to the characteristic value.
[0028] The means of calculating the characteristic value can calculate the slope of a power in a predetermined frequency band, of the input signal, according to the characteristic value.
[0029] The means of calculating the characteristic value can calculate a temporal variation of the slope of a power in a predetermined frequency band, of the input signal, according to the characteristic value.
[0030] High frequency subband power estimation means can calculate an estimated high frequency subband power value based on the characteristic value, and a coefficient for each high frequency subband obtained advance by learning.
[0031] The coefficient for each high frequency subband can be generated by agglomerating the residual vector of the high frequency signal component calculated with the coefficient for each high frequency subband obtained through regression analysis with multiple signals. learning, and performing regression analysis, for each cluster obtained through the cluster, using the learning signals belonging to the cluster.
[0032] The residual vector can be normalized with the dispersion value of each component of the multiple residual vectors, and the vector after normalization can be subjected to agglomeration.
[0033] The high frequency subband power estimation means can calculate an estimated high frequency subband power value based on the characteristic value, and the coefficient and constant for each of the subband bands. high frequency; with the constant being calculated from a center of gravity vector for the new clusters obtained, still calculating the residual vector using the coefficient for each high frequency subband obtained through regression analysis with the learning signals belonging to the cluster, and agglomerating the residual vector of the same for multiple new clusters.
[0034] The high frequency subband power estimation means can record the coefficient for each of the high frequency subbands, and an indicator that determines the coefficient for each high frequency subband, in a manner correlated, and also record multiple sets of the indicator and the constant, and some of the multiple sets can include an indicator having the same value.
[0035] The high frequency signal generating means can generate the high frequency signal component from a low frequency subband power which is a power of the multiple subband signals, and an estimated value of high frequency subband power.
[0036] A frequency band extension method according to the first aspect of the present invention includes: a signal splitting step arranged to divide an input signal into multiple subband signals; an arranged characteristic value calculation step for calculating the characteristic value expressing a characteristic of the input signal using at least one of the multiple subband signals divided by processing in the signal splitting step, and the input signal; a high frequency subband power estimation step arranged to calculate an estimated high frequency subband power ie the power of the subband signal having a higher frequency band than the entry based on the characteristic value calculated by processing in the characteristic value calculation step; and a step of generating a high frequency signal component arranged to generate a high frequency signal component based on the multiple subband signals divided by processing in the signal splitting step, and the estimated value of the subband power. high frequency band calculated by processing in the high frequency subband power estimation step; with the frequency band of the input signal being extended using the high frequency signal component generated by processing in the high frequency signal component generation step.
[0037] A program according to the first aspect of the present invention includes: a signal splitting step arranged to divide an input signal into multiple subband signals; an arranged characteristic value calculation step for calculating the characteristic value expressing a characteristic of the input signal using at least one of the multiple subband signals divided by processing in the signal splitting step, and the input signal; an arranged high frequency subband power estimation step to calculate an estimated high frequency subband power which is the power of a subband signal having a higher frequency band than the signal input based on the characteristic value calculated by processing in the characteristic value calculation step; and a high frequency signal component generation step arranged to generate a high frequency signal component based on the multiple subband signals divided by processing in the signal splitting step, and the estimated value of the subband power high frequency band calculated by processing in the high frequency subband power estimation step; causing a computer to perform processing to extend the frequency band of the input signal using the high frequency signal component generated by processing in the high frequency signal component generation step.
[0038] According to the first aspect of the present invention, an input signal is divided into multiple subband signals, a characteristic value expressing a characteristic of the input signal is calculated with at least one of the multiple subband signals split and the input signal, an estimated value of a high frequency subband power which is the power of a subband signal having a higher frequency band than the input signal is calculated based on the value of calculated characteristic, a high frequency signal component is generated based on multiple divided subband signals, and the estimated value of the calculated high frequency subband power, and the frequency band of the input signal is generated with the high frequency signal component generated.
[0039] A coding device according to a second aspect of the present invention includes: subband splitting means configured to divide an input signal into multiple subbands, and to generate a low frequency subband signal composed of multiple multiple subbands on one low frequency side and a high frequency subband signal composed of multiple subbands on one high frequency side; characteristic value calculation means configured to calculate the characteristic value expressing a characteristic of the input signal, using at least one of the low frequency subband signal generated by the subband splitting means, and the input; high frequency subband pseudopotency calculation means configured to calculate a high frequency subband pseudopotency which is a pseudopotency of the high frequency subband signal based on the characteristic value calculated by the value calculation means characteristic; high frequency subband pseudo-power difference means for calculating a high frequency subband power which is the power of the high frequency subband signal from the high frequency subband signal generated by the subband splitting means, and to calculate difference of high frequency subband pseudopotency which is difference in the high frequency subband pseudopotence calculated by the means of calculating high frequency subband pseudopotency ; high frequency encoding means configured to encode the high frequency subband pseudo-power difference calculated by the high frequency subband pseudo-power difference calculating means to generate high frequency encoded data; low frequency encoding means configured to encode a low frequency signal which is a low frequency signal from the input signal to generate low frequency encoded data; and multiplexing means configured to multiplex the low frequency encoded data generated by the low frequency encoding means, and the high frequency encoded data generated by the high frequency encoding means to obtain an output code sequence.
[0040] The encoding device may further include low frequency decoding means configured to decode low frequency encoded data generated by the low frequency encoding means to generate a low frequency signal; with the subband splitting means generating the low frequency subband signal from the low frequency signal generated by the low frequency decoding means.
[0041] High frequency encoding means can calculate similarity between the high frequency subband pseudopotency difference, and a representative vector or representative value in predetermined plurality of high frequency subband pseudopotence difference space for generate an index corresponding to a representative vector or representative value of which the similarity is the maximum, such as high frequency coded data.
[0042] The means of calculating the difference of high frequency subband pseudopotency can calculate a value evaluated based on the high frequency subband pseudopotency of each subband, and the high frequency subband power for each multiple coefficients to calculate the high frequency subband pseudopotency; with the high frequency coding means generating an index indicating the coefficient of the evaluated value which is the highest evaluated value, like the high frequency coded data.
[0043] The means of calculating the high frequency subband pseudo-power difference can calculate the value evaluated based on at least any of the sum of the squares of the high frequency subband pseudo-power difference of each subband, the maximum value of the absolute value of the high frequency subband pseudo-power, or the average value of the difference of high frequency subband pseudo-power of each subband.
[0044] The means of calculating the difference in high frequency sub-band pseudo-power can calculate the value evaluated based on the difference in high-frequency sub-band pseudo-power in different frames.
[0045] The high frequency subband pseudo-power difference calculation means can calculate the assessed value using the high frequency subband pseudo-power difference multiplies by the weighting factor which is the weighting factor for each sub- band such that the lower the frequency side that the subband is, the greater the weighting factor of the subband is.
[0046] The means for calculating the high frequency subband pseudo-power difference can calculate the mentioned value mentioned using the mentioned high frequency subband pseudo-power difference multiplied by the weighting factor which is the weighting factor for each subband such that the higher the power of the mentioned high frequency subband of the subband is, the greater the weighting factor of the subband is.
[0047] A coding method according to a second aspect of the present invention includes: a subband splitting step arranged to divide an input signal into a plurality of subbands, and to generate a subband signal low-frequency composed of multiple sub-bands on one low-frequency side and a high-frequency sub-band signal composed of multiple sub-bands on one high-frequency side; an arranged characteristic value calculation step to calculate the characteristic value expressing a characteristic of the input signal, using at least one of the low frequency subband signal generated by processing in the subband splitting step, and the input signal; an arranged high frequency subband pseudopotency calculation step to calculate a high frequency subband pseudopotency which is a pseudopotency of the high frequency subband signal based on the characteristic value calculated by processing in the calculation of characteristic value; a step of calculating the high frequency subband pseudo-power difference arranged to calculate a high frequency subband power which is the power of the high frequency subband signal from the high subband signal frequency generated by processing in the subband division step, and to calculate difference in high frequency subband pseudopotency which is the difference in the mentioned high frequency subband pseudopotence calculated by processing in the pseudopotence calculation step high frequency subband; a high frequency coding step arranged to encode the high frequency subband pseudo-power difference calculated by processing in the high frequency subband pseudo-power difference calculation step to generate high frequency encoded data; a low frequency encoding step arranged to encode a low frequency signal which is a low frequency signal from the input signal to generate low frequency encoded data; and a multiplexing step arranged to multiplex the low frequency coded data generated by processing in the low frequency coding step, and the high frequency coded data generated by processing in the high frequency coding step to obtain an output code sequence. .
[0048] A program according to the second aspect causing the computer to perform processing including: a sub-band splitting step arranged to divide an input signal into a plurality of sub-bands, and to generate a sub signal - low-frequency band composed of multiple sub-bands on one low-frequency side and a high-frequency sub-band signal composed of multiple sub-bands on one high-frequency side; an arranged characteristic value calculation step to calculate the characteristic value expressing a characteristic of the input signal, using at least one of the low frequency subband signal generated by processing in the subband splitting step, and of the input signal; a step of calculating high frequency subband pseudo-power arranged to calculate a high frequency subband pseudo-power which is a pseudo-power of the high frequency subband signal based on the calculated characteristic value by processing in the characteristic value calculation step; a step of calculating the high frequency subband pseudo-power difference arranged to calculate a high frequency subband power which is the power of the high frequency subband signal from the high subband signal frequency generated by processing in the sub-band split step, and to calculate difference in high-frequency sub-band pseudo-power which is the difference in high-frequency sub-band pseudo-power calculated by processing in the pseudo-power calculation step of high frequency subband; a high frequency coding step arranged to encode the high frequency subband pseudo-power difference calculated by processing in the high frequency subband pseudo-power difference calculation step to generate high frequency encoded data; a low frequency encoding step arranged to encode a low frequency signal which is a low frequency signal from the input signal to generate low frequency encoded data; and multiplexing step arranged to multiplex the low frequency encoded data generated by processing in the low frequency encoding step, and the high frequency encoded data generated by processing in the high frequency encoding step to obtain an output code sequence.
[0049] With the second aspect of the present invention, an input signal is divided into multiple sub-bands, a low-frequency sub-band signal composed of multiple sub-bands on one low-frequency side and a sub-signal high frequency band composed of multiple sub-bands on a high frequency side are generated, a characteristic value expressing a characteristic of the input signal is calculated with at least one of the generated low frequency subband signal and the signal input, a high frequency subband pseudo-power that is a high frequency subband signal pseudo-power is calculated based on the calculated characteristic value, a high frequency subband power which is the signal strength high-frequency sub-band signal is calculated from the generated high-frequency sub-band signal, a difference of high-frequency sub-band pseudo-power which is a difference in terms of the high-frequency sub-band pseudo-power calculated, the calculated high frequency subband pseudo-power difference is encoded to generate high frequency encoded data, a low frequency signal that is a low frequency signal from the input signal is encoded to generate low encoded data frequency, and the generated low frequency encoded data and multiplexed the generated high frequency encoded data to obtain an output code sequence.
A decoding device according to a third aspect of the present invention includes demultiplexing means configured to demultiplex the encoded data entered into at least low frequency encoded data and an index; low frequency decoding means configured to decode low frequency encoded data to generate a low frequency signal; subband splitting means configured to divide the low frequency signal band into multiple low frequency subbands to generate a low frequency subband signal for each of the low frequency subbands; and generation means configured to generate the high frequency signal based on the index and the low frequency subband signal.
[0051] The index can be obtained, in a device that encodes an input signal and emits the encoded data, based on the input signal before encoding, and the high frequency signal estimated from the input signal.
[0052] The index may not have been encrypted.
[0053] The index can be information indicating an estimate coefficient used to generate the high frequency signal.
[0054] The means of generation can generate the high frequency signal based on the multiple estimation coefficients, the estimate coefficient indicated by the index.
[0055] The generation means may include, characteristic value calculation means configured to calculate the characteristic value expressing a characteristic of the encoded data using at least one of the low frequency subband signal and the low frequency signal ; high frequency subband power calculation means configured to calculate the high frequency subband power of a high frequency subband signal from the high frequency subband by calculating using the characteristic value and the estimation coefficient considering each of the multiple high frequency sub-bands composing the high frequency signal band; and high frequency signal generating means configured to generate the high frequency signal based on the high frequency subband power and the low frequency subband signal.
[0056] The high frequency subband power calculation means can calculate the mentioned high frequency subband power of the high frequency subband by linearly combining a plurality of characteristic values using the estimate coefficient prepared for each of the high frequency sub-bands.
[0057] The means of calculating the characteristic value can calculate the low frequency subband power of the low frequency subband signal for each of the low frequency subbands according to the characteristic value.
[0058] The index can be information indicating the estimation coefficient and thereby the high frequency subband power that most closely matches the high frequency subband power obtained from the high frequency signal of the signal input before encoding is obtained as a result of the comparison between the high frequency subband power obtained from the high frequency signal of the input signal before encoding and the high frequency subband power generated on the basis of in the estimate coefficient of the multiple estimate coefficients.
[0059] The index can be information indicating the estimate coefficient and thereby the sum of the squares of the difference between the high frequency subband power obtained from the high frequency signal of the input signal before encoding, and the high frequency subband power generated based on the estimate coefficient obtained for each of the high frequency subbands, becomes minimal.
[0060] The encoded data may also include difference information indicating difference between the high frequency subband power obtained from the high frequency signal of the input signal before encoding, and the high frequency subband power. generated based on the estimate coefficient. The difference information may have been encoded.
[0061] The high frequency subband power calculation means can add the difference indicated with the difference information included in the coded data for the high frequency subband power obtained through calculation using the characteristic value and the estimate coefficient; with the high frequency signal generation means generating the high frequency signal based on the high frequency subband power to which the difference has been added, and the low frequency subband signal.
[0062] The estimate coefficient can be obtained through regression analysis using the least square method with the characteristic value as an explanatory variable and the high frequency subband power as an explained variable.
[0063] The decoding device can also include, with the index being information indicating a difference vector composed of the difference for each of the high frequency sub-bands characterized by the fact that the difference between the power of the high sub-band frequency obtained from the high frequency signal of the input signal before encoding, and the high frequency subband power generated based on the estimation coefficient as an element, coefficient emission means configured to obtain the distance between a representative vector or representative value in the difference characteristic space with the difference of the high frequency sub-bands as an element, obtained in advance for each of the estimation coefficients, and the difference vector indicated by the index, and to provide the coefficient of estimate of the representative vector or the representative value and thereby the distance is the shortest of the multiple estimation coefficients s, for the means of calculating high frequency subband power.
[0064] The index can be information indicating the estimate coefficient of a plurality of estimate coefficients and thereby as a result of comparison between the high frequency signal of the input signal before encoding, and the high frequency signal generated based on the estimation coefficient, the high frequency signal closest to the high frequency signal of the input signal before encoding is obtained.
[0065] The estimate coefficient can be obtained through regression analysis.
[0066] The generation means can generate the high frequency signal based on the information obtained by decoding the coded index.
[0067] The index may have been subjected to entropy coding.
[0068] A decoding method or program according to the third aspect includes: a demultiplexing step arranged to demultiplex the encoded data entered into at least low frequency encoded data and an index; a low frequency decoding step arranged to decode the low frequency encoded data to generate a low frequency signal; a subband splitting step arranged to divide the low frequency signal band into multiple low frequency subbands to generate a low frequency subband signal for each of the low frequency subbands; and a generation step arranged to generate the high frequency signal based on the index and the low frequency subband signal.
[0069] With the third aspect of the present invention, in the entered encoded data is demultiplexed transforming at least low frequency encoded data and an index, the low frequency encoded data is decoded to generate a low frequency signal, the signal band low frequency subband is divided into multiple low frequency subbands to generate a low frequency subband signal for each of the low frequency subbands, and the high frequency signal is generated based on the index and the signal low frequency subband.
[0070] A decoding device according to a fourth aspect of the present invention includes: demultiplexing means configured to demultiplex the encoded data entered into the low frequency encoded data and an index to obtain an estimate coefficient used to generate a signal. high frequency; low frequency decoding means configured to decode low frequency encoded data to generate a low frequency signal; subband splitting means configured to divide the low frequency signal band into multiple low frequency subbands to generate a low frequency subband signal for each of the low frequency subbands; characteristic value calculation means configured to calculate the characteristic value expressing a characteristic of the encoded data using at least one of the low frequency subband signal and the low frequency signal; high frequency subband power calculation means configured to calculate the high frequency subband power of the high frequency subband signal of the high frequency subband by undoing the multiplexing of the characteristic value by the coefficient of estimate determined by the index of the multiple coefficients of estimates prepared in advance considering each of the multiple high frequency sub-bands composing the high frequency signal band, and obtaining the sum of the characteristic value by which the estimate coefficient was multiplied; and high frequency signal generating means configured to generate the high frequency signal using the high frequency subband power and the low frequency subband signal.
[0071] The means of calculating the characteristic value can calculate the low frequency subband power of the low frequency subband signal for each of the low frequency subbands according to the characteristic value.
[0072] The index can be information to obtain the estimate coefficient of the multiple estimate coefficients and thereby the sum of the squares of the difference obtained for each of the high frequency sub-bands, which is the difference between the power of the sub -high frequency band obtained from the real value of the high frequency signal, and the high frequency subband power generated with the estimation coefficient, becomes the minimum.
[0073] The index may also include difference information indicating difference between the high frequency subband power obtained from the real value, and the high frequency subband power generated with the estimate coefficient; with the means of calculating high frequency subband power still adding the difference indicated by the difference information included in the index for the high frequency subband power obtained by obtaining the sum of the characteristic value by which the estimation coefficient was multiplied, and characterized by the fact that the high frequency signal generation means generate the high frequency signal using the high frequency subband power to which the difference was added by the sub frequency power calculation means. high frequency band, and the low frequency subband signal.
[0074] The index can be information indicating the estimate coefficient.
[0075] The index can be information obtained by the information indicating the coefficient of estimate being submitted to the entropy coding; with the high frequency subband power calculation means by calculating the high frequency subband power using the estimate coefficient indicated by the information obtained by decoding the index.
[0076] The multiple estimation coefficients can be obtained in advance through regression analysis using the least square method with the characteristic value as an explanatory variable and the high frequency subband power as an explained variable.
[0077] The decoding device can also include, with the index being information indicating a difference vector composed of the difference for each of the high frequency sub-bands characterized by the fact that the difference between the power of the high sub-band frequency obtained from the actual value of the high frequency signal, and the high frequency subband power generated with the estimate coefficient as an element, coefficient emission means configured to obtain the distance between a representative vector or representative value in the difference characteristic space with the difference of the high frequency sub-bands as an element, obtained in advance for each of the estimation coefficients, and the difference vector indicated by the index, and to provide the estimation coefficient of the representative vector or the representative value and thereby the distance is the shortest of the multiple estimation coefficients for the means of calculating pot high frequency subband.
[0078] A decoding method or program according to the fourth aspect of the present invention includes: a demultiplexing step arranged to undo the multiplexing of the encoded data entered into low frequency encoded data and an index to obtain an estimate coefficient used for generation of a high frequency signal; a low frequency decoding step arranged to decode the low frequency encoded data to generate a low frequency signal; a subband splitting step arranged to divide the low frequency signal band into multiple low frequency subbands to generate a low frequency subband signal for each of the low frequency subbands; an arranged characteristic value calculation step for calculating the characteristic value expressing a characteristic of the encoded data using at least one of the low frequency subband signal and the low frequency signal; a high frequency subband power calculation step arranged to calculate the high frequency subband power of the high frequency subband signal of the high frequency subband by multiplexing the characteristic value by the coefficient of estimation determined by the index of the multiple estimate coefficients prepared in advance considering each of the multiple high frequency sub-bands composing the high frequency signal band, and obtaining the sum of the characteristic value by which the estimate coefficient was multiplied; and a high frequency signal generation step arranged to generate the high frequency signal using the high frequency subband power and the low frequency subband signal.
[0079] With the fourth aspect of the present invention, it is demultiplexed from the encoded data entered and transformed into low frequency encoded data and an index to obtain an estimate coefficient used to generate a high frequency signal, the low frequency encoded data are decoded to generate a low frequency signal, the low frequency signal band is divided into multiple low frequency subbands to generate a low frequency subband signal for each of the low frequency subbands, the characteristic value expressing a characteristic of the encoded data is calculated with at least one of the low frequency subband signal and the low frequency signal, a high frequency subband power of the high frequency subband signal of the high frequency subband is calculated by multiplying the characteristic value by the estimation coefficient determined by the index of the multiple coefficients of est immative prepared in advance considering each of the multiple high frequency sub-bands composing the high frequency signal band, and obtain the sum of the characteristic value by which the estimation coefficient was multiplied, and the high frequency signal is generated with the high frequency subband power and the low frequency subband signal. Advantageous Effects of the Invention
[0080] According to the first aspect through the fourth aspect of the present invention, music signals can be performed with higher sound quality due to the extension of the frequency bands. Brief Description of the Drawings
[0081] [Fig. 1] Fig. 1 is a diagram illustrating an example of a low frequency power spectrum after decoding, serving as an input signal, and an estimated high frequency envelope.
[0082] [Fig. 2] Fig. 2 is a diagram illustrating an example of an original power spectrum of an “attack” music signal that accompanies a temporary sudden change.
[0083] [Fig. 3] Fig. 3 is a block diagram illustrating a functional example of configuring a frequency band extension device according to the first embodiment of the present invention.
[0084] [Fig. 4] Fig. 4 is a flowchart describing an example of frequency band extension processing by a frequency band extension device in Fig. 3.
[0085] [Fig. 5] Fig. 5 is a diagram illustrating the power spectrum of the signal input to a frequency band extension device in Fig. 3 and the positioning on a frequency axis of the bandpass filter.
[0086] [Fig. 6] Fig. 6 is a diagram illustrating an example of a frequency characteristic of a vocal segment and the estimated high frequency power spectrum.
[0087] [Fig. 7] Fig. 7 is a diagram illustrating an example of the signal power spectrum entered into a frequency band extension device in Fig. 3.
[0088] [Fig. 8] Fig. 8 is a diagram illustrating an example of a power spectrum after raising the input signal in Fig. 7.
[0089] [Fig. 9] Fig. 9 is a block diagram illustrating a functional example of configuring a coefficient learning device to effect learning coefficients used in a high frequency signal generation circuit of a frequency band extension device in Fig 3.
[0090] [Fig. 10] Fig. 10 is a flowchart describing an example of processing coefficient learning by the coefficient learning device in Fig. 9.
[0091] [Fig. 11] Fig. 11 is a block diagram illustrating a functional example of configuring a coding device according to a second embodiment of the present invention.
[0092] [Fig. 12] Fig. 12 is a flow chart describing an example of coding processing by the coding device in Fig. 11.
[0093] [Fig. 13] Fig. 13 is a block diagram illustrating a functional example of configuration of the decoding device according to the second embodiment of the present invention.
[0094] [Fig. 14] Fig. 14 is a flow chart describing an example of decoding processing by the decoding device in Fig. 13.
[0095] [Fig. 15] Fig. 15 is a block diagram illustrating a functional example of configuring a coefficient learning device for learning representative vectors used in the high frequency coding circuit of the coding device in Fig. 11 and estimation coefficients decoded high-frequency subband power used in the high-frequency decoding circuit of the decoding device in Fig. 13.
[0096] [Fig. 16] Fig. 16 is a flowchart describing an example of processing coefficient learning by the coefficient learning device in Fig. 15.
[0097] [Fig. 17] Fig. 17 is a diagram illustrating an example of a code sequence emitted by the coding device in Fig. 11.
[0098] [Fig. 18] Fig. 18 is a block diagram illustrating a functional example of configuring an encoding device.
[0099] [Fig. 19] Fig. 19 is a flow chart describing coding processing.
[00100] [Fig. 20] Fig. 20 is a block diagram illustrating a functional example of configuring a decoding device.
[00101] [Fig. 21] Fig. 21 is a flow chart describing the decoding processing.
[00102] [Fig. 22] Fig. 22 is a flow chart describing the coding processing.
[00103] [Fig. 23] Fig. 23 is a flow chart describing the decoding processing.
[00104] [Fig. 24] Fig. 24 is a flow chart describing the coding processing.
[00105] [Fig. 25] Fig. 25 is a flow chart describing the coding processing.
[00106] [Fig. 26] Fig. 26 is a flow chart describing coding processing.
[00107] [Fig. 27] Fig. 27 is a flow chart describing the coding processing.
[00108] [Fig. 28] Fig. 28 is a diagram illustrating an example of setting up a coefficient learning device.
[00109] [Fig. 29] Fig. 29 is a flowchart describing coefficient learning processing.
[00110] [Fig. 30] Fig. 30 is a block diagram illustrating an example of computer hardware configuration that performs processing for which the present invention has been applied, through a program. Description of Modalities
[00111] Modalities of the present invention will be described with reference to the attached diagrams. Note that the description will be in the following order. 1. First modality (in the case of applying the present invention to a frequency band extension device) 2. Second modality (in the case of applying the present invention to a coding device and decoding device) 3. Third modality (in the case of including coefficient index in high-frequency coded data) 4. Fourth modality (in case of including coefficient index and high-frequency sub-band pseudo-power difference in high-frequency coded data) 5. Fifth modality (in the case to select a coefficient index using an evaluated value) 6. Sixth modality (in case of sharing a portion of coefficients) <1. First modality>
[00112] According to the first modality, processing to extend a frequency band (hereinafter called frequency band extension processing) is performed for the components of the low frequency signal after decoding that are obtained by decoding data encoded with a high frequency elimination coding method. [Functional example of configuring a frequency band extension device]
[00113] Fig. 3 shows a functional example of configuring a frequency band extension device to which the present invention is applied.
[00114] With the low frequency signal components after decoding as an input signal, a frequency band extension device 10 performs frequency band extension processing for the input signal thereof, and outputs the signal after processing frequency band length obtained as a result of it as an output signal.
[00115] A frequency band extension device 10 consists of a low pass filter 11, a delay circuit 12, a band pass filter 13, a characteristic value calculation circuit 14, a power estimation circuit of frequency subband 15, a high frequency signal generation circuit 16, a high pass filter 17, and a signal addition unit 18.
[00116] The low-pass filter 11 filters the input signal with a predetermined cutoff frequency, and supplies the components of the low frequency signal that are components of a low frequency signal to the delay circuit 12 with a post signal -filtration.
[00117] In order to synchronize in the event of adding together the low frequency signal components from the low pass filter 11 and the high frequency signal components to be described later, the delay circuit 12 delays the signal components low frequency for a certain amount of delay time and then supply to the signal addition unit 18.
[00118] The bandpass filter 13 is composed of bandpass filters 13-1 to 13-N that each has different bandwidths. The bandpass filter 13-i (1 <i <N) allows a predetermined bandpass signal from the input signal to pass, and as one of the multiple subband signals, it supplies this to the value calculation circuit feature 14 and high frequency signal generation circuit 16.
[00119] The characteristic value calculation circuit 14 uses at least one of the multiple subband signals from the bandpass filter 13 and the input signal to calculate one or multiple characteristic values, and supplies it to the circuit of frequency subband power estimation 15. The characteristic value is now information indicating a signal characteristic of the input signal.
[00120] The frequency subband power estimation circuit 15 calculates an estimated value of a high frequency subband power which is a power of a high frequency subband signal, for each subband high frequency, based on one or multiple characteristic values from the characteristic value calculation circuit 14, and supplies them to the high frequency signal generation circuit 16.
[00121] The high frequency signal generation circuit 16 generates high frequency signal components that are high frequency signal components, based on the multiple subband signals from the bandpass filter 13 and the estimated values of multiple sub-band powers from the sub-band power estimation circuit of frequency 15, and supplies them to the high-pass filter 17.
[00122] The high pass filter 17 filters the high frequency signal components from the high frequency signal generation circuit 16 with the cutoff frequency corresponding to the cutoff frequency in the lowpass filter 11, and supplies it to the unit plus sign 18.
[00123] The signal addition unit 18 adds a low frequency signal component of the delay circuit 12 and a high frequency signal component of the high-pass filter 17, and outputs them as the output signal.
[00124] Note that according to the configuration in Fig. 3, the bandpass filter 13 is used to obtain a subband signal, but the configuration is not restricted to this, and, for example, a splitting filter band as disclosed in PTL 1 can be used.
[00125] Furthermore, similarly, according to the configuration in Fig. 3, the signal addition unit 18 is used to synthesize the subband signals, but the configuration is not restricted to this, and, for example, a band synthesis filter as disclosed in PTL 1 can be used. [Frequency Band Extension Processing of the Frequency Band Extension Device]
[00126] Next, a frequency band extension processing with a frequency band extension device in Fig. 3 will be described with reference to the flowchart in Fig. 4.
[00127] In step S1, the low pass filter 11 filters the input signal with a predetermined cutoff frequency, and supplies the low frequency signal component serving as a post-filter signal for the delay circuit 12.
[00128] The low pass filter 11 can configure an optional frequency as the cutoff frequency, but according to the present modality, with a predetermined band as the start extension band to be described later, the cutoff frequency is configured corresponding to a frequency of the lower end of the extension start band. Consequently, the low-pass filter 11 supplies the low frequency signal components to the delay circuit 12, which are signal components of a lower band than the start-extension band, with the post-filtering signal.
[00129] In addition, the low pass filter 11 can also configure an optimal frequency as the cutoff frequency, according to the encoding parameters such as the high frequency and bit rate elimination encoding method and so on. input signal. The secondary information used by the band extension method in PTL 1, for example, can be used as the encoding parameter.
[00130] In step S2, the delay circuit 12 delays the components of the low frequency signal from the low pass filter 11 by only a certain amount of delay time, and supplies them to the signal addition unit 18.
[00131] In step S3, the bandpass filter 13 (bandpass filters 13-1 to 13-N) divides the input signal into multiple subband signals, and supplies each post by dividing multiple subband signals band for a characteristic value calculation circuit 14 and a high frequency signal generation circuit 16. Note that details of the processing for splitting the input signal with the bandpass filter 13 will be described later.
[00132] In step S4, the characteristic value calculation circuit 14 uses at least one of the multiple subband signals from the bandpass filter 13 and the input signal to calculate one or multiple characteristic values, and the supply for the frequency subband power estimation circuit 15. Note that the details of the processing to calculate the characteristic value with the characteristic value calculation circuit 14 will be described later.
[00133] In step S5, the frequency subband power estimation circuit 15 calculates estimated values of the multiple high frequency subband powers, based on one or multiple characteristic values from the calculation circuit of characteristic value 14, and supplies them for the high frequency signal generation circuit 16. Note that processing details to calculate the estimated high frequency subband power values with the subband power estimation circuit frequency 15 will be described later.
[00134] In step S6, the high frequency signal generation circuit 16 generates components of the high frequency signal, based on the multiple subband signals from the bandpass filter 13 and the estimated values of the multiple sub powers -high frequency band from the frequency subband power estimation circuit 15, and supplies them to the high pass filter 17. The components of the high frequency signal here are signal components of a higher band than the extension start band. Note that details of the processing to generate the high frequency signal components with the high frequency signal generation circuit 16 will be described later.
[00135] In step S7, the high pass filter 17 filters the components of the high frequency signal from the high frequency signal generation circuit 16, and thereby removing noise from repeated components for the low frequency included in the components of the high frequency signal, and the like, and supplies the components of the high frequency signal to the signal addition unit 18.
[00136] In step S8, the signal addition unit 18 adds the low frequency signal components from the delay circuit 12 and the high frequency signal components from the high pass filter 17, and emits them as one exit sign.
[00137] According to the above processing, the frequency band can be extended as to the components of the low-frequency signal from post-decoding after decoding.
[00138] Next, details of the processing for each of the steps S3 to S6 in the flowchart in Fig. 4 will be described. [Processing Details by Bandpass Filter]
[00139] First, details of the processing by the bandpass filter 13 in step S3 of the flowchart in Fig. 4 will be described.
[00140] Note that for ease of description, Hereafter, the number N of bandpass filters 13 will be N = 4.
[00141] For example, one of the 16 sub-bands obtained by dividing the Nyquist frequency of the input signal into 16 equal parts can be configured as the beginning extension band, and of the 16 sub-bands, each of 4 sub-bands of the lower band than the beginning extension band are configured as bandpass bands for bandpass filters 13-1 to 13-4, respectively.
[00142] Fig. 5 shows the position of each of the pass bands of the band pass filters 13-1 to 13-4 on an axis of frequency axis of each.
[00143] As shown in Fig. 5, if the first subband index from the high frequency of a frequency band (subband) which is a lower band than the start extension band is represented as sb, and second subband index as sb-1, and I'th subband index as sb- (I-1), each of the bandpass filters 13-1 to 13-4 are assigned to be pass bands for each of the sub-bands having an index from sb to sb-3, outside the sub-bands lower than the beginning extension band.
[00144] Note that according to the present modality, each of the bands of the filters passing band 13-1 to 13-4 are described as being the four predetermined out of the 16 sub-bands obtained by dividing the Nyquist frequency of the signal input bands in 16 equal parts, but not restricted to this, the pass bands can be the four predetermined out of the 256 subbands obtained by dividing the Nyquist frequency of the input signal into 256 equal parts. In addition, the bandwidth of each of the 13-1 to 13-4 bandpass filters can each be different. [Details of Processing by the Characteristic Value Calculation Circuit]
[00145] Next, details of the processing by the characteristic value calculation circuit 14 in step S4 of the flowchart in Fig. 4 will be described.
[00146] The characteristic value calculation circuit 14 uses at least one of the multiple subband signals from the bandpass filter 13 and the input signal, and calculates one or multiple characteristic values that the estimation circuit of frequency subband power 15 uses to calculate the high frequency subband power estimate values.
[00147] More specifically, the characteristic value calculation circuit 14 calculates, as characteristic values, the subband signal power (subband power (hereinafter, also called low subband power) frequency)) for each subband, of the four subband signals from the bandpass filter 13, and supplies them to the frequency subband power estimation circuit 15.
[00148] This is to say, the characteristic value calculation circuit 14 finds the low frequency subband power in a given predetermined time frame, called power (ib, J), of the four sub signals -band x (ib, n) supplied from the bandpass filter 13, with Expression (1) below. Here, ib represents the subband index and n represents the dispersion time index. Note that the sample size of a frame is FSIZE and the power is expressed in decibels. [Expression 1]

[00149] Therefore, the low frequency subband power, power (ib, J), found with the characteristic value calculation circuit 14, is provided with a characteristic value for the power estimation circuit of frequency subband 15. [Details of Processing with the Frequency Subband Power Estimation Circuit]
[00150] Next, details of the processing with the frequency subband power estimation circuit 15 in step S5 of the flowchart in Fig. 4 will be described.
[00151] The frequency subband power estimation circuit 15 calculates the estimated value of the subband power (high frequency subband power) of the band to be extended (frequency extension band) in addition to the subband whose index is sb + 1 (extension start band), based on the four subband powers provided from the characteristic value calculation circuit 14.
[00152] This is to say, if we say that the subband index of the highest band in a frequency extension band is eb, the frequency subband power estimation circuit 15 estimates (eb-sb ) subband power numbers for subbands characterized by the fact that the index is sb + 1 à and b.
[00153] The estimated value of the subband power in a frequency extension band characterized by the fact that the index is ib, potency (ib, J), uses the four subband powers, power (ib, j), supplied from the characteristic value calculation circuit 14, and can be expressed with Expression (2) below, for example. [Expression 2]

[00154] Now, in Expression (2), the coefficients Aib (kb) and Bib are coefficients having values that differ for each ib subband. The coefficients Aib (kb) and Bib are coefficients configured appropriately such that favorable values can be obtained for the various input signals. In addition, the coefficients Aib (kb) and Bib are changed to optimal values by changing the subband sb. Note that production of the coefficients Aib (kb) and Bib will be described later.
[00155] In Expression (2), the high frequency subband power estimate values are calculated with the linear combination using the power for each of the multiple subband signals from the bandpass filter 13, but the arrangement is not restricted to this, and for example, calculation can be performed using a linear combination of multiple low-frequency subband powers of several frames before and after a J time frame, or using non-linear functions.
[00156] Therefore, the high frequency subband power estimate values calculated with the frequency subband power estimate circuit 15 is provided for the high frequency signal generation circuit 16. [Details of Processing by the High Frequency Signal Generation Circuit]
[00157] Next, details of processing by the high frequency signal generation circuit 16 in step S6 of the flowchart in Fig. 4 will be described.
[00158] The high frequency signal generation circuit 16 calculates a low frequency subband power, power (ib, J), of each subband of the multiple subband signals provided from the bandpass filter 13, based on Expression (1) described above. The high frequency signal generation circuit 16 uses the calculated multiple low frequency subband powers, power (ib, J), and the estimated high frequency subband power values, powerest (ib, J) , which are calculated based on Expression (2) described above by the frequency subband power estimation circuit 15 to find a gain value G (ib, J), according to Expression (3) below. [Expression 3]

[00159] Now, in Expression (3), sbmap (ib) represents a subband index of a source image source in the case that the subband ib is the subband of a target image, and is expressed in Expression (4) below. [Expression 4]

[00160] Note that in Expression (4), INT (a) is a function to round down the decimal point of a value a.
[00161] Next, the high frequency signal generation circuit 16 calculates a post-gain subband x2 (ib, n) signal, multiplying the gain value G (ib, J) found with the Expression (3) through the bandpass filter output 13, using Expression (5) below. [Expression 5]

[00162] Additionally, the high frequency signal generation circuit 16 calculates, using Expression (6) below, the sub-band signal x3 (ib, n) of gain post-adjustment that was submitted to the cosine transform , from the subband signal x2 (ib, n) of gain post-adjustment, making cosine adjustment for the frequency corresponding to the frequency at the upper end of the subband having a sb index, from a frequency corresponding to a frequency at the lower end of the subband having an index of sb-3. [Expression 6]

[00163] Note that in Expression (6), it represents the circumference ratio. Expression (6) here means that the post-gain subband signal x2 (ib, n) is shifted towards the frequency of the high frequency side, of four complete bands each.
[00164] The high frequency signal generation circuit 16 then calculates components of the high frequency signal xhigh (n) from subband signal x3 (ib, n) of the gain post-adjustment shifted towards the side high frequency, with Expression (7) below. [Expression 7]

[00165] Therefore, components of the high frequency signal are generated by the high frequency signal generation circuit 16, based on the four low frequency subband powers calculated based on the four subband signals from the bandpass filter 13, and an estimated high frequency subband power value from the frequency subband power estimation circuit 15, and are provided for the highpass filter 17.
[00166] According to the above processing, as for an input signal obtained after decoding the encoded data using a high frequency elimination encoding method, using the low frequency subband power calculated from the multiple signals subband values according to the characteristic value, based on this and an appropriately configured coefficient, a high frequency subband power estimate value is calculated, and components of the high frequency signal are appropriately generated from the low frequency subband and the high frequency subband power estimation value, and thereby the subband power of the frequency extension band can be estimated with high precision, and music signals can be performed with higher sound quality ..
[00167] Descriptions have been given above an example characterized by the fact that the characteristic value calculation circuit 14 calculates only the low frequency subband power calculated from the multiple subband signals according to the value of characteristic, but in this case, depending on the type of input signal, the subband power of the frequency extension band may not be able to be estimated with high accuracy.
[00168] Therefore, the characteristic value calculation circuit 14 calculates a characteristic value having a strong correlation with the shape of the subband power of the frequency extension band (shape of the high frequency power spectrum), and thereby, estimating the subband power of the frequency extension band in the frequency subband power estimation circuit 15 can be performed with greater precision. [Another Example of Characteristic Value Calculated by the Characteristic Value Calculation Circuit]
[00169] Fig. 6 shows, with respect to a given input signal, an example of a frequency characteristic in a vocal segment that is a segment characterized by the fact that the vocal takes up a large portion of it, and the spectrum of high frequency power obtained by calculating the low frequency subband power only as a characteristic value to estimate the high frequency subband power.
[00170] As shown in Fig. 6, in a frequency characteristic in a vocal segment, the estimated high frequency power spectrum is often positioned higher than the high frequency power spectrum of the original signal. The discomfort of a singing person's voice is readily perceived by the human ear, so the high frequency subband power estimation needs to be particularly precisely performed on a vocal segment.
[00171] Furthermore, as shown in Fig. 6, in a frequency characteristic in a vocal segment, a large recess is often seen between 4.9 kHz and 11.025 kHz.
[00172] Now, an example will be described below an example to apply the degree of recess between 4.9 kHz and 11.025 kHz in a frequency region, serving as the characteristic value used to estimate the power of the high subband frequency in a vocal segment. Note that the characteristic value that indicates the degree of recess from now on will be called dip.
[00173] An example of calculating the dip, dip (J), in time frame J will be described below.
[00174] First, FFT (Fast Fourier Transform) of 2048 points is performed for signals in 2048 sample segments included in a range of several frames before and after, including time frame J, of the input signal, and coefficients on a frequency axis are calculated. A power spectrum is obtained by carrying out the db transform in the absolute values of the various calculated coefficients.
[00175] Fig. 7 shows an example of a spectrum of power obtained as described above. Now, in order to remove thin components from the power spectrum, elevation processing is performed in order to remove components that are 1.3 kHz or less, for example. According to the elevation processing, the various dimensions of the power spectrum are visualized as time series, and filtering processing is carried out by applying a low-pass filter, thereby flattening the thin components of the peak of the spectrum.
[00176] Fig. 8 shows an example of a power spectrum of a post-elevation input signal. In the post-elevation power spectrum in Fig. 8, the difference between the minimum and maximum value of the power spectrum included in an interval corresponding to 4.9 kHz to 11.025 kHz is configured as dip, dip (J).
[00177] Therefore, a characteristic value having a characteristic value that is strongly correlated with the subband power of the frequency extension band is calculated. Note that the example of calculating dip dip (J) is not restricted to the example described above, and you can use another method.
[00178] In the following, another example of calculating a characteristic value having a strong correlation with the subband power of a frequency extension band will be described. [Yet Another Example of a Characteristic Value Calculated with Characteristic Value Calculation Circuit]
[00179] For a frequency characteristic of an "attack" segment, which is a segment including a "attack" type music signal, the power spectrum on the high frequency side is often approximately flat on a given input signal , as described with reference to Fig. 2. With the method to calculate only the low frequency subband power according to the characteristic value, the subband power of the frequency extension band is estimated without using the value of characteristic showing a unique time variation for the input signal that includes the "attack" segment, thus estimating a subband power of the approximately flat frequency extension band as seen in a "attack" segment, with high accuracy, it's difficult.
[00180] Therefore, an example of applying a temporal variation of low frequency subband power serving as a characteristic value used in estimating high frequency subband power in an "attack" segment will be described below .
[00181] The time variation of the low frequency subband power, potency (J) in a given time frame J is found with Expression (8) below, for example. [Expression 8]

[00182] According to Expression (8), the temporal variation of the low frequency subband power of the powerd (J) expresses a proportion of the sum of the four low frequency subband powers in the time frame J and the sum of the four low-frequency subband powers in the time frame (J-1) which is a frame prior to the time frame J, and the higher this value is, the greater the temporal variation in power between frames, ie how much strongest is the "attack", it is considered to be the signal included in time frame J.
[00183] In addition, comparing a statistically average power spectrum shown in Fig. 1 and a power spectrum in an "attack" segment (musical signal of the "attack" type) shown in Fig. 2, the power spectrum in the “attack” segment it appears to the right at a medium frequency. This type of frequency characteristic is often shown in the “attack” segments.
[00184] Now, an example of applying a slope on the middle frequency will be described below, as a characteristic value used to estimate the high frequency subband power in an “attack” segment.
[00185] The slope, slope (J), in the average frequency of a given time frame J is obtained with Expression (9) below, for example. [Expression 9]

[00186] In Expression (9), the coefficient w (ib) is a weighted coefficient that is adjusted to be weighted by a high frequency subband power. According to Expression (9), the slope (J) expresses the ratio between the sum of the four low frequency subband powers weighted by the high frequency and the sum of the four low frequency subband powers. For example, in the event that the four low frequency subband powers become a power corresponding to a medium frequency subband, the slope (J) takes on a greater value when the medium frequency power spectrum appears for the right, and a lower value when falling to the right.
[00187] Furthermore, in many cases the mean frequency slope varies widely before and after an “attack” segment, and thereby, the temporal variation of the slope, the sloped (J), expressed with Expression (10) below it can be configured as the characteristic value used to estimate the high frequency subband power of an “attack” segment. [Expression 10]

[00188] In addition, similarly, the temporal variation, dipd (J), of the dip described above, dip (J), expressed in the following Expression (11), can be configured as a characteristic value used to estimate the high frequency subband power of an “attack” segment. [Expression 11]

[00189] According to the method above, a characteristic value having a strong correlation with the subband power of the frequency extension band is calculated, thus using these, the estimate of the subband power of the extension band with the frequency subband power estimation circuit 15 can be carried out with greater precision.
[00190] An example to calculate a characteristic value having a strong correlation with the subband power of the frequency extension band is described above, but an example of estimating the high frequency subband power using the value of characteristic thus being calculated will be described below. [Processing Details with Frequency Subband Power Estimating Circuit]
[00191] Now, an example to estimate a high frequency subband power, using the dip described with reference to Fig. 8 and a low frequency subband power according to the characteristic values, will be described.
[00192] This is to say, in step S4 in the flowchart in Fig. 4, the characteristic value calculation circuit 14 calculates a low frequency and dip subband power as characteristic values for each subband, the from the four subband signals from the bandpass filter 13, and supplies them to the frequency subband power estimation circuit 15.
[00193] In step S5, the frequency subband power estimation circuit 15 calculates an estimate value of the high frequency subband power, based on the four low frequency subband powers from the circuit for calculating the value of characteristic 14 and the dip.
[00194] Now, with the subband and dip power, since the intervals (scale) of the values that can be considered different, the frequency subband power estimation circuit 15 transforms the dip values according to shown below, for example.
[00195] The frequency subband power estimation circuit 15 calculates the maximum frequency subband power of the four low frequency subband powers, and the dip values, for a large number of frequency signals. entries in advance, and find mean values and standard deviations for each. Now, the average value of the subband powers is represented by powerave, the standard deviation of the subband powers as powerstd, the average value of the dips as dipave, and the standard deviation of the dips as dipstd.
[00196] The frequency subband power estimation circuit 15 transforms the value of dip, dip (J) as shown in Expression (12) below, using these values, and obtains a post-transformation dip, dips (J ). [Expression 12]

[00197] Carrying out the transformation shown in Expression (12), the frequency subband power estimation circuit 15 can transform the dip dip value (J) into dips (J) variables (J) equivalent to the average and dispersion of low-frequency subband powers, and can cause the range of values that can be considered from the dips to be approximately the same as the range of values that can be taken from the subband powers.
[00198] An estimated value of the potency of sub-band potency (ib, J) having an index of ib in a frequency extension band is expressed with Expression (13) below, for example, using a linear combination of the four powers low frequency subband, power (ib, J), from the characteristic value calculation circuit 14 and the dips, dips (J), shown in Expression (12). [Expression 13]

[00199] Now, in Expression (13), the coefficients Cib (kb), Dib, and Eib are coefficients having values that differ for each subband ib. The coefficients Cib (kb), Dib, and Eib are coefficients appropriately configured such that favorable values can be obtained for the various input signals. In addition, depending on the variation of a subband sb, the coefficients Cib (kb), Dib, and Eib can also be varied to be optimal values. Note that production of the coefficients Cib (kb), Dib, and Eib will be described later.
[00200] In Expression (13), the estimated value of the high frequency subband power is calculated with a linear combination, but not restricted to this, it can be calculated using a linear combination of multiple characteristic values of several frames before and after the time frame J, or it can be calculated using a non-linear function, for example.
[00201] According to the above processing, the exclusive dip value for the vocal segment is used as a characteristic value in the estimation of the high frequency subband power, and thereby the precision estimate of high frequency subband of the vocal segment can be improved when compared to the case characterized by the fact that only the low frequency subband power is the characteristic value, and discomfort readily perceived by the human ear, which is generated by a high frequency power spectrum being estimated to be greater than the high frequency power spectrum of the original signal with the method characterized by the fact that only low frequency subband power is the characteristic value , is reduced, and thereby, music signals can be played with higher sound quality.
[00202] Now, considering the dips (degree of recess in a vocal segment frequency characteristic) calculated as characteristic values with the method described above, in the case that the number of subband divisions is 16, the resolution of frequency is low, so the degree of recess here cannot be expressed only with the low frequency subband power.
[00203] Now, increasing the number of subband divisions (eg 16 times, which is 256 divisions), increasing the number of band divisions with the 13 bandpass filter (eg 16 times, which is 64), and by increasing the number of low-frequency sub-band powers (eg 16 times, which is 64) calculated with characteristic value calculation circuit 14, the frequency resolution can be improved, and the degree of recess here can be expressed only with the low frequency subband power.
[00204] Therefore, it can be thought that the high frequency subband power can be estimated with approximately the same precision as the estimate of a high frequency subband power using the dip described above as a characteristic value. , using only the low frequency subband power.
[00205] However, increasing the number of sub-band divisions, and the number of low-frequency sub-band powers, the amount of calculations increases. If we consider that the high frequency subband power can be estimated with similar precision or method, the method that does not increase the number of subband divisions and that uses dip as a characteristic value to estimate the power of high frequency subband is more efficient from the perspective of design values.
[00206] The above description has been about a method for estimating high frequency subband power using dip and low frequency subband power, but the characteristic value used in estimating a sub frequency power high frequency band is not restricted to this combination, and one or multiple of the characteristic values described above (low frequency subband power, dip, time variation of low frequency subband power, slope, time variation of slope, and temporal dip variation) can be used. Therefore, accuracy to estimate the high frequency subband power can be further improved.
[00207] In addition, as described above, in an input signal, using exclusive parameters for a segment characterized by the fact that the estimation of the high frequency subband power is difficult according to the characteristic value used for the estimation of the high frequency subband power, the accuracy of the segment estimate can be improved. For example, temporal variation of low frequency subband power, slope, temporal slope variation, and temporal dip variation, are parameters exclusive to the “attack” segment, and using these parameters with characteristic values, the precision estimation of high frequency subband power in the “attack” segment can be improved.
[00208] Note that in case of estimating the high frequency subband power using the characteristic value other than the low frequency and dip subband power, ie using the time variation of the subband power low frequency, slope, time slope variation, and time dip variation, the high frequency subband power can be estimated with the same method as described above.
[00209] Note that each of the methods of calculating the characteristic values shown here are not restricted to the methods described above, and that other methods can be used. [Method of Finding Coefficients Cib (kb), Dib, Eib]
[00210] Next, a method for finding the coefficients Cib (kb), Dib, and Eib in Expression (13) above will be described.
[00211] As a method to find the coefficients Cib (kb), Dib, and Eib, a method is used and thereby learning is done in advance with a learning signal having a broadband (hereinafter called a broadband learning), such that, in estimating the subband power of the frequency extension band, the coefficients Cib (kb), Dib, Eib can be favorable values for the various input signals, and can be determined with based on its learning outcomes.
[00212] In the event of learning the coefficients Cib (kb), Dib, and Eib, a coefficient learning device that positions a bandpass filter having a pass bandwidth similar to bandpass filters 13-1 to 13- 4 described above with reference to Fig. 5, with a higher frequency than the start extension band, is used. When a broadband learning signal is being input, the coefficient learning device performs learning. [Functional Example of Setting the Coefficient Learning Device]
[00213] Fig. 9 shows a functional example of configuring a coefficient learning device to perform learning of the coefficients Cib (kb), Dib, and Eib.
[00214] With respect to the signal components of a frequency lower than the broadband start signal extension band entered for the coefficient learning device 20 in Fig. 9, it is favorable for an input signal restricted bandwidth that is entered into a frequency band extension device 10 in Fig. 3 to be a signal encoded with the same format as the encoding format performed in the encoding event.
[00215] The coefficient learning device 20 is composed of a bandpass filter 21, a high frequency subband power calculation circuit 22, a characteristic value calculation circuit 23, and a value estimation circuit coefficient 24.
[00216] The bandpass filter 21 is composed of bandpass filters 21-1 to 21- (K + N), each of which has different bandwidths. The bandpass filter 21-i (1 <i <K + N) allows a predetermined bandpass signal from the input signal to pass through it, and supplies it as one of the multiple subband signals for the calculation circuit of high frequency subband power 22 or for the characteristic value calculation circuit 23. Note that bandpass filters 21-1 to 21-K, bandpass filters 21-1 to 21- (K + N ), allow signals to pass through a higher frequency than the start extension band.
[00217] The high frequency subband power calculation circuit 22 calculates the high frequency subband power for each subband for each given time frame as to the multiple high frequency subband signals to from the bandpass filter 21, and supplies them to the coefficient estimation circuit 24.
[00218] The characteristic value calculation circuit 23 calculates a characteristic value which is the same as the characteristic value calculated by the characteristic value calculation circuit 14 of the frequency band extension device 10 in Fig. 3, for each time frame which is the same as the given time frame calculated for the high frequency subband power by the high frequency subband power calculation circuit 22. That is to say, the calculation circuit characteristic value 23 uses at least one of the multiple subband signals from the bandpass filter 21 and broadband learning signal to calculate one or multiple characteristic values, and supplies it to the coefficient estimation circuit 24 .
[00219] The coefficient estimation circuit 24 estimates a coefficient used with the frequency subband power estimation circuit 15 of the frequency band extension device 10 in Fig. 3, based on the subband power high frequency from the high frequency subband power calculation circuit 22 and the characteristic value from the characteristic value calculation circuit 23 of each given time frame. [Coefficient Learning Processing of the Coefficient Learning Device]
[00220] Next, the processing of coefficient learning by the coefficient learning device in Fig. 9 will be described with reference to the flowchart in Fig. 10.
[00221] In step S11, the bandpass filter 21 divides the input signal (broadband learning signal) into (K + N) subband signal numbers. Bandpass filters 21-1 to 21-K provide multiple subband signals having a higher frequency than the start extension band for the high frequency 22 subband power calculation circuit. the bandpass filter 21- (K + 1) to 21- (K + N) provides the multiple subband signals having a lower frequency than the start extension band for the characteristic value calculation circuit 23 .
[00222] In step S12, the high frequency subband power calculation circuit 22 calculates the high frequency subband power, power (ib, J) for each subband, for each given time frame , for multiple high frequency subband signals from the bandpass filter 21 (bandpass filters 21-1 to 21-K). The high frequency subband power, power (ib, J), is found with Expression (1) described above. The high frequency subband power calculation circuit 22 supplies the high frequency subband power calculated for the coefficient estimation circuit 24.
[00223] In step S13, the characteristic value calculation circuit 23 calculates the characteristic value for each time frame which is the same as the given time frame calculated for the high frequency subband power by the high frequency subband power calculation 22.
[00224] Note that in the characteristic value calculation circuit 14 of the frequency band extension device 10 in Fig. 3, it is assumed that the four low frequency subband powers and the dip are calculated according to the values of characteristic, and similar to the characteristic value calculation circuit 23 of the coefficient learning device 20, description is given below as calculating the four low frequency subband powers and the dip.
[00225] This is to say, the characteristic value calculation circuit 23 uses the four subband signals, each having the same band as the four subband signals introduced in the characteristic value calculation circuit 14 of the frequency band extension device 10, from the bandpass filter 21 (bandpass filters 21- (K + 1) to 21- (K + 4), to calculate the four low frequency subband powers. In addition, the characteristic value calculation circuit 23 calculates a dip from the broadband learning signal, and calculates the dip, dips (J) based on the Expression (12) described above. of characteristic 23 supplies the four calculated low-frequency sub-band powers and dip, dips (J), as characteristic values for the coefficient estimation circuit 24.
[00226] In step S14, the coefficient estimation circuit 24 performs DOS estimation coefficients Cib (kb), Dib, and Eib, based on multiple combinations of the (eb-sb) number of high frequency subband powers provided for the same time frame from the high frequency subband power calculation circuit 22 and the characteristic value calculation circuit 23 and the characteristic values (four low frequency subband powers and dip dips (J)). For example, for a given high frequency subband, the coefficient estimation circuit 24 sets five characteristic values (four low frequency subband powers and dip dips (J)) as explanatory variables, and the power high frequency sub-band power (ib, J) with an explained variable, and performs regression analysis using a minimum square method, and thereby determining the coefficients Cib (kb), Dib, and Eib in Expression ( 13).
[00227] Note that, of course, the method of estimating the coefficients Cib (kb), Dib, and Eib is not restricted to the method described above, and several types of methods for identifying general parameters can be used.
[00228] According to the processing described above, learning of the coefficients used to estimate the high frequency subband power is performed using a broadband learning signal in advance, and thereby, favorable output results can be obtained as for the various input signals introduced into a frequency band extension device 10, and therefore, music signals can be played with higher sound quality.
[00229] Note that the coefficients Aib (kb) and Bib in Expression (2) described above can also be obtained with the coefficient learning method described above.
[00230] A coefficient learning processing is described above, with the premise that in the frequency subband power estimation circuit 15 of a frequency band extension device 10, each of the power estimation values of high frequency subband is calculated with a linear combination of the four low frequency subband powers and the dip. However, a method of estimating the high frequency subband power in the frequency subband power estimating circuit 15 is not restricted to the example described above, and for example, the characteristic value calculation circuit 14 can calculate one or multiple characteristic values other than the dip (low frequency subband power, time variation, slope, time slope variation, and time dip variation) to calculate the high frequency subband power, or Linear combinations of multiple values of characteristics of the multiple frames before and after the time frame J can be used, or non-linear functions can be used. This is to say, in the processing of coefficient learning, the coefficient estimation circuit 24 must be able to calculate (learn) the coefficients, with conditions similar to the conditions for the characteristic values, time frame, and functions used in the event to calculate the high frequency subband power with the frequency subband power estimation circuit 15 of a frequency band extension device 10. <2. Second modality>
[00231] With the second mode, the encoding processing and decoding processing are carried out with a high frequency characteristic encoding method, with an encoding device and decoding device. [Functional Example of Configuring the Coding Device]
[00232] Fig. 11 shows a functional example of configuration of the coding device to which the present invention is applied.
[00233] A coding device 30 is composed of a low-pass filter 31, a low-frequency coding circuit 32, a subband division circuit 33, a characteristic value calculation circuit 34, a calculation circuit high frequency subband pseudopotency 35, a high frequency subband pseudopotence difference calculation circuit 36, a high frequency encoding circuit 37, a multiplexing circuit 38, and a low decoding circuit frequency 39.
[00234] The low-pass filter 31 filters the input signal at a predetermined cutoff frequency, and supplies signals having a frequency lower than the cutoff frequency (hereinafter called low frequency signals) for the coding circuit. low frequency 32, for the subband division circuit 33, and for the characteristic value calculation circuit 34, as a post-filter signal.
[00235] The low frequency coding circuit 32 encodes the low frequency signal from the low pass filter 31, and supplies the low frequency coded data obtained as a result of it for the multiplexing circuit 38 and for the control circuit. low frequency decoding 39.
[00236] The subband division circuit 33 divides the low frequency signal from the input signal and low pass filter 31 into multiple equal subband signals having a predetermined bandwidth, and supplies them to the circuit characteristic value calculation circuit 34 or high frequency subband pseudo-power difference calculation circuit 36. More specifically, the subband division circuit 33 supplies the multiple subband signals obtained with low frequency signals as the input (hereinafter referred to as low frequency subband signals) for characteristic value calculation circuit 34. In addition, subband division circuit 33 supplies the subband signals having a frequency higher than the cutoff frequency configured by the low pass filter 31 (hereinafter called high frequency subband signals), of the multiple subband signals obtained with the input signal as the input, for the calculation circuit the difference in pseudopotency of high frequency subband 36.
[00237] The characteristic value calculation circuit 34 uses at least one of the multiple subband signals of the low frequency subband signals from the subband split circuit 33 or low frequency signals from of the low-pass filter 31 to calculate one or multiple characteristic values, and supplies it for the high-frequency subband pseudopotence calculation circuit 35.
[00238] The high frequency subband pseudopotency calculation circuit 35 generates a high frequency subband pseudopotency based on one or multiple characteristic values from the characteristic value calculation circuit 34, and supplies it for the high frequency subband pseudopotence difference calculation circuit 36.
[00239] The high frequency subband pseudo-power difference calculation circuit 36 calculates the high frequency subband pseudo-power difference described later, based on the high frequency subband signals from the circuit of sub-band splitting 33 and in the high-frequency sub-band pseudo-power from the high-frequency sub-band pseudo-power calculation circuit 35, and supplies it for the high-frequency coding circuit 37.
[00240] The high frequency coding circuit 37 encodes the high frequency subband pseudopotency difference from the high frequency subband pseudopotence difference calculation circuit 36, and supplies the high frequency coded data obtained as a result of it for the multiplexing circuit 38.
[00241] Multiplexing circuit 38 multiplexes low-frequency encoded data from low-frequency encoding circuit 32 and high-frequency encoded data from high-frequency encoding circuit 37, and outputs them as a sequence exit code.
[00242] Low-frequency decoding circuit 39 decodes low-frequency encoded data from low-frequency encoding circuit 32 as appropriate, and supplies the decoded data obtained as a result thereof for the sub-division circuit. band 33 and for the characteristic value calculation circuit 34. [Coding Device Coding Processing]
[00243] In the following, coding processing with the coding device 30 in Fig. 11 will be described with reference to the flowchart in Fig. 12.
[00244] In step S111, the low-pass filter 31 filters the input signal with a predetermined cutoff frequency, and supplies the low frequency signal serving as a post-filtering signal for the low frequency coding circuit 32, for the subband division circuit 33, and for the characteristic value calculation circuit 34.
[00245] In step S112, the low frequency coding circuit 32 encodes the low frequency signal from the low pass filter 31, and supplies the low frequency coded data obtained as a result of it for the multiplexing circuit 38.
[00246] Note that as for low frequency signal coding in step S112, it is sufficient that an appropriate coding format is selected according to the scope of the circuit to be found and coding efficiency, and the present invention does not depend on this format coding.
[00247] In step S113, the subband division circuit 33 also divides the input signal and the low frequency signal into multiple subband signals having a predetermined bandwidth. The subband division circuit 33 supplies the low frequency subband signals, obtained with the low frequency signal as input, for the characteristic value calculation circuit 34. In addition, the multiple subband signals band obtained with the input signal as input, the subband division circuit 33 supplies the high frequency subband signals having a band greater than a restricted frequency of the band configured by the low pass filter 31 for the calculation of pseudopotency difference of high frequency subband 36.
[00248] In step S114, the characteristic value calculation circuit 34 uses at least one of the multiple subband signals of the low frequency subband signals from the subband split circuit 33 or the signal low frequency filter from the low pass filter 31 to calculate one or multiple characteristic values, and supplies it for the high frequency subband pseudo-power calculation circuit 35. Note that the characteristic value calculation circuit 34 in Fig. 11 has basically the same configuration and functionality as the characteristic value calculation circuit 14 in Fig. 3, so processing in step S114 is basically the same as processing in step S4 of the flowchart in Fig. 4, then description detailed information will be omitted.
[00249] In step S115, the high frequency subband pseudo-power calculation circuit 35 generates a high frequency subband pseudo-power, based on one or multiple characteristic values from the value calculation circuit. characteristic 34, and supplies it for the high frequency subband pseudo-power difference calculation circuit 36. Note that the high frequency subband pseudo-power calculation circuit 35 in Fig. 11 has basically the same configuration and function of the frequency subband power estimation circuit 15 in Fig. 3, and processing in step S115 is basically the same as processing in step S5 in the flowchart in Fig. 4, so detailed description will be omitted.
[00250] In step S116, the high frequency subband pseudopotence difference calculation circuit 36 calculates the high frequency subband pseudopotence difference, based on the high frequency subband signal from the subband splitting circuit 33 and the high frequency subband pseudopotency from the high frequency subband pseudopotency calculation circuit 35, and supplies it to the high frequency coding circuit 37.
[00251] More specifically, the high frequency subband pseudo-power difference calculation circuit 36 calculates the (high frequency) subband power, power (ib, J), in a given time frame J, of the high-frequency sub-band signal from sub-band split circuit 33. Note that according to the present modality, all sub-bands of the low-frequency sub-band signal and sub-bands of the high frequency subband are identified using the ib index. The subband power calculation method can be a similar method for the first modality, i.e. the method used for Expression (1) can be applied.
[00252] Next, the high frequency sub-band pseudo-power difference calculation circuit 36 finds the difference (high-frequency sub-band pseudo-power difference) diff power (ib, J) between the sub band power high frequency, power (ib, J), and the high frequency subband pseudopotency, powerlh (ib, J), from the high frequency subband pseudopotence calculation circuit 35 in time frame J The difference in pseudopotency of the high frequency subband, powerdiff (ib, J), is found with Expression (14) below. [Expression 14]

[00253] In Expression (14), the sb + 1 index represents a minimum frequency subband index in the high frequency subband signal. In addition, the eb index represents a maximum frequency subband index encoded in the high frequency subband signal.
[00254] Therefore, the high frequency subband pseudopotency difference calculated with the high frequency subband pseudopotence difference calculation circuit 36 is provided for the high frequency coding circuit 37.
[00255] In step S117, the high frequency coding circuit 37 encodes the high frequency subband pseudopotency difference from the high frequency subband pseudopotence difference calculation circuit 36, and supplies the data high frequency encodings obtained as a result of it for multiplexing circuit 38.
[00256] More specifically, the high frequency coding circuit 37 determines which cluster, of multiple clusters in a characteristic space of a preconfigured difference of high frequency subband pseudopotency, owes the difference of sub pseudopotency -high frequency band vectorized from the high frequency subband pseudopotence difference calculation circuit 36 (hereinafter called the high frequency subband pseudopotence difference vector) belongs. Now, the high frequency subband pseudo-power difference vector in a given time frame J indicates a vector dimension (eb-sb) that has high frequency powerdiff sub-band pseudo-power difference values (ib, J) for each ib index, as the elements for the vectors. In addition, the characteristic space for the high frequency subband pseudo-power difference similarly has a dimension space (eb-sb).
[00257] In the characteristic space for the high frequency subband pseudopotency difference, the high frequency coding circuit 37 measures the distance between the various representative vectors of preconfigured multiple clusters and the pseudopotence difference vector of high frequency subband, and finds an index for the cluster with the shortest distance (hereinafter called the high frequency subband pseudo-power difference ID), and supplies it to the multiplexing circuit 38 as encoded data high frequency.
[00258] In step S118, the multiplexing circuit 38 multiplexes the low frequency coded data emitted from the low frequency coding circuit 32 and the high frequency coded data emitted from the high frequency coding circuit 37, and issues an exit code sequence.
[00259] Now, considering a coding device for the high frequency characteristic coding method, a technique is disclosed in Japanese Untested Patent Publication Application No. 2007-17908 in which a subband pseudo-signal high-frequency signal is generated from a low-frequency sub-band signal, the pseudo-power of the high-frequency sub-band and high-frequency sub-band are compared for each sub-band, the power gain for each subband is calculated to match the high frequency subband signal pseudo-power and the high frequency subband signal power, and this is included in a code sequence as high frequency characteristic information.
[00260] On the other hand, according to the processing described above, in the decoding event, only the high frequency subband pseudo-power difference ID has to be included in the output code sequence as information to estimate the power high frequency subband. This is to say, in the case that the number of preconfigured clusters is 64 for example, as information to decode the high frequency signal with a decoding device, only 6-bit information has to be added to a code sequence for a time frame, and compared to the method disclosed in Untested Japanese Patent Application Publication No. 2007-17908, the amount of information to be included in the code sequence can be reduced, the coding efficiency can be improved, and therefore, music signals can be played with higher sound quality.
[00261] Furthermore, with the processing described above, if there is a margin of maneuver in the calculation amount, the low frequency decoding circuit 39 can input the low frequency signal obtained by decoding the low frequency encoded data from the circuit low-frequency coding 32 in the subband division circuit 33 and in the characteristic value calculation circuit 34. For the decoding processing by the decoding device, the characteristic value is calculated from the obtained low-frequency signals having decoded the low frequency coded data, and the high frequency subband power is estimated based on its characteristic value. Therefore, also with the encoding processing, including the high frequency subband pseudo-power difference ID which is calculated based on the characteristic value calculated from the low frequency signal decoded in the code sequence, makes it possible to estimate high frequency subband power with greater accuracy in decoding processing with the decoding device. Consequently, music signals can be played with higher sound quality. [Functional Example of Decoding Device Configuration]
[00262] In the following, a functional example of configuration of the decoding device corresponding to the encoding device 30 in Fig. 11 will be described with reference to Fig. 13.
[00263] The decoding device 40 is composed of a demultiplex circuit 41, a low frequency decoding circuit 42, a subband division circuit 43, a characteristic value calculation circuit 44, a decoding circuit high band 45, a decoded high frequency subband power calculation circuit 46, a decoded high frequency signal generation circuit 47, and a synthesis circuit 48.
[00264] Demultiplex circuit 41 undo multiplexes the input code sequence into high frequency encoded data and low frequency encoded data, and supplies the low frequency encoded data to the low frequency decoding circuit 42 and supplies the data high frequency encoded for high frequency decoding circuit 45.
[00265] The low frequency decoding circuit 42 performs decoding of the low frequency encoded data from the demultiplexing circuit 41. The low frequency decoding circuit 42 supplies the low frequency signals obtained as a result of decoding (hereinafter hereinafter called decoded low frequency signals) for subband division circuit 43, for characteristic value calculation circuit 44, and for synthesis circuit 48.
[00266] Subband division circuit 43 also divides the low frequency signal decoded from the low frequency decoding circuit 42 into multiple subband signals having a predetermined bandwidth, and supplies the sub signals -band (decoded low frequency subband signal) obtained for the characteristic value calculation circuit 44 and for the decoded high frequency signal generation circuit 47.
[00267] Feature value calculation circuit 44 uses at least one of multiple subband signals from the low frequency subband signals decoded from the subband split circuit 43 and the low frequency signal decoded from the low frequency decoding circuit 42 to calculate one or multiple characteristic values, and supplies it to the decoded high frequency subband power calculation circuit 46.
[00268] The high frequency decoding circuit 45 performs decoding of the high frequency encoded data from the demultiplexing circuit 41, and uses the high frequency subband pseudo-power difference ID obtained as a result of it to provide the coefficient (hereinafter called decoded high-frequency subband power estimation coefficient) to estimate a high-frequency sub-band power prepared in advance for each ID (index) for the sub-power calculation circuit decoded high frequency band 46.
[00269] The decoded high frequency subband power calculation circuit 46 calculates the decoded high frequency subband power based on one or multiple characteristic values from the characteristic value calculation circuit 44 and the high frequency subband power estimation coefficient decoded from the high frequency decoding circuit 45, and supplies it for the decoded high frequency signal generation circuit 47.
[00270] The decoded high frequency signal generation circuit 47 generates a decoded high frequency signal based on the decoded low frequency subband signal from the subband division circuit 43 and the subpower power decoded high frequency band from the decoded high frequency subband power calculation circuit 46, and supplies it to the synthesis circuit 48.
[00271] The synthesis circuit 48 synthesizes the decoded low frequency signal from the low frequency decoding circuit 42 and the high frequency signal decoded from the decoded high frequency signal generation circuit 47, and outputs as an exit signal. [Decoding Device Decoding Processing]
[00272] In the following, decoding processing with the decoding device in Fig. 13 will be described with reference to the flowchart in Fig. 14.
[00273] In step S131, demultiplexing circuit 41 undo multiplexes the input code sequence into high frequency encoded data and low frequency encoded data, supplies the low frequency encoded data to the low frequency decoding circuit 42, and supplies the high frequency encoded data for the high frequency decoding circuit 45.
[00274] In step S132, the low frequency decoding circuit 42 performs low frequency coded data decoding from the demultiplexing circuit 41, and supplies the decoded low frequency signal obtained as a result of it for a splitting circuit. subband 43, for a characteristic value calculation circuit 44, and for a synthesis circuit 48.
[00275] In step S133, the subband division circuit 43 divides the decoded low frequency signal from the low frequency decoding circuit 42 equally into multiple subband signals having predetermined bandwidths, and supplies the decoded low-frequency subband signal obtained for the characteristic value calculation circuit 44 and for the decoded high frequency signal generation circuit 47.
[00276] In step S134, the characteristic value calculation circuit 44 calculates one or multiple characteristic values from at least one of the multiple subband signals of the low frequency subband signals decoded from the circuit subband splitter 43 and the low frequency signals decoded from the low frequency decoding circuit 42, and supplies it to the decoded high frequency subband power calculation circuit 46. Note that the calculation of characteristic value 44 in Fig. 13 has basically the same configuration and functionality as the circuit of calculation of characteristic value 14 in Fig. 3, and the processing in step S134 is basically the same as the processing in step S4 in the flowchart in Fig. 4, so its detailed description will be omitted.
[00277] In step S135, the high frequency decoding circuit 45 performs decoding of the high frequency encoded data from the demultiplexing circuit 41, and using the high frequency subband pseudo-power difference ID obtained as a result likewise, it supplies the coefficients of decoded high frequency subband power estimates that are prepared for each ID (index) in advance for the decoded high frequency subband power calculation circuit 46.
[00278] In step S136, the decoded high frequency subband power calculation circuit 46 calculates the decoded high frequency subband power based on the one or multiple characteristic values from the calculation circuit of characteristic value 44 and high frequency subband power estimate coefficient encoded from high frequency decode circuit 45. Note that the decoded high frequency subband power calculation circuit 46 in Fig. 13 has basically the same configuration and functionality as the frequency subband power estimation circuit 15 in Fig. 3, and the processing in step S136 is basically the same as processing in step S5 in the flowchart in Fig. 4, so detailed description will be omitted.
[00279] In step S137, the decoded high frequency signal generation circuit 47 emits a decoded high frequency signal, based on the low frequency subband signal decoded from the subband division circuit 43 and the decoded high frequency subband power from the decoded high frequency subband power calculation circuit 46. Note that the decoded high frequency signal generation circuit 47 in Fig. 13 has basically the same configuration and functionality as the high frequency signal generation circuit 16 in Fig. 3, and the processing in step S137 is basically the same as the processing in step S6 of the flowchart in Fig. 4, so its detailed description will be omitted.
[00280] In step S138, synthesis circuit 48 synthesizes the low frequency signal decoded from the low frequency decoding circuit 42 and the high frequency signal decoded from the decoded high frequency signal generation circuit 47 , and output it as an exit signal.
[00281] According to the processing described above, using the high frequency subband power estimation coefficient in the decoding event that corresponds to the characteristics of the difference between the high frequency subband pseudo-power calculated in advance in the event encoding and the effective high frequency subband power, the accuracy of estimating the high frequency subband power in the decoding event can be improved, and consequently, music signals can be played with higher sound quality.
[00282] In addition, according to the processing described above, the only information to generate the high frequency signals included in a code sequence is the high frequency subband pseudo-power difference ID, which is not much, then decoding processing can be performed efficiently.
[00283] The above description was made considering the encoding processing and decoding processing for which the present invention is applied, but representative vectors for each of the multiple clusters in a characteristic space of the high subband pseudo-power difference frequency that is preconfigured with the high frequency encoding circuit 37 of the encoding device 30 in Fig. 11, and a method of calculating the decoded high frequency subband power estimation coefficient emitted by the decoding circuit of high frequency 45 of the decoding device 40 in Fig. 13 will be described below. [Vector Representative of Multiple Clusters in the High Frequency Subband Pseudopotence Difference Characteristic Space, and Method of Calculating the Decoded High Frequency Power Estimate Coefficient corresponding to each Cluster]
[00284] As a method to find representative vectors of multiple clusters and the decoded high frequency subband power coefficients of each cluster, the coefficients that can precisely estimate the high frequency subband power in the event of decoding, according to the high frequency subband pseudopotence difference vector calculated in the coding event, need to be prepared. Therefore, the technique is applied characterized by the fact that learning is carried out in advance with a broadband learning signal, and these are determined based on their learning results. [Functional Example of Setting the Coefficient Learning Device]
[00285] Fig. 15 shows a functional example of configuration of a coefficient learning device that performs learning of the representative vectors of multiple clusters and the coefficients of high frequency sub-band power estimates decoded for each cluster.
[00286] The signal component below a cutoff frequency configured by the low-pass filter 31 of the coding device 30, of the broadband learning signal entered in the coefficient learning device 50 in Fig. 15 is favorable when the signal of input to the coding device 30 passes through the low-pass filter 31 and is encoded by the low-frequency coding circuit 32, and is still a low-frequency signal decoded by the low-frequency decoding circuit 42 of the decoding device 40.
[00287] The coefficient learning device 50 is composed of a low-pass filter 51, a subband division circuit 52, a characteristic value calculation circuit 53, a subband pseudo-power calculation circuit of high frequency 54, a high frequency subband pseudopotence difference calculation circuit 55, a high frequency subband pseudopotence difference agglomeration circuit 56, and a coefficient estimation circuit 57.
[00288] Note that each of the low-pass filter 51, the subband division circuit 52, the characteristic value calculation circuit 53, and the high frequency subband pseudo-power calculation circuit 54 of the device coefficient learning method 50 in Fig. 15 has basically the same configuration and functionality as the respective low-pass filter 31, the subband division circuit 33, the characteristic value calculation circuit 34, and the high frequency subband pseudopotency 35 in the coding device 30 in Fig. 11, so the description of it will be omitted as appropriate.
[00289] This is to say, the high frequency subband pseudopotency difference calculation circuit 55 has similar configuration and functionality as the high frequency subband pseudopotence difference calculation circuit 36 in Fig. 11 , but the calculated high frequency subband pseudo-power difference is provided for the high frequency subband pseudo-power difference agglomeration circuit 56, and the high frequency subband power calculated in the event of calculating the difference of high frequency subband pseudopotence is provided for the coefficient estimation circuit 57.
[00290] The high-frequency sub-band pseudo-power difference agglomeration circuit 56 clusters the high-frequency sub-band pseudo-power difference vectors obtained from the high-frequency sub-band pseudo-power difference from the high-frequency sub-band pseudo-power difference computing circuit 55, and calculates the representative vectors for each cluster.
[00291] Coefficient estimation circuit 57 calculates the high frequency subband power estimation coefficient for each cluster that has been clustering with the high frequency subband pseudo-power difference clustering circuit 56, based on in the high frequency subband power from the high frequency subband pseudo-power difference circuit 55, and the one or multiple characteristic values from the characteristic value calculation circuit 53. [Learning Processing Coefficient Learning Device Coefficient]
[00292] Next, processing the coefficient learning with the coefficient learning device 50 in Fig. 15 will be described with reference to the flowchart in Fig. 16.
[00293] Note that the processing in steps S151 to S155 in the flowchart in Fig. 16 is similar to the processing in steps S111 and S113 to S116 in the flowchart in Fig. 12, other than the signal being entered in the coefficient learning device 50 being a sign of broadband learning, so its description will be omitted.
[00294] This is to say, in step S156, the high frequency subband pseudopotence difference agglomeration circuit 56 multiple clusters (a plurality of time periods) high frequency subband pseudopotence difference vectors obtained from the high frequency subband pseudopotency difference from the high frequency subband pseudopotence difference calculation circuit 55 in 64 clusters, for example, and calculates representative vectors for each cluster. An example of an agglomeration method may be to use agglomeration by k means, for example. The high frequency subband pseudopotence difference agglomeration circuit 56 sets up a center of gravity vector for each cluster, which is obtained as a result of agglomerating by k means, as the representative vector for each cluster. Note that the method of agglomeration and number of clusters is not restricted to the above descriptions, and that other methods can be used.
[00295] In addition, the high-frequency sub-band pseudo-power difference clustering circuit 56 uses a high-frequency sub-band pseudo-power difference vector obtained from the high-frequency sub-band pseudo-power difference from the high frequency sub-band pseudopotence difference calculation circuit 55 in a time frame J to measure the distance from the 64 representative vectors, and determines a CID index (J) for the cluster to which the vector representative having the shortest distance belongs. Note that the CID index (J) assumes integer values from 1 for the number of clusters (64 in this example,). The high frequency sub-band pseudo-power difference agglomeration circuit 56 thus emits the representative vector, and supplies the CID index (J) for the coefficient estimation circuit 57.
[00296] In step S157, the coefficient estimation circuit 57 calculates a decoded high frequency subband power estimation coefficient for each cluster, for each group having the same CID (J) index (belonging to the same agglomerate), multiple combinations of the characteristic value and (eb-sb) high frequency subband power number provided for the same time frame from the high frequency subband pseudo-power difference circuit 55 and characteristic value calculation circuit 53. Note that the method for calculating coefficients with the coefficient estimation circuit 57 is similar to the method of the coefficient estimation circuit 24 of the coefficient learning device 20 in Fig. 9, but it goes without saying that another method can be used.
[00297] According to the processing described above, learning is performed for the representative vectors for each of the multiple clusters in the resource space of the pre-configured high-frequency sub-band difference in the high-frequency coding circuit 37 of the encoding device 30 in Fig. 11, and for the decoded high-frequency subband power estimation coefficient emitted by the high-frequency decoding circuit 45 of the decoding device 40 in Fig. 13 using a learning signal from broadband in advance, and thereby favorable output results for the various input signals that are input to the encoding device 30 and various input code sequences entered into the decoding device 40 can be obtained, and therefore, music can be played with higher sound quality.
[00298] Additionally, the coefficient data for calculating the high frequency subband power in the high frequency subband pseudo-power calculation circuit 35 of the coding device 30 and in the subband power calculation circuit High frequency decoded 46 of decoding device 40 can be treated as follows with respect to encoding and decoding the signal. This is to say, using coefficient data that differ by the type of input signal, the coefficient of the same signal can be recorded at the beginning of the code sequence.
[00299] For example, modifying the coefficient data according to signals for a speech or jazz and so on, the efficiency of coding efficiency can be improved.
[00300] Fig. 17 shows a code sequence obtained in this way.
[00301] The code sequence A in Fig. 17 is that of a coded speech, and data of α coefficient, optimal for a speech is recorded in the header.
[00302] Conversely, the code sequence B in Fig. 17 is that of encoded jazz, and data of coefficient β, great for jazz, is recorded in the header.
[00303] Such multiple types of coefficient data can be prepared by learning with similar types of music signals in advance, and coefficient data can be selected by the coding device 30 with the gender information such as that recorded in the input signal header. . Alternatively, the gender can be determined by performing an analysis of the signal waveform, and thus selecting the coefficient data. That is to say, such a method of gender analysis for signs is not particularly restricted.
[00304] Furthermore, if the calculation time allows, the learning device described above can be built into the coding device 30, the processing carried out using the coefficients of a dedicated signal thereof, and as shown in the C code sequence in Fig. 17, finally, its coefficient can be recorded in the header.
[00305] Advantages of using this method will be described below.
[00306] There are many locations in an input signal characterized by the fact that the forms of high-frequency sub-band powers are similar. Using this feature that many input signals have, learning the coefficient to estimate the high frequency subband power, individually for each input signal, allows redundancy caused by the existence of similar high frequency subband power locations to be reduced, and allows coding efficiency to be increased. In addition, high frequency subband power estimation can be performed with greater precision than can learn coefficients to estimate high frequency subband power statistically with multiple signals.
[00307] Furthermore, as shown above, an arrangement can be made characterized by the fact that coefficient data learned from the input signal in the coding event is inserted once in several frames. <3. Third modality> [Functional Example of Configuring the Encoding Device]
[00308] Note that according to the above description, the high frequency subband pseudo-power difference ID is output as high frequency encoded data, from the encoding device 30 to the decoding device 40, but the coefficient index to obtain the decoded high frequency subband power estimate coefficient can be configured as the high frequency encoded data.
[00309] In such a case, the encoding device 30 is configured as shown in Fig. 18, for example. Note that in Fig. 18, the portions corresponding to the case in Fig. 11 have the same reference numerals attached to it, and their description will be omitted as appropriate.
[00310] The coding device 30 in Fig. 18 differs from the coding device 30 in Fig. 11 in that the low frequency decoding circuit 39 is not provided, and elsewhere it is the same.
[00311] With the encoding device 30 in Fig. 18, the characteristic value calculation circuit 34 uses the low frequency subband signal provided from the subband division circuit 33 to calculate the power of low-frequency sub-band as a characteristic value, and supplies it for the high-frequency sub-band pseudo-power calculation circuit 35.
[00312] In addition, multiple decoded high frequency subband power estimate coefficients found through regression analysis in advance and the coefficient indices that identify such decoded high frequency subband power estimate coefficients are correlated and recorded in the high frequency sub-band pseudo-power calculation circuit 35.
[00313] Specifically, multiple sets of coefficients Aib (kb) and Bib coefficient for the various sub-bands used to compute the Expression (2) described above are prepared in advance, as coefficients of decoded high-frequency sub-band power estimates. . For example, these coefficients Aib (kb) and Bib coefficient are found in advance with regression analysis using a minimum square method, with low frequency subband power as explanatory variables, and high frequency subband power. as an explained variable. In regression analysis, an input signal composed of low frequency subband signals and high frequency subband signals is used as the broadband learning signal.
[00314] The high frequency subband pseudo-power calculation circuit 35 uses the decoded high frequency subband power estimate coefficient and the characteristic value from the characteristic value calculation circuit 34 for each recorded decoded high-frequency subband power estimate coefficient to calculate the high-frequency sub-band pseudo-power of each sub-band on the high-frequency side, and supplies them for the sub-pseudo-power difference calculation circuit -high frequency band 36.
[00315] The high frequency subband pseudopotence difference calculation circuit 36 compares the high frequency subband power obtained from the high frequency subband signal provided from the sub division circuit -band 33 and the high frequency subband pseudopotency from the high frequency subband pseudopotency calculation circuit 35.
[00316] As a result of the comparison, of the multiple coefficients of decoded high-frequency sub-band power estimates, the high-frequency sub-band pseudopotence difference calculation circuit 36 supplies, for the high-coding circuit frequency 37, a coefficient index of the high frequency subband power estimation coefficient decoded having obtained the high frequency subband pseudopotency closest to the high frequency subband power. In other words, a coefficient index of the decoded high frequency subband power estimation coefficients, for which a high frequency signal of the input signal to be performed at the decoding time, ie a decoded high frequency signal closest to the actual value is obtained, is selected. [Coding Device Coding Processing]
[00317] Next, coding processing carried out by the coding device 30 in Fig. 18 will be described with reference to the flowchart in Fig. 19. Note that the processing in step S181 to step S183 is similar to step S111 to S113 in Fig. 12, so its description will be omitted.
[00318] In step S184, the characteristic value calculation circuit 34 uses the low frequency subband signal from the subband division circuit 33 to calculate the characteristic value, and supplies it to the circuit for calculating high frequency subband pseudopotency 35.
[00319] Specifically, the characteristic value calculation circuit 34 performs the computation in Expression (1) described above to calculate, according to the characteristic value, the low frequency subband power, power (ib, J), of frame J (where 0 <J) for each subband ib (where sb-3 <ib <sb) on the low frequency side. That is to say, the low frequency subband power, power (ib, J), is calculated by taking the average square root of the sample values for each sample of the low frequency subband signals composing table J as a logarithm.
[00320] In step S185, the high frequency subband pseudo-power calculation circuit 35 calculates the high frequency subband pseudo-power, based on the characteristic value provided from the characteristic value calculation circuit 34 , and supplies it for the high frequency subband pseudopotence difference calculation circuit 36.
[00321] For example, the high frequency subband pseudopotency calculation circuit 35 uses the coefficient Aib (kb) and coefficient Bib which are recorded in advance as the coefficient of estimation of decoded high frequency subband power and the low frequency subband power, power (kb, J) (where sb-3 <kb <sb), to perform the computation in Expression (2) described above, and calculates the high frequency subband pseudopotency, potency (ib, J).
[00322] This is to say, the coefficient Aib (kb) for each sub-band is multiplied by the power of the low-frequency sub-band, power (kb, J), for each sub-band on the sub-low-frequency side band, supplied according to the characteristic value, plus the Bib coefficient is added to the sum of the low-frequency sub-band powers multiplied by the coefficients, and becomes the high-frequency sub-band pseudo-power, potest (ib, J ). The high frequency subband pseudopotence is calculated for each high frequency side subband characterized by the fact that the index is sb + 1 à and b.
[00323] In addition, the high frequency subband pseudopotency calculation circuit 35 performs high frequency subband pseudopotency calculation for each decoded high frequency subband power estimate coefficient recorded in advance. For example, let's say that the coefficient index is 1 to K (where 2 <K), and K coefficients of decoded high-frequency subband power estimates are prepared in advance. In this case, for each of the K coefficients of decoded high-frequency subband power estimates, the high-frequency sub-band pseudo-powers are calculated for each sub-band.
[00324] In step S186, the high frequency subband pseudo-power difference calculation circuit 36 calculates the high frequency subband pseudo-power difference, based on the high frequency subband signal from the subband split circuit 33 and the high frequency subband pseudopotency from the high frequency subband pseudopotency calculation circuit 35.
[00325] Specifically, the high frequency subband pseudo-power difference calculation circuit 36 performs computation similar to that in Expression (1) described above for the high frequency subband signals from the sub division circuit -band 33, and calculates the high-frequency subband power, power (ib, J) in table J. Note that according to the present modality, all sub-bands of the low-frequency subband signals and sub-bands of high frequency sub-band signals are identified using an ib index.
[00326] Next, the high frequency subband pseudo-power difference calculation circuit 36 performs a calculation similar to that in Expression (14) described above, and finds the difference between the high frequency subband power, power (ib, J) in table J, and the high frequency subband pseudopotency, potency (ib, J). Therefore, for each decoded high frequency subband power estimation coefficient, the difference of high frequency subband pseudo-power, diff power (ib, J), is obtained for each subband on the high frequency side characterized by the fact that the index is sb + 1 à and b.
[00327] In step S187, the high frequency subband pseudo-power difference calculation circuit 36 calculates the following expression (15) for each decoded high frequency subband power estimate coefficient, and calculates the sum quadratic difference of high frequency subband pseudopotence. [Expression 15]

[00328] Note that in Expression (15), the sum of the square E (J, id) differences shows the square sum of the high frequency subband pseudopotency difference of frame J, found for the power estimation coefficient of decoded high frequency subband characterized by the fact that the coefficient index is id. Furthermore, in Expression (15), potencydiff (ib, J, id) represents the difference in pseudopotency of high frequency potencydiff (ib, J) of frame J of the subband characterized by the fact that the index is ib, which is found for the decoded high frequency subband power estimation coefficient characterized by the fact that the coefficient index is id. The sum of the squared differences E (J, id) is calculated for each of the K coefficients of decoded high frequency subband power estimates.
[00329] The sum of the square differences E (J, id) thus obtained shows the degree of similarity between the high frequency subband power calculated from the effective high frequency signal and the high subband pseudopotency frequency calculated using the decoded high frequency subband power estimation coefficient characterized by the fact that the coefficient index is id.
[00330] This is to say, the error of estimate values as to the real value of the high frequency subband power is indicated. Consequently, the smaller the sum of the square differences E (J, id) is, the closer to the high frequency signal is the decoded high frequency signal obtained by computing using the decoded high frequency subband power estimate coefficient. In other words, the decoded high frequency subband power estimation coefficient having a sum of squared differences E (J, id) can be said to be the optimal estimation coefficient for frequency band extension processing that is performed when decoding an exit code sequence.
[00331] Therefore, the high frequency subband pseudo-power difference calculation circuit 36 selects the sum of square differences of the K sums of differences E (J, id) of which the value is the smallest, and supplies the coefficient index indicating the decoded high-frequency subband power estimate coefficient corresponding to the sum of square differences of the same, for the high-frequency coding circuit 37.
[00332] In step S188, the high frequency coding circuit 37 encodes the coefficient index provided from the high frequency subband pseudopotence difference calculation circuit 36, and supplies the high frequency coded data obtained as a result of it for the multiplexing circuit 38.
[00333] For example, in step S188, entropy coding or the like is performed as to the coefficient index. Accordingly, the amount of high frequency encoded data information output to the decoding device 40 can be compressed. Note that the high frequency encoded data can be any type of information while the information can obtain an optimal decoded high frequency subband power estimate coefficient, and for example, the coefficient index can be used as data encoded data. high frequency, without change.
[00334] In step S189, the multiplexing circuit 38 multiplexes the low frequency encoded data provided from the low frequency encoding circuit 32 and the high frequency encoded data provided from the high frequency encoding circuit 37, issues the output code sequence obtained as a result of it, and ends the encoding processing.
[00335] Therefore, by emitting the high frequency encoded data, obtained by encoding the coefficient index, as an output code sequence, together with the low frequency encoded data, the decoding device 40 that receives the input of this code sequence output can obtain the decoded high frequency subband power estimation coefficient this is great for frequency band extension processing. Therefore, signals with higher sound quality can be obtained. [Functional Example of Decoding Device Configuration]
[00336] Also, the decoding device 40 for entering, as an input code sequence, and decoding, the output code sequence emitted from the encoding device 30 in Fig. 18, is configured as shown in Fig. 20, for example. Note that in Fig. 20, the portions corresponding to the case in Fig. 13 have the same reference numerals attached to it, and their description will be omitted.
[00337] The decoding device 40 in Fig. 20 is the same as the decoding device 40 in Fig. 13, from the point of being composed of demultiplexing circuit 41 to synthesis circuit 48, but differs from the decoding device decoding 40 in Fig. 13 from the point that the low frequency signal decoded from the low frequency decoding circuit 42 is not provided for the characteristic value calculation circuit 44.
[00338] In the decoding device 40 in Fig. 20, the high frequency decoding circuit 45 records in advance the same decoded high frequency subband power estimate coefficient as the subband power estimation coefficient of decoded high frequency recorded by the high frequency subband pseudopotency calculation circuit 35 in Fig. 18. This is to say, a set of the coefficient Aib (kb) and coefficient Bib serving as the coefficient of estimation of sub-power decoded high frequency band found by regression analysis in advance is correlated with the coefficient index and recorded.
[00339] The high frequency decoding circuit 45 decodes the high frequency encoded data provided from the demultiplex circuit 41, and supplies the decoded high frequency subband power estimate coefficient shown with the obtained coefficient index as a result of it for the decoded high-frequency subband power calculation circuit 46. [Decoding Device Decoding Processing]
[00340] Next, decoding processing performed with decoding device 40 in Fig. 20 will be described with reference to the flowchart in Fig. 21.
[00341] Decoding processing is initiated when the exit code sequence issued from the encoding device 30 is provided as an input code sequence to the decoding device 40. Note that processing in step S211 through step S213 is similar to the processing in step S131 to step S133 in Fig. 14, therefore, its description will be omitted.
[00342] In step S214, the characteristic value calculation circuit 44 uses the low frequency subband signal decoded from the subband division circuit 43 to calculate the characteristic value, and supplies it for the decoded high frequency subband power calculation circuit 46. Specifically, the characteristic value calculation circuit 44 computes the Expression (1) described above, and calculates the low frequency subband power, power ( ib, J) of frame J (where 0 <J) according to the characteristic value, for the various sub-bands ib on the low frequency side.
[00343] In step S215, the high frequency decoding circuit 45 performs decoding of the high frequency coded data provided from the demultiplexing circuit 41, and supplies the decoded high frequency subband power estimation coefficient shown by coefficient index obtained as a result of it for the decoded high frequency subband power calculation circuit 46. That is to say, of the multiple decoded high frequency subband power estimate coefficients recorded in advance in the circuit of high frequency decoding 45, the decoded high frequency subband power estimation coefficient shown in the coefficient index obtained through decoding is output.
[00344] In step S216, the decoded high frequency subband power calculation circuit 46 calculates the decoded high frequency subband power based on the characteristic value provided from the value calculation circuit feature 44 and the decoded high-frequency subband power estimation coefficient provided from the high-frequency decoding circuit 45, and supplies it for the decoded high-frequency signal generation circuit 47.
[00345] This is to say, the decoded high frequency subband power calculation circuit 46 uses the coefficients Aib (kb) and Bib serving as the decoded high frequency subband power estimate coefficients, and the low frequency subband power, power (kb, J), (where sb-3 <kb <sb) according to the characteristic value, to perform the computation in Expression (2) described above, and calculates the power of decoded high frequency subband. Therefore, the decoded high frequency subband power is obtained for each subband of the high frequency side characterized by the fact that the index is sb + 1 à and b.
[00346] In step S217, the decoded high frequency signal generation circuit 47 generates a decoded high frequency signal, based on the decoded low frequency subband signal supplied from the subband division circuit 43 and the supplied high frequency subband power decoded from the decoded high frequency subband power calculation circuit 46.
[00347] Specifically, the decoded high frequency signal generation circuit 47 performs the computation in Expression (1) described above, using the decoded low frequency subband signal, and calculates the low frequency subband power for each sub-band on the low frequency side. The decoded high frequency signal generation circuit 47 then uses the obtained low frequency subband power and the decoded high frequency subband power to perform the expression (3) described above, and calculates a value of G gain (ib, J) for each sub-band on the high frequency side.
[00348] Additionally, the decoded high frequency signal generation circuit 47 uses the gain value G (ib, J) and the decoded low frequency subband signal to perform Expression (5) and Expression (6) computation ) described above, and generates a high frequency subband signal x3 (ib, n) for each high frequency side subband.
[00349] That is to say, the decoded high frequency signal generation circuit 47 submits the decoded low frequency subband signal x (ib, n) to the amplitude adjustment, according to the proportion of the sub power - low frequency band and the decoded high frequency subband power, and as a result of it, still subject the obtained decoded low frequency subband signal x2 (ib, n) to frequency modulation. Therefore, the signal of the subband frequency component of the low frequency side is converted to a signal of the subband frequency component of the high frequency side, and a high frequency subband signal x3 ( ib, n) is obtained.
[00350] The processing thus obtaining the high frequency subband signals for each subband is as described in more detail below.
[00351] Let's say that four sub-bands organized in a matrix continuously in a frequency region is called a band block, and the frequency band is divided such that a band block (hereinafter particularly called low frequency block) it is composed of four sub-bands characterized by the fact that the indices on the low frequency side are sb to sb-3. At this moment, for example, the band composed of sub-bands characterized by the fact that the indices on the high frequency side are sb + 1 to that of sb + 4 is considered a band block. Note that hereinafter, a band block on the high frequency side, i.e. composed of sub-bands characterized by the fact that the indices are sb + 1 or greater, is particularly called a high frequency block.
[00352] Now, we are going to focus on a subband that makes up a high frequency block, and generates a high frequency subband signal from the same subband (hereinafter called the focus subband). First, the decoded high frequency signal generation circuit 47 identifies the subband of the low frequency block that is in the same position relationship as the position of the subband of interest in the high frequency block.
[00353] For example, if the index of the subband of interest is sb + 1, the subband of interest is the band having the lowest frequency of the high frequency block, and thereby a subband of low frequency block in the same position relation as the subband of interest becomes the subband characterized by the fact that the index is sb-3.
[00354] Therefore, after the low-frequency block sub-band in the same position relationship as the sub-band of interest having been identified, the low-frequency sub-band power and low-frequency sub-band signal decoded subband, and the decoded high frequency subband power of the subband of interest, are used to generate the high frequency subband signal of the subband of interest.
[00355] This is to say, the decoded high frequency subband power and low frequency subband power are replaced in expression (3), and a gain value according to the proportion of the same power is calculated. The calculated gain value is multiplied by the decoded low frequency subband signal, and the decoded low frequency subband signal that has been multiplied by the gain value is subjected to frequency modulation with the computation in Expression ( 6), and becomes the high frequency subband signal of the subband of interest.
[00356] With the above processing, a high frequency subband signal is obtained for each subband on the high frequency side. Subsequently, the decoded high frequency signal generation circuit 47 still performs computation in Expression ( 7) described above, finds the sum of the various high frequency subband signals obtained, and generates the decoded high frequency signal. The decoded high frequency signal generation circuit 47 supplies the decoded high frequency signal obtained for synthesis circuit 48, and processing is advanced from step S217 to step S218.
[00357] In step S218, synthesis circuit 48 synthesizes the low-frequency signal decoded from the low-frequency decoding circuit 42 and the high-frequency signal decoded from the decoded high-frequency signal generation circuit 47 , and output it as an exit signal. Subsequently, the decoding processing is then completed.
[00358] As described above, according to decoding device 40, a coefficient index is obtained from the high frequency encoded data that are obtained by undoing the multiplexing of the input code sequence, and the power estimate coefficient decoded high frequency subband data shown by the coefficient index of the same is used to calculate decoded high frequency subband power, and thereby the estimation accuracy for the high frequency subband power can be enhanced. Therefore, music signals can be played with higher sound quality. <4. Fourth Mode> [Coding Device Coding Processing]
[00359] Also, an example is described above of a case characterized by the fact that only the coefficient index is included in the high frequency coded data, but other information can be included.
[00360] For example, if the coefficient index is included in the encoded high frequency data, the decoded high frequency subband power estimate coefficient, which obtains the decoded high frequency subband power closest to the high frequency subband power of the effective high frequency signal can be known on the decoding device side 40.
[00361] However, the difference of approximately the same value as the difference in high frequency subband pseudo-power, diff power (ib, J), calculated with the high frequency subband difference calculation circuit 36 , occurs at the effective high frequency subband power (real value) and the decoded high frequency subband power (estimated value) obtained on the side of the decoding device 40.
[00362] Now, if not only the coefficient index, but also the difference in high frequency subband pseudo-power of each subband is included in the high frequency coded data, the general error of the high subband power frequency decoded as to the effective high frequency subband power can be known on the decoding device side 40. Therefore, the estimation accuracy for the high frequency subband power can be further improved using this error.
[00363] The encoding processing and decoding processing in the case of the high frequency subband pseudopotency difference being included in the high frequency encoded data will be described below with reference to the flowcharts in Fig. 22 and Fig. 23.
[00364] First, the coding processing performed with the coding device 30 in Fig. 18 will be described with reference to the flowchart in Fig. 22. Note that the processing in step S241 to S246 is similar to the processing in step S181 to S186 in Fig. 19, so its description will be omitted.
[00365] In step S247, the high frequency sub-band pseudo-power difference calculation circuit 36 computes the Expression (15) described above, and calculates the sum of the square difference E (J, id) for each coefficient of decoded high-frequency subband power estimate.
[00366] The high frequency subband pseudopotence difference calculation circuit 36 selects a sum of square differences that has the lowest value of the sums of square differences (J, id), and supplies, for the encoding circuit of high frequency 37, the coefficient index showing the power estimate coefficient of the high frequency coded subband corresponding to the sum of the square differences of the same.
[00367] Additionally, the high frequency sub-band pseudo-power difference calculation circuit 36 supplies the high-frequency sub-band pseudo-power difference (ib, J) for each sub-band, found for the coefficient of decoded high-frequency subband power estimate corresponding to the selected sum of the square differences, for the high-frequency coding circuit 37.
[00368] In step S248, the high frequency coding circuit 37 encodes the high frequency sub-band coefficient index and pseudo-power difference, provided from the high sub-band pseudo-power difference calculation circuit frequency 36, and supplies the high frequency encoded data obtained as a result of it for the multiplexing circuit 38.
[00369] Therefore, the difference in high frequency subband pseudopotency for each subband on the high frequency side, characterized by the fact that the index is sb + 1 à and b, ie the estimation error in the power of high frequency subband, is provided as high frequency encoded data for the decoding device 40.
[00370] After the high frequency coded data has been obtained, subsequently, processing in step S249 is performed and coding processing is finished, but processing in step S249 is similar to processing in step S189 in Fig. 19 thus description will be omitted.
[00371] As described above, when the high frequency subband pseudo-power difference is included in the high frequency encoded data, the accuracy estimate of the high frequency subband power can be further improved in the decoding device 40 , and music signals with higher sound quality can be obtained. [Decoding Device Decoding Processing]
[00372] Next, the decoding processing performed with the decoding device 40 in Fig. 20 will be described with reference to the flowchart in Fig. 23. Note that the processing in step S271 to S274 is similar to the processing in step S211 to S214 in Fig. 21, so its description will be omitted.
[00373] In step S275, the high frequency decoding circuit 45 performs decoding of the high frequency encoded data provided from the demultiplexing circuit 41. The high frequency decoding circuit 45 then supplies the sub power estimation coefficient - decoded high frequency band indicated by the coefficient index obtained through decoding, and the difference in high frequency subband pseudo-power of each subband obtained through decoding, for the subband power calculation circuit of high frequency decoded 46.
[00374] In step S276, the decoded high-frequency subband power calculation circuit 46 calculates the decoded high-frequency subband power based on the characteristic value provided from the value calculation circuit feature 44 and the decoded high-frequency subband power estimation coefficient provided from the high-frequency decoding circuit 45. Note that in step S276, processing similar to that in step S216 in Fig. 21 is performed.
[00375] In step S277, the decoded high-frequency subband power calculation circuit 46 adds the difference of high-frequency sub-band pseudo-power provided from the high-frequency decoding circuit 45 to the sub-power - decoded high frequency band, configures this as the final decoded high frequency subband power, and supplies it for the decoded high frequency signal generation circuit 47. That is to say, the subband power of decoded high frequency for each calculated subband is added to the difference in high frequency subband pseudo-power of the same subband.
[00376] Subsequently, processing in step S278 and step S279 is carried out and the decoding processing is finished, but the processing here is the same as that in step S217 and step S218 in Fig. 21, so its description will be omitted .
[00377] As described above, the decoding device 40 obtains the coefficient index and the high frequency subband pseudo-power difference from the high frequency coded data obtained by undoing the multiplexing of the input code sequence. The decoding device 40 then calculates the decoded high frequency subband power, using the decoded high frequency subband power estimate coefficient indicated by the coefficient index and the high frequency subband pseudo-power difference . Therefore, the accuracy of estimating the high frequency subband power can be improved, and music signals can be played with higher sound quality.
[00378] Note that the difference in the estimated values of the high-frequency subband power occurring between the coding device 30 and the decoding device 40, ie the difference in the high-frequency subband pseudo-power and the sub-power high frequency band (hereinafter called the difference of intra-device estimation) can be considered.
[00379] In such a case, for example, the difference in high frequency subband pseudo-power serving as the high frequency encoded data can be corrected with the difference in intra-device estimation, or the difference in intra-device estimation. can be included in the high frequency encoded data, and the difference in high frequency subband pseudo-power can be corrected by the difference in intra-device estimation on the side of the decoding device 40. Additionally, the difference in intra-device estimation can be recorded in advance on the decoding device 40 side, where the decoding device 40 adds the intra-device estimate difference to the high frequency subband pseudo-power difference, and makes corrections. Therefore, a high frequency signal decoded closer to the effective high frequency signal can be obtained. <5. Fifth modality>
[00380] Note that the coding device 30 in Fig. 18 is described such that the high frequency subband pseudo-power difference calculation circuit 36 selects, as the sum of squared differences E (J, id) as a indicator, an optimal sum of the square differences of multiple coefficient indices, but a different indicator from the sum of square differences can be used to select the coefficient index.
[00381] For example, an evaluated value that considers the mean square value, the maximum value, and the mean value, and so on the residual difference between the high frequency subband power and high subband pseudopotency frequency can be used as the indicator to select the coefficient index. In such a case, the coding device 30 in Fig. 18 performs coding processing shown in the flowchart in Fig. 24.
[00382] The coding processing with the coding device 30 will be described below with reference to the flow chart in Fig. 24. Note that the processing in step S301 to step S305 is similar to the processing in step S181 to step S185 in Fig. 19, so its description will be omitted. After processing in step S301 to step S305 having been performed, the high frequency subband pseudopotency for each subband is calculated for each of the K decoded high frequency subband power estimation coefficients.
[00383] In step S306, the high frequency subband pseudopotence difference calculation circuit 36 calculates an evaluated value Res (id, J) using the current frame J that is submitted to processing, for each of the K coefficients decoded high frequency subband power estimation.
[00384] Specifically, the high frequency subband pseudo-power difference calculation circuit 36 uses the high frequency subband signal for each subband provided from the subband splitting circuit 33 to effect computation similar to that in Expression (1) described above, and calculates the high-frequency subband power, power (ib, J) in table J. Note that according to the present modality, all sub-bands of the low frequency subband and the subband of high frequency subband signals are identified using the ib index.
[00385] After the high frequency subband power, power (ib, J) having been obtained, the high frequency subband pseudo-power difference calculation circuit 36 calculates the following expression (16), and calculates the residual mean frame value Resstd (id, J). [Expression 16]

[00386] This is to say, for each subband on the high frequency side characterized by the fact that the index is sb + 1 à and b, the difference of the high frequency subband power, power (ib, J) of frame J and the high frequency subband pseudopotency, potency (ib, id, J) is found, and the square sum of its difference becomes the mean square value Resstd (id, J). Note that the high frequency subband pseudopotency, potency (ib, id, J), represents a high frequency subband pseudopotency of frame J of the subband characterized by the fact that the index is ib, which is found for a decoded high frequency subband power estimation coefficient characterized by the fact that the coefficient index is id.
[00387] Next, the high frequency sub-band pseudo-power difference calculation circuit 36 calculates the following expression (17), and calculates the maximum residual value Resmax (id, J). [Expression 17]

[00388] Note that in Expression (17), maxib {| power (ib, J) - potest (ib, id, J) |} represents the largest of the absolute values of the difference between the high frequency subband power, power (ib, J), of each subband characterized by the fact that the index is sb + 1 to eb, and the high frequency subband pseudo-power, potest (ib, id, J). Consequently, the maximum value of the absolute values of the difference between the high frequency subband power, power (ib, J), in table J and the high frequency subband pseudo-power, potest (ib, id, J) , becomes the maximum residual value Resmax (id, J).
[00389] In addition, the high frequency subband pseudopotence difference calculation circuit 36 calculates the next Expression (18), and calculates the mean residual value Res (id, J). [Expression 18]

[00390] This is to say, for each subband on the high frequency side characterized by the fact that the index is sb + 1 à and b, the difference between the high frequency subband power, power (ib, J ) of table J, and the high frequency subband pseudopotency, potency (ib, id, J) is found, and the total sum of these differences is found. The absolute value of the values obtained by dividing the sum obtained from differences by the number of sub-bands (eb-sb) on the high frequency side becomes the average residual value Resave (id, J). The average residual value Resave (id, J) here represents the size of the average values of the estimated difference of several sub-bands from which the signal was taken into account.
[00391] Additionally, after obtaining the average square value Resstd (id, J), the maximum residual value Resmax (id, J), and the average residual value Resave (id, J), the circuit for calculating the difference of pseudopotency of high frequency subband 36 calculates the following expression (19), and calculates a final evaluated value Res (id, J). [Expression 19]

[00392] This is to say, the average square value Resstd (id, J), the maximum residual value Resmax (id, J), and the average residual value Resave (id, J) are added with the weighting factor, and if make a final evaluated value Res (id, J). Note that in Expression (19), Wmax and Wwave are pre-configured weightings, and for example it can be Wmax = 0.5, Wwave = 0.5 or the like.
[00393] The high frequency subband pseudopotence difference calculation circuit 36 performs the processing described above, and calculates the evaluated value Res (id, J) for each of the K subband power estimation coefficients high frequency decoded, ie for each of the K id coefficient indices.
[00394] In step S307, the high frequency subband pseudopotence difference calculation circuit 36 selects an id coefficient index, based on the evaluated value Res (id, J) for each id coefficient index found.
[00395] The evaluated value Res (id, J) obtained with the above processing indicates the degree of similarity between the high frequency subband power calculated from the effective high frequency signal, and the subband pseudopotence of high frequency calculated using the decoded high frequency subband power estimation coefficient characterized by the fact that the coefficient index is id. This is to say, it shows the size of the high frequency component estimation error.
[00396] Consequently, the lower the evaluated value Res (id, J) is, a decoded high frequency signal will be obtained that is closer to the effective high frequency signal, due to computation using the sub-power estimation coefficient. decoded high frequency band. Therefore, the high frequency subband pseudopotence difference calculation circuit 36 selects an evaluated value characterized by the fact that, of the K values evaluated Res (id, J), the value is minimal, and supplies, for the high frequency coding circuit 37, the coefficient index indicating the decoded high frequency subband power estimation coefficient corresponding to its evaluated value.
[00397] After the coefficient index being issued to the high frequency coding circuit 37, subsequently processing in step S308 and step S309 is carried out and the coding processing is finished, but this processing is similar to that in step S188 and step S189 in Fig. 19, so its description will be omitted.
[00398] As shown above, with the coding device 30, the evaluated value Res (id, J) calculated from the mean square value Resstd (id, J), the maximum residual value Resmax (id, J), and the Resave mean residual value (id, J) is used, and an optimal coefficient index for the decoded high frequency subband power estimation coefficient is selected.
[00399] Using the evaluated value Res (id, J), the precision estimation of the high frequency subband power can be evaluated using more evaluation scales when compared to the case of using the sum of the square differences, and by in between, a more appropriate decoded high-frequency subband power estimation coefficient can be selected. Accordingly, with the decoding device 40 receiving input from the output code sequence, a decoded high frequency subband power estimation coefficient that is optimal for frequency band extension processing can be obtained, and higher sound quality signals can be obtained. <Modification 1>
[00400] In addition, by performing the coding processing described above for each frame of the input signal, coefficient indices that differ for each consecutive frame may be high from the input signal.
[00401] This is to say, with consecutive frames that make up a constant region of the input signal, the high frequency subband power is approximately the same value for each frame, so for those frames the same coefficient index must be selected continuously. However, in the segments of these consecutive frames, the coefficient index selected per frame may change, and consequently, the high frequency component of the audio played on the side of the decoding device 40 may no longer be constant. Discomfort from a listening perspective can occur from the audio played.
[00402] Now, in case of selecting a coefficient index with the coding device 30, results of estimation of the high frequency component with the frame that is temporarily previous can also be considered. In such a case, the coding device 30 in Fig. 18 performs the coding processing shown in the flowchart in Fig. 25.
[00403] Coding processing with coding device 30 will be described below with reference to the flowchart in Fig. 25. Note that processing in step S331 to S336 is similar to processing in step S301 through step S306 in Fig. 24, therefore, the description will be omitted.
[00404] In step S337, the high frequency subband pseudo-power difference calculation circuit 36 calculates the evaluated value ResP (id, J) using a past frame and current frame.
[00405] Specifically, the high frequency subband pseudopotence difference calculation circuit 36 records the high frequency subband pseudopotency for each subband, obtained using the subband power estimation coefficient of high frequency decoded from the coefficient index finally selected for frame (J-1) which is temporarily a frame prior to frame J to be processed. The finally selected coefficient index is now the coefficient index which is encoded by the high frequency encoding circuit 37 and emitted by the decoding device 40.
[00406] From now on, let's say that the index coefficient id selected particularly in the table (J-1) is idselected (J-1). In addition, the description will be continued of where the subband's high frequency subband pseudopotency having the ib index (where sb + 1 <ib <eb), obtained using the subband power estimate coefficient high frequency decoded from the idselected coefficient index (J-1), as potency (ib, idselected (J-1), J-1).
[00407] The high frequency subband pseudopotence difference calculation circuit 36 first calculates the next Expression (20), and calculates an estimated mean square value ResPstd (id, J). [Expression 20]

[00408] This is to say, for each subband on the high frequency side characterized by the fact that the index is sb + 1 à and b, the difference is found between the high frequency subband pseudo-power, potest (ib , idselected (J-1), J-1) of the frame (J-1) and the high frequency subband pseudopotency, potency (ib, id, J) of the frame J. The square sum of their difference then becomes the estimated mean ResPstd square value (id, J). Note that the high frequency subband pseudopotency, potency (ib, id, J), represents the high frequency subband pseudopotency of frame J of the subband characterized by the fact that the index is ib, which is found for the decoded high frequency subband power estimation coefficient characterized by the fact that the coefficient index is id.
[00409] The estimated mean ResPstd square value (id, J) here is the sum of the square differences of the high frequency subband pseudo-power between temporarily consecutive frames, and thereby, the lower the estimated mean ResPstd square value (id , J) is, the less time change there will be in the estimated value of the high frequency component.
[00410] Next, the high frequency sub-band pseudo-power difference calculation circuit 36 calculates the following expression (21), and calculates an estimated maximum residual value ResPmax (id, J). [Expression 21]

[00411] Note that in Expression (21), maxib {| potestest (ib, idselected (J-1), J-1) -potestest (ib, id, J) |} represents the largest of the absolute values of the difference between high-frequency sub-band pseudopotency, potency (ib, idselected (J-1), J-1) of each sub-band characterized by the fact that the index is sb + 1 à and b, and the sub-band pseudopotence high frequency, potency (ib, id, J). Consequently, the maximum value of the absolute values of the difference in the high frequency subband pseudopower between temporarily consecutive frames becomes the maximum estimated residual value ResPmax (id, J).
[00412] The lower the maximum residual value estimated ResPmax (id, J) is, the closer the estimation results will be to the high frequency components between consecutive frames.
[00413] After the estimated maximum residual value ResPmax (id, J) having been obtained, then the high frequency sub-band pseudo-power difference calculation circuit 36 calculates the following expression (22), and calculates a residual value estimated average ResPave (id, J). [Expression 22]

[00414] This is to say, for each subband on the high frequency side characterized by the fact that the index is sb + 1 to b, the difference is found between the high frequency subband pseudo-power, potest (ib , idselected (J-1), J-1) of the frame (J-1) and the high-frequency subband pseudopotency, potent (ib, id, J) of frame J. The absolute value of the value obtained by dividing the sum of the differences in the various sub-bands by the number of sub-bands on the high frequency side (eb-sb) becomes the estimated average residual value ResPave (id, J). The estimated average residual value ResPave (id, J) here represents the average size of the difference in the estimated values of the sub-bands between frames from which the signal is taken into account.
[00415] Additionally, after obtaining the estimated average square value ResPstd (id, J), the maximum estimated residual value ResPmax (id, J), and the estimated average residual value ResPave (id, J), the circuit for calculating the difference in high frequency subband pseudopotence 36 calculates the following expression (23), and calculates the evaluated value ResP (id, J). [Expression 23]

[00416] This is to say, the estimated square average value ResPstd (id, J), the maximum estimated residual value ResPmax (id, J), and the estimated average residual value ResPave (id, J) are added with the weighting factor , and become the evaluated value ResP (id, J). Note that in Expression (23), Wmax and Wonda are preconfigured weights, and, for example, can be Wmax = 0.5, Wonda = 0.5 or the like.
[00417] Therefore, after the evaluated value ResP (id, J) that uses a past frame and a current frame having been calculated, the processing is advanced from step S337 to step S338.
[00418] In step S338, the high frequency subband pseudopotence difference calculation circuit 36 calculates the following expression (24), and calculates a final evaluated value Resall (id, J). [Expression 24]

[00419] This is to say, the evaluated value Res (id, J) and the evaluated value ResP (id, J) found are added with the weighting factor. Note that in Expression (24), Wp (J) is a weighting factor that is defined by the following expression (25), for example. [Expression 25]

[00420] Furthermore, the potency (J) in Expression (25) is a value defined by the following expression (26). [Expression 26]

[00421] The potency (J) here represents the average of the differences in the high frequency subband power of the frame (J-1) and frame J. Furthermore, from Expression (25), when Wp (J) is a value in a predetermined interval where the potency (J) is close to 0, Wp (J) becomes a value closer to 1 are the potency (J) becomes smaller, and becomes 0 when the potency (J ) is a value greater than the predetermined range.
[00422] Now, in the case that the potency (J) is a value within the predetermined range close to 0, the average of the difference of the high frequency subband power between consecutive frames frames becomes small of a certain value . In other words, the time variation of the high frequency components of the input signal is small, and thereby the current frame of the input signal is a constant region.
[00423] The more stable the frequency components of the input signal are, the closer the weighting factor Wp (J) is to a value that becomes closer to 1, and conversely, the more the high frequency components do not are stable, the closer the value becomes 0. Consequently, with the evaluated value Resall (id, J) shown in Expression (24), the less time variation of the high frequency components of the input signal, frequency, the greater the proportion of contribution of the evaluated value ResP (id, J), characterized by the fact that the result of the comparison from the estimation results of the high frequency components with the immediately preceding table serves as the evaluation scale, becomes.
[00424] Consequently, with the constant region of the input signal, a decoded high frequency subband power estimation coefficient, which can obtain estimation results close to the high frequency components in the immediately preceding frame is selected, and audio can be performed more naturally with high quality sound quality on the 40 decoding device side. Conversely, with a non-constant region of the input signal, the item for the evaluated value ResP (id, J) in the evaluated value Resall (id , J) becomes 0, and a decoded high frequency signal that is closer to the effective high frequency signal is obtained.
[00425] The high frequency subband pseudopotence difference calculation circuit 36 performs the above processing, and calculates an evaluated Resall value (id, J) for each of the K subband power estimation coefficients of high frequency decoded.
[00426] In step S339, the high frequency subband pseudo-power difference calculation circuit 36 selects an id coefficient index, based on the evaluated value Resall (id, J) for each of the power estimation coefficients of decoded high-frequency subband that is found.
[00427] The evaluated value Resall (id, J) obtained with the above processing linearly combines the evaluated value Res (id, J) and i evaluated value ResP (id, J), using the weighting factor. As described above, the lower the value of the evaluated value Res (id, J) is, a decoded high frequency signal can be obtained that is closer to the effective high frequency signal. In addition, the lower the value of the evaluated value ResP (id, J) is, a decoded high frequency signal can be obtained that is closer to the decoded high frequency signal of the immediately preceding frame.
[00428] Consequently, the lower the evaluated value Resall (id, J) is, the more appropriate decoded high frequency signal can be obtained. Therefore, of the K values evaluated Resall (id, J), the high frequency subband pseudopotence difference calculation circuit 36 selects an evaluated value having the lowest value, and supplies the coefficient index indicating the estimate coefficient of decoded high frequency subband power corresponding to the evaluated value of the same, for the high frequency coding circuit 37.
[00429] After the coefficient index has been selected, the processing in step S340 and step S341 is subsequently performed and the coding processing is finished, but the processing here is similar to step S308 and step S309 in Fig. 24, being thus its description will be omitted.
[00430] As shown above, with the coding device 30, the evaluated value Resall (id, J) which is obtained linearly by combining the evaluated value Res (id, J) and the evaluated value ResP (id, J) is used, and an optimal coefficient index of the decoded high frequency subband power estimation coefficient is selected.
[00431] Using the evaluated value Resall (id, J), similar to the case of using the evaluated value Res (id, J), a more appropriate coefficient of estimation of decoded high frequency subband power can be selected through more rating scales. Additionally, using the evaluated value Resall (id, J), temporal variations in the constant region of the high frequency components of the signal to be performed can be suppressed on the side of the decoding device 40, and a signal with higher sound quality can be obtained . <Modification 2>
[00432] Now, with frequency band extension processing, if higher sound quality for audio is to be achieved, the more sub-bands on the low frequency side become important from the perspective of listening. This is to say, of the various subbands on the high frequency side, the greater the accuracy of estimation of the subband closest to the low frequency side is, the higher the audio quality that can be played.
[00433] Now, in the case that an evaluated value is calculated for each decoded high frequency subband power estimation coefficient, the subbands on the far low frequency side can be weighted. In such a case, the coding device 30 in Fig. 18 performs coding processing shown in the flowchart in Fig. 26.
[00434] Coding processing by coding device 30 will be described below with reference to the flowchart in Fig. 26. Note that processing in step S371 through step S375 is similar to processing in step S331 through step S335 in Fig. 25, therefore, the description will be omitted.
[00435] In step S376, the high frequency subband pseudopotence difference calculation circuit 36 calculates an evaluated value ResWband (id, J) using a current J frame to be processed, for each of the K estimation coefficients decoded high-frequency subband power
[00436] Specifically, the high frequency subband pseudo-power difference calculation circuit 36 uses the high frequency subband signal from the various subbands provided from the subband division circuit 33 to effect computation similar to that in Expression (1) described above, and calculates the high frequency subband power, power (ib, J) in table J.
[00437] After the high frequency subband power, power (ib, J) having been obtained, the high frequency subband pseudo-power difference calculation circuit 36 calculates the following expression (27), and calculates an average residual value ResstdWband (id, J). [Expression 27]

[00438] This is to say, for each subband of the high frequency side characterized by the fact that the index is sb + 1 à and b, the difference between the high frequency subband power, power (ib, J ) of table J and the high frequency sub-band pseudopotency, potency (ib, id, J) is found, and the weighting factor Wband (ib) for each sub-band is multiplied by their difference. The square sum of the difference that is multiplied by the weighting factor Wband (ib) becomes the average square value ResstdWband (id, J).
[00439] Now, the weighting factor Wband (ib) (characterized by the fact that sb + 1 <ib <eb) is defined by the following expression (28), for example. The closer to the low frequency side the subband is, the higher the value of the weighting factor Wband (ib) becomes. [Expression 28]

[00440] Next, the high frequency subband pseudo-power difference calculation circuit 36 calculates the maximum residual value ResmaxWband (id, J). Specifically, the maximum value of the absolute value of those who have had the weighting factor Wband (ib) multiplied by the difference in the power of the high frequency sub-band, power (ib, J), of the various sub-bands characterized by the fact that the index is sb + 1 à and b and the high frequency subband pseudo-power, potest (ib, id, J), becomes the maximum residual value ResmaxWband (id, J).
[00441] In addition, the high frequency subband pseudopotence difference calculation circuit 36 calculates the average residual value ResaveWband (id, J).
[00442] Specifically, for each sub-band characterized by the fact that the index is sb + 1 to eb, the differences between the high-frequency sub-band power, power (ib, J) and sub-band pseudopotence of high frequency, potency (ib, id, J) are found and multiplied by the weighting factor Wband (ib), and the sum total of the differences multiplied by the weighting factor Wband (ib) is found. The absolute value of the value obtained by dividing the total sum of the differences obtained by the number of sub-bands (eb-sb) on the high frequency side is the average residual value ResaveWband (id, J).
[00443] Additionally, the high frequency subband pseudo-power difference calculation circuit 36 calculates the evaluated value ResWband (id, J). This is to say, the sum of the average square value ResstdWband (id, J), the maximum residual value ResmaxWband (id, J) which was multiplied by the weighting factor Wmax, and the average residual value ResaveWband (id, J) which was multiplied by the Wave weighting factor, it is the evaluated value ResWband (id, J).
[00444] In step S377, the high frequency subband pseudo-power difference calculation circuit 36 calculates the evaluated value ResPWband (id, J) using a past frame and a current frame.
[00445] Specifically, the high frequency sub-band pseudo-power difference calculation circuit 36 records the high-frequency sub-band pseudo-power for each sub-band, obtained using the sub-band power estimate coefficient of high frequency decoded from the coefficient index finally selected, to a frame (J-1) which is temporarily a frame preceding the J frame to be processed.
[00446] The high frequency subband pseudo-power difference calculation circuit 36 first calculates an estimated mean square value ResPstdWband (id, J). This is to say, for each subband on the high frequency side characterized by the fact that the index is sb + 1 to eb, the differences between the high frequency subband pseudopotency, potest (ib, idselected (J- 1), J-1), and the high frequency subband pseudopotency, potency (ib, id, J), are found and multiplied by the weighting factor Wband (ib). The square sum of the differences multiplied by the weighting factor Wband (ib) is the estimated average square value ResPstdWband (id, J).
[00447] Next, the high frequency subband pseudopotence difference calculation circuit 36 calculates an estimated maximum residual value ResPmaxWband (id, J). Specifically, the maximum value of the absolute values obtained by multiplying the weighting factor Wband (ib) by the differences between the high frequency subband pseudopotency, potency (ib, idselected (J-1), J-1) for each sub-band characterized by the fact that the index is sb + 1 à and b, and the high frequency sub-band pseudopotency, potent (ib, id, J), is considered as the maximum residual value estimated ResPmaxWband (id, J).
[00448] Next, the high frequency subband pseudopotence difference calculation circuit 36 calculates an estimated average residual value ResPaveWband (id, J). Specifically, the differences between the high frequency sub-band pseudopotency, potency (ib, idselected (J-1), J-1) for each sub-band characterized by the fact that the index is sb + 1 à and b, and the high frequency subband pseudopotency, potency (ib, id, J), are found, and multiplied by the weighting factor Wband (ib). The absolute value of the value obtained by dividing the total sum of the differences that are multiplied by the weighting factor Wband (ib) by the number of sub-bands (eb-sb) on the high frequency side is the estimated average residual value ResPaveWband (id, J ).
[00449] Additionally, the high frequency subband pseudopotence difference calculation circuit 36 finds the sum of the estimated average square value ResPstdWband (id, J), of the maximum estimated residual value ResPmaxWband (id, J) which has been multiplied by the weighting factor Wmax, and the estimated average residual value ResPaveWband (id, J) which was multiplied by the weighting factor Wonda is considered as the evaluated value ResPWband (id, J).
[00450] In step S378, the high frequency subband pseudo-power difference calculation circuit 36 adds the evaluated value ResWband (id, J) and the evaluated value ResPWband (id, J) which was multiplied by the weighting factor Wp (J) in Expression (25), and calculates a final evaluated value ResallWband (id, J). The evaluated value ResallWband (id, J) here is calculated for each of the K coefficients of decoded high frequency subband power estimates.
[00451] Subsequently, the processing in step S379 to step S381 is carried out and the coding processing is finished, but the processing here is similar to the processing in step S339 to step S341 in Fig. 25, thus describing it will be omitted. Note that in step S379, of the K coefficient indexes, the one with the lowest evaluated value ResallWband (id, J) is selected.
[00452] Therefore, each sub-band is weighted such that the weighting factor will be placed further away towards the sub-band on the low band side, and thereby audio with higher sound quality can be obtained on the side decoding device 40.
[00453] Note that with the above description, selection of the decoded high frequency subband power estimation coefficient is performed based on the evaluated value ResallWband (id, J), but the subband power estimation coefficient High frequency decoded can be selected based on the evaluated value ResallWband (id, J). <Modification 3>
[00454] Additionally, the human ear has the nature to better feel the frequency band when the amplitude (power) of the frequency band is large, so the evaluated value can be calculated for each subband power estimation coefficient high frequency decoded such that the weighting factor is placed in a subband having greater power.
[00455] In such a case, the coding device 30 in Fig. 18 performs the coding processing shown in the flowchart in Fig. 27. The coding processing with the coding device 30 will be described below with reference to the flowchart in Fig. 27. Note that the processing in step S401 to step S405 is similar to the processing in step S331 to step S335 in Fig. 25, so its description will be omitted.
[00456] In step S406, the high frequency subband pseudo-power difference calculation circuit 36 calculates an evaluated value ResWpower (id, J) using the current frame J that is submitted to processing, for each of the K coefficients of decoded high-frequency subband power estimates.
[00457] Specifically, the high frequency subband pseudo-power difference calculation circuit 36 uses a high frequency subband signal for each subband provided from the subband splitting circuit 33 to effect computation similar to Expression (1) described above, and calculates the power of the high frequency subband, power (ib, J), in table J.
[00458] After the high frequency subband power, power (ib, J), having been obtained, the high frequency subband pseudopotence difference calculation circuit 36 calculates the following expression (29), and calculates an average square value ResstdWpower (id, J). [Expression 29]

[00459] This is to say, the differences between the high frequency subband power, power (ib, J), and the high frequency subband pseudopotency, potency (ib, id, J), for each sub-band on the high frequency side characterized by the fact that the index is sb + 1 to b, are found, and the weighting factor Wpotency (power (ib, J)) for each sub-band is multiplied by these differences. The square sum of the differences multiplied by the weighting factor Wpotency (power (ib, J)) is the mean square value Resstd Wpotency (id, J).
[00460] Now, the weighting factor Wpotency (power (ib, J)) (where sb + 1 <ib <eb) is defined by the following expression (30), for example. The value of the weighting factor Wpotency (power (ib, J)) increases as the power of the high frequency subband, power (ib, J) of the same subband increases. [Expression 30]

[00461] Next, the high frequency subband pseudopower difference calculation circuit 36 calculates a maximum residual value Resmax Wpower to (id, J). Specifically, the maximum value of the absolute values obtained by multiplying the weighting factor Wpower (power (ib, J)) by the differences between the power of the high frequency subband, power (ib, J) for each subband characterized by the fact that the index is sb + 1 à and b, and the high frequency subband pseudopotency, potency (ib, id, J), is the maximum residual value Resmax Wpotency (id, J).
[00462] In addition, the high frequency subband pseudopotence difference calculation circuit 36 calculates an average residual value Resave Wpower (id, J).
[00463] Specifically, the differences between the high frequency subband power, power (ib, J) for each subband characterized by the fact that the index is sb + 1 à and b, and the subband pseudopotency high frequency, potency (ib, id, J), are found, and multiplied by the weighting factor Wpotency (power (ib, J)), and the sum total of the differences multiplied by the weighting factor Wpotency (power (ib, J )) is found. The absolute value of the value obtained by dividing the total sum obtained from the differences by the number of sub-bands (eb-sb) on the high frequency side is the average residual value Resave Wpotência (id, J).
[00464] In addition, the high frequency subband pseudo-power difference calculation circuit 36 calculates the evaluated value Res Wpotency a (id, J). This is to say, the sum of the mean square value Resstd Wpotency (id, J), the maximum residual value Resmax Wpotency (id, J) which was multiplied by the weighting factor Wmax, and the average residual value Resave Wpotency (id, J ) which was multiplied by the Wave weighting factor, is the evaluated value Res Wpotency (id, J).
[00465] In step S407, the high frequency subband pseudopotence difference calculation circuit 36 calculates an evaluated value ResP Wpotency (id, J) using a past frame and a current frame.
[00466] Specifically, the high frequency subband pseudopotence difference calculation circuit 36 records high frequency subband pseudopotency for each subband, obtained using the high subband power estimate coefficient decoded frequency of the finally selected coefficient index, for frame (J-1) which is temporarily a frame prior to frame J to be processed.
[00467] The high frequency subband pseudopotence difference calculation circuit 36 first calculates an estimated mean square value ResPstd Wpower (id, J). This is to say, for each subband on the high frequency side characterized by the fact that the index is sb + 1 to eb, the differences between the high frequency subband pseudopotency, potest (ib, idselected (J- 1), J-1), and high frequency subband pseudopotency, potency (ib, id, J), are found and multiplied by the weighting factor Wpotency (power (ib, J)). The square sum of the differences multiplied by the weighting factor Wpotency (power (ib, J)) is the estimated average square value ResPstdWpotency (id, J).
[00468] Next, the high frequency subband pseudopotence difference calculation circuit 36 calculates an estimated maximum residual value ResPmax Wpower (id, J). Specifically, that which is the absolute value of the maximum value of the differences between the high frequency sub-band pseudopotency, potency (ib, idselected (J-1), J-1) for each sub-band characterized by the fact that the index is sb + 1 à and b, and the high frequency sub-band pseudo-power, potest (ib, id, J), multiplied by the weighting factor Wpotency (power (ib, J)), is the maximum estimated residual value ResPmaxWpotency (id, J).
[00469] Next, the high frequency subband pseudo-power difference calculation circuit 36 calculates an estimated average residual value ResPaveWpotency (id, J). Specifically, the differences between the high frequency sub-band pseudopotency, potency (ib, idselected (J-1), J-1) for each sub-band characterized by the fact that the index is sb + 1 à and b, and the high frequency sub-band pseudopotency, potency (ib, id, J), are found, and multiplied by the weighting factor Wpotency (power (ib, J)). The absolute value of the value obtained by dividing the total sum of the differences that are multiplied by the weighting factor Wpotency (power (ib, J)) by the number of sub-bands (eb-sb) on the high frequency side is the estimated average residual value ResPaveWpotência (id, J).
[00470] In addition, the high frequency sub-band pseudo-power difference calculation circuit 36 finds the sum of the estimated average square value ResPstdWpotency (id, J), of the maximum residual value estimated ResPmaxWpower (id, J) than multiplied by the weighting factor Wmax, and the estimated average residual value ResPaveWpotency (id, J) which was multiplied by the weighting factor Wave, and considers this as the evaluated value ResWpotency (id, J).
[00471] In step S408, the high frequency sub-band pseudopower difference calculation circuit 36 adds the evaluated value ResWpotency (id, J) and the evaluated value ResPWpower (id, J) which were multiplied by the weighting factor Wp (J) in Expression (25), and calculates a final evaluated value ResallWpotency (id, J). The evaluated value ResallWpotency (id, J) here is calculated for each of the K coefficients of decoded high frequency subband power estimates.
[00472] Subsequently, the processing in step S409 to step S411 is carried out and the coding processing is finished, but the processing here is similar to the processing in step S339 to step S341 in Fig. 25, thus describing it will be omitted. Note that in step S409, of the K coefficient indexes, the one with the lowest evaluated value ResallWpotência (id, J) is selected.
[00473] Therefore, such that the weighting factor will be placed further away in a sub-band having greater power, each sub-band is weighted, and thereby, audio with higher sound quality can be obtained on the device side. decoding method 40.
[00474] Note that with the above description, the selection of the decoded high frequency subband power estimation coefficient is made based on the evaluated value ResallWpotency (id, J), but the subvalue power estimation coefficient decoded high frequency band can be selected based on the evaluated value ResWpotency (id, J). <6. Sixth modality> [Coefficient Learning Device Configuration]
[00475] Now, a set of coefficient Aib (kb) and coefficient Bib serving as the coefficients of decoded high-frequency subband power estimates is correlated to the coefficient index and recorded on decoding device 40 in Fig. 20. For example, when the coefficients of high frequency subband power estimates decoded from 128 coefficient indices having been recorded on the decoding device 40, a large region is required with the recording region for the memory that records these coefficients of decoding. decoded high-frequency subband power estimates and the like.
[00476] Therefore, a portion of several decoded high frequency subband power estimate coefficients can be caused to be shared coefficients, and the recording region required to record the subband power estimate coefficients of high frequency decoded can be made smaller. In such a case, the coefficient learning device that finds coefficients of high frequency subband power estimates decoded through learning is configured as shown in Fig. 28, for example.
[00477] The coefficient learning device 81 is composed of a subband division circuit 91, a high frequency subband power calculation circuit 92, a characteristic value calculation circuit 93, and a coefficient estimation circuit 94.
[00478] Multiple pieces of data tuning or the like used for learning are provided to the coefficient learning device 81 as broadband learning signals. A broadband learning signal is a signal that includes multiple high frequency subband components and multiple low frequency subband components.
[00479] The subband division circuit 91 is composed of a bandpass filter or the like, divides the broadband learning signal provided into multiple subband signals, and supplies them to the power calculation circuit of high frequency subband 92 and for the characteristic value calculation circuit 93. Specifically, the high frequency subband signal of each subband on the high frequency side characterized by the fact that the index is sb +1 à and b is provided for the high frequency subband power calculation circuit 92, and the low frequency subband signal of each subband on the low frequency side characterized by the fact that the index is sb-3 à is provided for the characteristic value calculation circuit 93.
[00480] The high frequency subband power calculation circuit 92 calculates the high frequency subband power of the various high frequency subband signals provided from the subband split circuit 91, and supplies it for the coefficient estimation circuit 94. The characteristic value calculation circuit 93 calculates the low frequency subband power as a characteristic value, based on the various low frequency subband signals provided from subband division circuit 91, and supplies it to coefficient estimation circuit 94.
[00481] The coefficient estimation circuit 94 generates a decoded high frequency subband power estimation coefficient using the high frequency subband power from the high frequency subband power calculation circuit 92 and the characteristic value from the characteristic value calculation circuit 93 to perform regression analysis, and output it to the decoding device 40. [Description of Coefficient Learning Processing]
[00482] Next, the coefficient learning processing carried out by the coefficient learning device 81 will be described with reference to the flowchart in Fig. 29.
[00483] In step S431, the subband division circuit 91 divides each of the multiple broadband learning signals provided into multiple subband signals. The subband split circuit 91 supplies the high frequency subband signal of the subband characterized by the fact that the index is sb + 1 à and b for the high frequency subband power calculation circuit 92, and supplies the low frequency subband signal of the subband characterized by the fact that the index is sb-3 to sb for the characteristic value calculation circuit 93.
[00484] In step S432, the high frequency subband power calculation circuit 92 performs computation similar to Expression (1) described above and calculates the high frequency subband power for the various subband signals high frequency inputs supplied from subband division circuit 91, and supplies them to coefficient estimation circuit 94.
[00485] In step S433, the characteristic value calculation circuit 93 performs computation similar to Expression (1) described above and calculates the low frequency subband power as a characteristic value for the various subband signals low frequency inputs supplied from subband division circuit 91, and supplies them to coefficient estimation circuit 94.
[00486] Therefore, high frequency subband power and low frequency subband power are provided for coefficient estimation circuit 94 for the various frames of the multiple broadband learning signals.
[00487] In step S434, the coefficient estimation circuit 94 performs regression analysis using a minimum square method, and calculates the coefficient Aib (kb) and coefficient Bib for each subband ib on the high frequency side (where sb +1 <ib <eb) characterized by the fact that the index is sb + 1 à eb.
[00488] Note that with regression analysis, the low frequency subband power provided from the characteristic value calculation circuit 93 is an explanatory variable, and the high frequency subband power provided from the high frequency subband power calculation circuit 92 is an explained variable. In addition, regression analysis is performed using low frequency subband power and high frequency subband power for all frames, which make up all of the broadband learning signals provided for the coefficient learning device 81 .
[00489] In step S435, the coefficient estimation circuit 94 uses the coefficient Aib (kb) and coefficient Bib found for each subband ib to find the residual vector for each frame of the broadband learning signal.
[00490] For example, the coefficient estimation circuit 94 subtracts the sum of the total sum of the low frequency subband power, power (kb, J), which was multiplied by the coefficient Aib (kb) (where sb-3 <kb <sb), and the Bib coefficient, of a high frequency subband power, power (ib, J), for each subband ib (where sb + 1 <ib <eb) of frame J, and gets the residue. The vector composed of the residues of each sub-band ib in table J is the residual vector.
[00491] Note that the residual vector is calculated for all frames that make up the entire broadband learning signal provided for the coefficient learning device 81.
[00492] In step S436, the coefficient estimation circuit 94 normalizes the residual vectors found from the various frames. For example, the coefficient estimation circuit 94 normalizes the residual vector by finding the residual dispersion value of an ib subband of the residual vectors for all frames, and divides the residual of an ib subband of the various residual vectors by square root of the dispersion value for each subband.
[00493] In step S437, the coefficient estimation circuit 94 agglomerates the residual vectors for all frames normalized by k means or the like.
[00494] For example, an average frequency envelope for all frames, obtained when the estimation of the high frequency subband power is performed using the coefficient Aib (kb) and coefficient Bib, it is called a medium frequency envelope SA . In addition, we will say that a predetermined frequency envelope having a higher power than the average frequency envelope SA is an SH enveloped frequency, and that a predetermined frequency envelope having a lower power than the frequency envelope medium SA is an SL enveloped frequency.
[00495] At this time, residual vector agglomeration is performed such that each of the residual vectors of the coefficients, for which a frequency envelope close to the medium frequency envelope SA, the frequency envelope SH, and frequency envelope SL is obtained , belongs to a CA agglomerate, CH agglomerate, and CL agglomerate, respectively. In other words, agglomeration is performed such that the residual vector for each frame belongs to one of the CA agglomerates, CH agglomerates, or CL agglomerates.
[00496] With a frequency band extension processing that estimates high frequency components based on the correlation between low frequency components and high frequency components, when calculating the residual vector using the Aib coefficient (kb) and Bib coefficient obtained with the regression analysis, the farther the subband is towards the high frequency side, the greater the residue becomes, based on its characteristics. Therefore, if the residual vector is agglomerated without change, a greater weighting factor is placed in the more distant sub-bands on the high frequency side, and processing is performed.
[00497] Conversely, with the coefficient learning device 81, normalizing the residual vector with the dispersion value of the residual value for each subband, the dispersion of the residues of each subband at first glance are equal, and agglomeration is carried out by weighting the various sub-bands equally.
[00498] In step S438, the coefficient estimation circuit 94 selects one of the agglomerates of the CA agglomerate, CH agglomerate, or CL agglomerate, as an agglomerate to be processed.
[00499] In step S439, the coefficient estimation circuit 94 uses the residual vector frame belonging to the selected cluster with the cluster to be processed, to calculate the Aib coefficient (kb) and Bib coefficient of the various ib sub-bands (where sb + 1 <ib <eb), with regression analysis.
[00500] This is to say, if we say that the frame of the residual vector belonging to the cluster to be processed is called a frame to be processed, the low frequency subband power and the high frequency subband power for all the frames to be processed there are then explanatory variables and explained variables, and regression analysis using a minimum square method is performed. Therefore, an Aib coefficient (kb) and Bib coefficient are obtained for each subband ib.
[00501] In step S440, the coefficient estimation circuit 94 uses the coefficient Aib (kb) and coefficient Bib obtained with the processing in step S439 for all frames to be processed, and finds the residual vector. Note that in step S440, processing similar to that in step S435 is performed, and the residual vectors for the various frames to be processed are found.
[00502] In step S441, the coefficient estimation circuit 94 normalizes the residual vectors of the various frames to be processed that are obtained in the processing in step S440, performing similar processing as that in step S436. That is to say, the residue is divided by the square root of the dispersion value and normalization of the residual vectors is carried out by each sub-band.
[00503] In step S442, the coefficient estimation circuit 94 agglomerates the residual vectors for all the frames to be processed that have been normalized, by k means or the like. The number of clusters here is defined as follows. For example, in the coefficient learning device 81, in the case of generating 128 coefficients indexes of decoded high-frequency subband power estimates, the number of frames to be processed is multiplied by 128, and the number obtained by dividing it. o by the number of all frames is the number of clusters. Now, the number of all frames is the total number of all frames of all broadband learning signals provided for coefficient learning device 81.
[00504] In step S443, the coefficient estimation circuit 94 finds a vector of center of gravity vector for the various clusters obtained with the processing in step S442.
[00505] For example, a cluster obtained by the cluster in step S442 corresponds to the coefficient index, and in the coefficient learning device 81, a coefficient index is assigned to each cluster, and the subband power estimate coefficient high frequency decoded from each coefficient index is found.
[00506] Specifically, let's say that in step S438 the CA agglomerate is selected as the agglomerate to be processed, and in step S442 F the number of agglomerates are obtained by the agglomeration in step S442. Now, if we focus on a CF cluster outside of the F clusters, the number of high frequency subband power estimate coefficients decoded from the CF cluster coefficient index is configured as the Aib coefficient (kb) which is an item correlation of the coefficient Aib (ib) found for the CA cluster in step S439. In addition, the sum of the vector performing reverse normalization processing (reverse normalization) performed in step S441 for the CF cluster center of gravity vector found in step S443 and the Bib coefficient found in step S439 is the Bib coefficient which is an item constant of the decoded high frequency subband power estimation coefficient. The reverse normalization here is, in the case that the normalization performed in step S441 divides the residue with the square root of the dispersion value for each sub-band, for example, processing that multiplies the same value as the normalization time (square root of the dispersion value for each subband) the elements of the CF cluster center of gravity vector.
[00507] This is to say, the set of the coefficient Aib (kb) obtained in step S439 and the coefficient Bib found as described above becomes the estimated coefficient of the coefficient index of the high frequency sub-band power decoded from the CF cluster . Consequently, each of the F number of clusters obtained through the cluster has a shared Aib coefficient (kb) found for the CA cluster, as a linear correlation item of the decoded high frequency subband power estimation coefficient.
[00508] In step S444, coefficient learning device 81 determines whether all agglomerates of CA agglomerate, CH agglomerate, and CL agglomerate form or not processed as agglomerates to be processed. In step S444, if the determination is made even though all the agglomerates have not been processed, the processing returns to step S438, and the processing described above is repeated. That is to say, the next cluster is selected as the one to be processed, and a decoded high-frequency subband power estimate coefficient is calculated.
[00509] Conversely, in step S444, in case the determination is made that all clusters have been processed, a predetermined number of decoded high-frequency subband power estimation coefficients to be found is obtained, and by means of in addition, processing is advanced to step S445.
[00510] In step S445, the coefficient estimation circuit 94 emits the decoded high frequency subband coefficient index and power estimate coefficient found for the decoding device 40 and causes it to be recorded, and the processing of coefficient learning is completed.
[00511] For example, of the decoded high-frequency subband power estimation coefficients issued to the decoding device 40, several have the same Aib coefficient (kb) as the linear correlation item. Therefore, as for the Aib coefficient (kb) that they share, the coefficient learning device 81 corresponds to an index of the linear correlation item (indicator) that is information identifying the Aib coefficient (kb) of the same, and as to the index of coefficient, corresponds to the index of the linear correlation item and Bib coefficient, which is a constant item.
[00512] The coefficient learning device 81 supplies the corresponding index of the linear correlation item (indicator) and Aib coefficient (kb) and the corresponding coefficient index and index of the linear correlation item (indicator) and Bib coefficient for the device decoder 40, and writes them to memory within the high frequency decoding circuit 45 of the decoding device 40. Therefore, in recording multiple coefficients of decoded high frequency subband power estimates, considering shared linear correlation items , if an index of the linear correlation item (indicator) is stored in the recording region for the various decoded high frequency subband power estimation coefficients, the recording region can be kept considerably smaller.
[00513] In this case, the index of the linear correlation item and coefficient Aib (kb) are correlated and recorded in memory within the high frequency decoding circuit 45, and thereby, the index of the linear correlation item and coefficient Bib it can be obtained from the coefficient index, and the Aib coefficient (kb) can be obtained from the index of the linear correlation item.
[00514] Note that as a result of analysis by the present applicant, we can see that even if three patterns or more or less of the linear correlation items of the decoded high-frequency subband power estimation coefficients are shared, there is little deterioration of sound quality from the perspective of listening to audio submitted to frequency band extension processing. Consequently, according to coefficient learning device 81, the sound quality of the vocals after a frequency band extension processing is not deteriorated, and a recording region needed to record the subband power estimate coefficient decoded high frequency can be less.
[00515] As shown above, coefficient learning device 81 generates and outputs the decoded high frequency subband power estimate coefficient for each coefficient index from the provided broadband learning signal.
[00516] Note that the coefficient learning processing in Fig. 29 is described as normalizing a residual vector, but in one or both of step S436 or step S441, normalizing the residual vector does not have to be performed.
[00517] Furthermore, an arrangement can be made characterized by the fact that normalization of the residual vector is performed, and sharing of the linear correlation items of the decoded high frequency subband power estimate coefficient is not performed. In such a case, after normalization processing in step S436, the normalized residual vector is clustered in the same number of clusters as the number of decoded high-frequency subband power estimate coefficients to be found. Tables of residual vectors belonging to the various clusters are used, regression analysis is performed for each cluster, and coefficients of decoded high-frequency subband power estimates are generated for the various clusters.
[00518] The series of processing described above can be performed by hardware or can be performed by software. In the case of executing the series of processing by software, a program composing the software of the same is installed from a program recording medium on a computer that was built with dedicated hardware or a general purpose personal computer or the like, for example. example, which can perform various types of functions by various types of programs being installed.
[00519] Fig. 30 is a block diagram showing an example of hardware configuration of the computer that performs a series of processing described above with a program.
[00520] In the computer, a CPU 101, ROM (Read-Only Memory) 102, and RAM (Random Access Memory) 103 are mutually connected through one of the multiple communication paths 104.
[00521] An input / output interface 105 is still connected to bus 104. An input unit 106 composed of a keyboard, mouse, microphone or the like, an output unit 107 composed of a monitor, speaker or the like , a storage unit 108 composed of a hard disk or non-volatile memory or the like, a communication unit 109 composed of a network interface or the like, and an operating mechanism 110 for operating a removable media 111 such as a magnetic disk , optical disk, magnetic optical disk, or semiconductor memory or the like, are connected to the input / output interface 105.
[00522] With a computer configured as described above, for example, CPU 101 loads the program stored in storage unit 108 to RAM 103, via the input / output interface 105 and bus 104, and executes it, and through In addition, the series of processing described above is carried out.
[00523] The program that the computer (CPU 101) runs is recorded on removable media 111 which is a media package composed of a magnetic disk (including floppy disk), optical disk (CD-ROM (Compact Disk - Read Only Memory) ), DVD (Digital Versatile Disc) or the like), magnetic optical disc, or semiconductor memory or the like, for example, or is provided via a cable or wireless transmission medium such as a local area network, the Internet , or broadcast by digital satellite broadcast.
[00524] The program is installed on the storage unit 108 via the input / output interface 105, mounting the removable media 111 on the operating mechanism 110. In addition, the program can be received with the communication unit 109 via a means of cable or wireless transmission, and installed on storage unit 108. Additionally, the program can be installed in advance on a ROM 102 or storage unit 108.
[00525] Note that the program that the computer executes may be a program that performs processing in a serial manner over time in the order described in this Specification, or it may be a program characterized by the fact that the processing is carried out in parallel, or in necessary time such as when called, or the like.
[00526] Note that the modalities of the present invention are not restricted to the modalities described above, and, several modifications can be made within the essence of the present invention. Reference Signal List 10 frequency band extension device 11 low-pass filter 12 delay circuit 13, 13-1 to 13-N band-pass filters 14 characteristic value calculation circuit 15 sub-power estimation circuit frequency band 16 high frequency signal generation circuit 17 high pass filter 18 signal addition unit 20 coefficient learning device 21, 21-1 to 21- (K + N) band pass filter 22 power calculation circuit high frequency subband band 23 characteristic value calculation circuit 24 coefficient estimation circuit 30 coding device 31 low pass filter 32 low frequency coding circuit 33 subband division circuit 34 value calculation circuit characteristic 35 high frequency subband pseudopotency calculation circuit 36 high frequency subband pseudopotence difference circuit 37 high frequency coding circuit ta frequency 38 multiplexing circuit 40 decoding device 41 demultiplexing circuit 42 low frequency decoding circuit 43 subband division circuit 44 characteristic value calculation circuit 45 high frequency decoding circuit 46 power calculation circuit decoded high-frequency sub-band circuit 47 decoded high-frequency signal generation circuit 48 synthesis circuit 50 coefficient learning device 51 low-pass filter 52 sub-band division circuit 53 characteristic value calculation circuit 54 circuit high-frequency sub-band pseudopotency calculation circuit 55 high-frequency sub-band pseudopotency difference calculation circuit 56 high-frequency sub-band pseudo-power difference cluster circuit 57 coefficient estimation circuit 101 CPU 102 ROM 103 RAM 104 BUS 105 INPUT / OUTPUT INTERFACE 106 INPUT UNIT 107 OUTPUT UNIT 108 STORAGE UNIT 109 COMMUNICATION UNIT 110 DRIVE 111 REMOVABLE MEDIA

权利要求:
Claims (11)
[0001]
1. Decoding device (40) for extending frequency bands, characterized by the fact that it comprises: demultiplexing means (41) configured to demultiplex encoded data introduced in at least low frequency encoded data and an index; low frequency decoding means (42) configured to decode low frequency encoded data to generate a low frequency music signal; subband splitting means (43) configured to divide the low frequency signal band into a plurality of low frequency subbands to generate a low frequency subband signal for each of the low frequency subbands frequency; and generation means (47) configured to generate a high frequency signal based on the low frequency index and subband signal; where the index is information that indicates an estimate coefficient used to generate the high frequency signal; and wherein the generation means comprise characteristic value calculation means (44) configured to calculate the characteristic value expressing a characteristic of the encoded data using at least one of the low frequency subband signal and the low signal frequency; high frequency subband power calculation means (46) configured to calculate a high frequency subband power from a high frequency subband signal of the high frequency subband through calculation using the value characteristic and the estimation coefficient considering each one of a plurality of high frequency sub-bands composing the high frequency signal band; and high frequency signal generation means (47) configured to generate the high frequency signal based on the high frequency subband power and low frequency subband signal; wherein the high frequency subband power calculation means (46) are configured to calculate the high frequency subband power of the high frequency subband by linearly combining a plurality of characteristic values using the estimate coefficient prepared for each of the high frequency sub-bands.
[0002]
2. Decoding device according to claim 1, characterized by the fact that the index is obtained, in a device that encodes an input music signal and outputs the encoded data, based on the input music signal before encoding, and the estimated high frequency signal of the input music signal.
[0003]
3. Decoding device according to claim 1, characterized by the fact that the means of calculating the characteristic value calculate a low frequency subband power of the low frequency subband signal for each of the sub-bands. low frequency bands as the characteristic value increases.
[0004]
4. Decoding device according to claim 1, characterized by the fact that the index is information that indicates the estimation coefficient by which the high frequency subband power is closest to the high frequency subband power obtained from the high frequency signal of the signal entered before encoding is obtained as a result of comparison between the high frequency subband power obtained from the high frequency signal of the music signal introduced before encoding and the sub power -high band generated based on the estimate coefficient of a plurality of estimate coefficients.
[0005]
5. Decoding device according to claim 4, characterized by the fact that the encoded data still includes different information indicating the difference between the high frequency subband power obtained from the high frequency signal of the music signal introduced before the coding, and the high frequency subband power generated based on the estimation coefficient.
[0006]
6. Decoding device according to claim 5, characterized by the fact that the difference information has been encoded.
[0007]
7. Decoding device according to claim 5, characterized by the fact that the high frequency subband power calculation means add the difference indicated with the difference information included in the encoded data to the subband power of high frequency obtained by calculating using the characteristic value and the estimation coefficient; and wherein the high frequency signal generating means generates the high frequency signal based on the high frequency subband power to which the difference has been added, and the low frequency subband signal.
[0008]
8. Decoding device according to claim 1, characterized by the fact that the generating means generate the high frequency signal based on information obtained when decoding the encoded indica.
[0009]
9. Decoding device according to claim 8, characterized by the fact that the index was subjected to entropy coding.
[0010]
10. Method of decoding to extend frequency bands, characterized by comprising: a demultiplexing step (S211) arranged to demultiplex encoded data introduced in at least low frequency encoded data and an index; a low frequency decoding step (S212) arranged to decode the low frequency encoded data to generate a low frequency music signal; a subband splitting step (S213) arranged to divide the low frequency music signal band into a plurality of low frequency subbands to generate a low frequency subband signal for each of the subband low frequency bands; and a generation step arranged to generate a high frequency signal based on the low frequency index and subband signal; where the index is information indicating an estimate coefficient used to generate the high frequency signal; and the generating step comprises: calculating (S214) a characteristic value that expresses a characteristic of the encoded data using at least one of the low frequency subband signal and low frequency music signal; calculate (S216) a high frequency subband power of a high frequency subband signal from a high frequency subband by calculating using a characteristic value and an estimation coefficient each with respect to a plurality high frequency sub-bands forming the high frequency signal band; and generating (S217) the high frequency signal based on the high frequency subband power and the low frequency subband signal; where the high frequency subband power of the high frequency subband signal of the high frequency subband is calculated by linearly combining the plurality of characteristic values using the estimate coefficient prepared for each of the sub- high frequency bands.
[0011]
11. Computer-readable storage media characterized by the fact that it comprises instructions that, when executed by a processor, cause the computer to perform a method as defined in claim 10.

类似技术:

---

公开号 | 公开日 | 专利标题

---

BR112012007389B1|2020-12-22|device and decoding methods to extend frequency bands, and computer-readable storage media

---

RU2589293C2|2016-07-10|Encoding device and method, decoding device and method and program

---

RU2571565C2|2015-12-20|Signal processing device and signal processing method, encoder and encoding method, decoder and decoding method and programme

---

BR112012025570B1|2020-11-17|DECODER APPLIANCE AND DECODING METHOD

---

JP6179571B2|2017-08-16|Decoding apparatus and method, and program

同族专利:

---

公开号 | 公开日

---

KR20190058705A|2019-05-29|

---

HK1172139A1|2013-04-12|

---

CN103996402B|2017-05-24|

---

KR101654402B1|2016-09-05|

---

US9691410B2|2017-06-27|

---

CN103996401A|2014-08-20|

---

EP2993667B1|2017-08-09|

---

AU2021215291A1|2021-09-02|

---

BR112012007389A2|2016-12-06|

---

EP3584794A1|2019-12-25|

---

KR101665283B1|2016-10-11|

---

KR102110727B1|2020-05-13|

---

US20160019911A1|2016-01-21|

---

HK1200237A1|2015-07-31|

---

WO2011043227A1|2011-04-14|

---

TW201131555A|2011-09-16|

---

AU2016253695B2|2019-04-18|

---

KR101982999B1|2019-05-27|

---

CN103996402A|2014-08-20|

---

TWI480862B|2015-04-11|

---

EP3584794B1|2021-10-27|

---

CN102576544B|2014-06-25|

---

KR101882002B1|2018-07-26|

---

CO6541531A2|2012-10-16|

---

HK1200236A1|2015-07-31|

---

EP2472512B1|2015-11-04|

---

CN102576544A|2012-07-11|

---

RU2549116C2|2015-04-20|

---

KR101786416B1|2017-10-17|

---

MY161609A|2017-04-28|

---

JP2011237751A|2011-11-24|

---

JP5754899B2|2015-07-29|

---

EP3232438A1|2017-10-18|

---

EP3232438B1|2019-09-11|

---

CA2775387C|2019-06-04|

---

EP2472512A4|2013-02-20|

---

CN103996401B|2018-01-16|

---

EP2993667A1|2016-03-09|

---

KR20170117210A|2017-10-20|

---

EP3968322A2|2022-03-16|

---

US20120243526A1|2012-09-27|

---

AU2019206091A1|2019-08-08|

---

KR20150140877A|2015-12-16|

---

AU2019206091B2|2021-05-13|

---

KR101681860B1|2016-12-01|

---

RU2012112445A|2013-10-27|

---

EP2472512A1|2012-07-04|

---

KR20160140965A|2016-12-07|

---

KR20180085831A|2018-07-27|

---

AU2010304440A1|2012-05-31|

---

KR20150140878A|2015-12-16|

---

KR20120082414A|2012-07-23|

---

US9208795B2|2015-12-08|

---

AU2016253695A1|2016-11-24|

---

CA2775387A1|2011-04-14|

引用文献:

---

公开号 | 申请日 | 公开日 | 申请人 | 专利标题

---

---

US4628529A|1985-07-01|1986-12-09|Motorola, Inc.|Noise suppression system|

---

JPH03254223A|1990-03-02|1991-11-13|Eastman Kodak Japan Kk|Analog data transmission system|

---

JP2655485B2|1994-06-24|1997-09-17|日本電気株式会社|Voice cell coding device|

---

JP3498375B2|1994-07-20|2004-02-16|ソニー株式会社|Digital audio signal recording device|

---

JP3189598B2|1994-10-28|2001-07-16|松下電器産業株式会社|Signal combining method and signal combining apparatus|

---

JPH1020888A|1996-07-02|1998-01-23|Matsushita Electric Ind Co Ltd|Voice coding/decoding device|

---

JP3328532B2|1997-01-22|2002-09-24|シャープ株式会社|Digital data encoding method|

---

US6073100A|1997-03-31|2000-06-06|Goodridge, Jr.; Alan G|Method and apparatus for synthesizing signals using transform-domain match-output extension|

---

SE512719C2|1997-06-10|2000-05-02|Lars Gustaf Liljeryd|A method and apparatus for reducing data flow based on harmonic bandwidth expansion|

---

US6415251B1|1997-07-11|2002-07-02|Sony Corporation|Subband coder or decoder band-limiting the overlap region between a processed subband and an adjacent non-processed one|

---

JP4132154B2|1997-10-23|2008-08-13|ソニー株式会社|Speech synthesis method and apparatus, and bandwidth expansion method and apparatus|

---

US6445750B1|1998-04-22|2002-09-03|Lucent Technologies Inc.|Technique for communicating digitally modulated signals over an amplitude-modulation frequency band|

---

US6424938B1|1998-11-23|2002-07-23|Telefonaktiebolaget L M Ericsson|Complex signal activity detection for improved speech/noise classification of an audio signal|

---

SE9903553D0|1999-01-27|1999-10-01|Lars Liljeryd|Enhancing conceptual performance of SBR and related coding methods by adaptive noise addition and noise substitution limiting |

---

US6829360B1|1999-05-14|2004-12-07|Matsushita Electric Industrial Co., Ltd.|Method and apparatus for expanding band of audio signal|

---

JP3454206B2|1999-11-10|2003-10-06|三菱電機株式会社|Noise suppression device and noise suppression method|

---

CA2290037A1|1999-11-18|2001-05-18|Voiceage Corporation|Gain-smoothing amplifier device and method in codecs for wideband speech and audio signals|

---

SE0001926D0|2000-05-23|2000-05-23|Lars Liljeryd|Improved spectral translation / folding in the subband domain|

---

AU6274801A|2000-06-14|2001-12-24|Kenwood Corp|Frequency interpolating device and frequency interpolating method|

---

SE0004163D0|2000-11-14|2000-11-14|Coding Technologies Sweden Ab|Enhancing perceptual performance or high frequency reconstruction coding methods by adaptive filtering|

---

US7430212B2|2001-02-13|2008-09-30|Paradyne Corporation|System and method for improved data transmission speed by fixing the lower corner frequency at a frequency above voice band in a symmetric DSL transmission system|

---

JP2002268698A|2001-03-08|2002-09-20|Nec Corp|Voice recognition device, device and method for standard pattern generation, and program|

---

SE0101175D0|2001-04-02|2001-04-02|Coding Technologies Sweden Ab|Aliasing reduction using complex-exponential-modulated filter banks|

---

JP4231987B2|2001-06-15|2009-03-04|日本電気株式会社|Code conversion method between speech coding / decoding systems, apparatus, program, and storage medium|

---

MXPA03002115A|2001-07-13|2003-08-26|Matsushita Electric Ind Co Ltd|Audio signal decoding device and audio signal encoding device.|

---

US6988066B2|2001-10-04|2006-01-17|At&T Corp.|Method of bandwidth extension for narrow-band speech|

---

US6895375B2|2001-10-04|2005-05-17|At&T Corp.|System for bandwidth extension of Narrow-band speech|

---

CN100395817C|2001-11-14|2008-06-18|松下电器产业株式会社|Encoding device and decoding device|

---

JP3926726B2|2001-11-14|2007-06-06|松下電器産業株式会社|Encoding device and decoding device|

---

KR100587517B1|2001-11-14|2006-06-08|마쯔시다덴기산교 가부시키가이샤|Audio coding and decoding|

---

JP3870193B2|2001-11-29|2007-01-17|コーディングテクノロジーズアクチボラゲット|Encoder, decoder, method and computer program used for high frequency reconstruction|

---

EP1470550B1|2002-01-30|2008-09-03|Matsushita Electric Industrial Co., Ltd.|Audio encoding and decoding device and methods thereof|

---

JP2003255973A|2002-02-28|2003-09-10|Nec Corp|Speech band expansion system and method therefor|

---

US20030187663A1|2002-03-28|2003-10-02|Truman Michael Mead|Broadband frequency translation for high frequency regeneration|

---

US7447631B2|2002-06-17|2008-11-04|Dolby Laboratories Licensing Corporation|Audio coding system using spectral hole filling|

---

DE60327039D1|2002-07-19|2009-05-20|Nec Corp|AUDIO DEODICATION DEVICE, DECODING METHOD AND PROGRAM|

---

KR100723753B1|2002-08-01|2007-05-30|마츠시타 덴끼 산교 가부시키가이샤|Audio decoding apparatus and audio decoding method based on spectral band replication|

---

JP4728568B2|2002-09-04|2011-07-20|マイクロソフトコーポレーション|Entropy coding to adapt coding between level mode and run length / level mode|

---

JP3881943B2|2002-09-06|2007-02-14|松下電器産業株式会社|Acoustic encoding apparatus and acoustic encoding method|

---

SE0202770D0|2002-09-18|2002-09-18|Coding Technologies Sweden Ab|Method of reduction of aliasing is introduced by spectral envelope adjustment in real-valued filterbanks|

---

CN100492492C|2002-09-19|2009-05-27|松下电器产业株式会社|Audio decoding apparatus and method|

---

US7330812B2|2002-10-04|2008-02-12|National Research Council Of Canada|Method and apparatus for transmitting an audio stream having additional payload in a hidden sub-channel|

---

EP2665294A2|2003-03-04|2013-11-20|Core Wireless Licensing S.a.r.l.|Support of a multichannel audio extension|

---

US7318035B2|2003-05-08|2008-01-08|Dolby Laboratories Licensing Corporation|Audio coding systems and methods using spectral component coupling and spectral component regeneration|

---

US20050004793A1|2003-07-03|2005-01-06|Pasi Ojala|Signal adaptation for higher band coding in a codec utilizing band split coding|

---

KR20050027179A|2003-09-13|2005-03-18|삼성전자주식회사|Method and apparatus for decoding audio data|

---

US7844451B2|2003-09-16|2010-11-30|Panasonic Corporation|Spectrum coding/decoding apparatus and method for reducing distortion of two band spectrums|

---

DE10345995B4|2003-10-02|2005-07-07|Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V.|Apparatus and method for processing a signal having a sequence of discrete values|

---

CN101556800B|2003-10-23|2012-05-23|松下电器产业株式会社|Acoustic spectrum coding method and apparatus, spectrum decoding method and apparatus, acoustic signal transmission apparatus, acoustic signal reception apparatus|

---

KR100587953B1|2003-12-26|2006-06-08|한국전자통신연구원|Packet loss concealment apparatus for high-band in split-band wideband speech codec, and system for decoding bit-stream using the same|

---

WO2005111568A1|2004-05-14|2005-11-24|Matsushita Electric Industrial Co., Ltd.|Encoding device, decoding device, and method thereof|

---

CN1954363B|2004-05-19|2011-10-12|松下电器产业株式会社|Encoding device and method thereof|

---

AT474310T|2004-05-28|2010-07-15|Nokia Corp|MULTI-CHANNEL AUDIO EXPANSION|

---

KR100608062B1|2004-08-04|2006-08-02|삼성전자주식회사|Method and apparatus for decoding high frequency of audio data|

---

US7716046B2|2004-10-26|2010-05-11|Qnx Software Systems , Inc.|Advanced periodic signal enhancement|

---

US20060106620A1|2004-10-28|2006-05-18|Thompson Jeffrey K|Audio spatial environment down-mixer|

---

SE0402651D0|2004-11-02|2004-11-02|Coding Tech Ab|Advanced methods for interpolation and parameter signaling|

---

EP2752849B1|2004-11-05|2020-06-03|Panasonic Intellectual Property Management Co., Ltd.|Encoder and encoding method|

---

BRPI0517780A2|2004-11-05|2011-04-19|Matsushita Electric Ind Co Ltd|scalable decoding device and scalable coding device|

---

KR100657916B1|2004-12-01|2006-12-14|삼성전자주식회사|Apparatus and method for processing audio signal using correlation between bands|

---

JP5224017B2|2005-01-11|2013-07-03|日本電気株式会社|Audio encoding apparatus, audio encoding method, and audio encoding program|

---

KR100708121B1|2005-01-22|2007-04-16|삼성전자주식회사|Method and apparatus for bandwidth extension of speech|

---

TWI319565B|2005-04-01|2010-01-11|Qualcomm Inc|Methods, and apparatus for generating highband excitation signal|

---

CN101138274B|2005-04-15|2011-07-06|杜比国际公司|Envelope shaping of decorrelated signals|

---

US20070005351A1|2005-06-30|2007-01-04|Sathyendra Harsha M|Method and system for bandwidth expansion for voice communications|

---

JP4899359B2|2005-07-11|2012-03-21|ソニー株式会社|Signal encoding apparatus and method, signal decoding apparatus and method, program, and recording medium|

---

KR100813259B1|2005-07-13|2008-03-13|삼성전자주식회사|Method and apparatus for encoding/decoding input signal|

---

WO2007026821A1|2005-09-02|2007-03-08|Matsushita Electric Industrial Co., Ltd.|Energy shaping device and energy shaping method|

---

US8396717B2|2005-09-30|2013-03-12|Panasonic Corporation|Speech encoding apparatus and speech encoding method|

---

BRPI0617447A2|2005-10-14|2012-04-17|Matsushita Electric Ind Co Ltd|transform encoder and transform coding method|

---

KR100958144B1|2005-11-04|2010-05-18|노키아 코포레이션|Audio Compression|

---

JP4876574B2|2005-12-26|2012-02-15|ソニー株式会社|Signal encoding apparatus and method, signal decoding apparatus and method, program, and recording medium|

---

JP4863713B2|2005-12-29|2012-01-25|富士通株式会社|Noise suppression device, noise suppression method, and computer program|

---

US7953604B2|2006-01-20|2011-05-31|Microsoft Corporation|Shape and scale parameters for extended-band frequency coding|

---

US7590523B2|2006-03-20|2009-09-15|Mindspeed Technologies, Inc.|Speech post-processing using MDCT coefficients|

---

JP4976381B2|2006-03-31|2012-07-18|パナソニック株式会社|Speech coding apparatus, speech decoding apparatus, and methods thereof|

---

EP2012305B1|2006-04-27|2011-03-09|Panasonic Corporation|Audio encoding device, audio decoding device, and their method|

---

EP2017830B9|2006-05-10|2011-02-23|Panasonic Corporation|Encoding device and encoding method|

---

JP2007316254A|2006-05-24|2007-12-06|Sony Corp|Audio signal interpolation method and audio signal interpolation device|

---

KR20070115637A|2006-06-03|2007-12-06|삼성전자주식회사|Method and apparatus for bandwidth extension encoding and decoding|

---

JP2007333785A|2006-06-12|2007-12-27|Matsushita Electric Ind Co Ltd|Audio signal encoding device and audio signal encoding method|

---

WO2007148925A1|2006-06-21|2007-12-27|Samsung Electronics Co., Ltd.|Method and apparatus for adaptively encoding and decoding high frequency band|

---

US8260609B2|2006-07-31|2012-09-04|Qualcomm Incorporated|Systems, methods, and apparatus for wideband encoding and decoding of inactive frames|

---

EP2063418A4|2006-09-15|2010-12-15|Panasonic Corp|Audio encoding device and audio encoding method|

---

JP4918841B2|2006-10-23|2012-04-18|富士通株式会社|Encoding system|

---

US8295507B2|2006-11-09|2012-10-23|Sony Corporation|Frequency band extending apparatus, frequency band extending method, player apparatus, playing method, program and recording medium|

---

JP5141180B2|2006-11-09|2013-02-13|ソニー株式会社|Frequency band expanding apparatus, frequency band expanding method, reproducing apparatus and reproducing method, program, and recording medium|

---

KR101565919B1|2006-11-17|2015-11-05|삼성전자주식회사|Method and apparatus for encoding and decoding high frequency signal|

---

JP5339919B2|2006-12-15|2013-11-13|パナソニック株式会社|Encoding device, decoding device and methods thereof|

---

JP4984983B2|2007-03-09|2012-07-25|富士通株式会社|Encoding apparatus and encoding method|

---

JP2008261978A|2007-04-11|2008-10-30|Toshiba Microelectronics Corp|Reproduction volume automatically adjustment method|

---

US8015368B2|2007-04-20|2011-09-06|Siport, Inc.|Processor extensions for accelerating spectral band replication|

---

KR101355376B1|2007-04-30|2014-01-23|삼성전자주식회사|Method and apparatus for encoding and decoding high frequency band|

---

WO2009001874A1|2007-06-27|2008-12-31|Nec Corporation|Audio encoding method, audio decoding method, audio encoding device, audio decoding device, program, and audio encoding/decoding system|

---

WO2009004727A1|2007-07-04|2009-01-08|Fujitsu Limited|Encoding apparatus, encoding method and encoding program|

---

JP5045295B2|2007-07-30|2012-10-10|ソニー株式会社|Signal processing apparatus and method, and program|

---

US8041577B2|2007-08-13|2011-10-18|Mitsubishi Electric Research Laboratories, Inc.|Method for expanding audio signal bandwidth|

---

CA2697920C|2007-08-27|2018-01-02|Telefonaktiebolaget L M Ericsson |Transient detector and method for supporting encoding of an audio signal|

---

MX2010001394A|2007-08-27|2010-03-10|Ericsson Telefon Ab L M|Adaptive transition frequency between noise fill and bandwidth extension.|

---

DK3401907T3|2007-08-27|2020-03-02|Ericsson Telefon Ab L M|Method and apparatus for perceptual spectral decoding of an audio signal comprising filling in spectral holes|

---

EP2209116B8|2007-10-23|2014-08-06|Clarion Co., Ltd.|Device and method for high-frequency range interpolation of an audio signal|

---

KR101373004B1|2007-10-30|2014-03-26|삼성전자주식회사|Apparatus and method for encoding and decoding high frequency signal|

---

JP4733727B2|2007-10-30|2011-07-27|日本電信電話株式会社|Voice musical tone pseudo-wideband device, voice musical tone pseudo-bandwidth method, program thereof, and recording medium thereof|

---

EP2214163A4|2007-11-01|2011-10-05|Panasonic Corp|Encoding device, decoding device, and method thereof|

---

US20090132238A1|2007-11-02|2009-05-21|Sudhakar B|Efficient method for reusing scale factors to improve the efficiency of an audio encoder|

---

RU2449386C2|2007-11-02|2012-04-27|Хуавэй Текнолоджиз Ко., Лтд.|Audio decoding method and apparatus|

---

US8515767B2|2007-11-04|2013-08-20|Qualcomm Incorporated|Technique for encoding/decoding of codebook indices for quantized MDCT spectrum in scalable speech and audio codecs|

---

WO2009059631A1|2007-11-06|2009-05-14|Nokia Corporation|Audio coding apparatus and method thereof|

---

JP2009116275A|2007-11-09|2009-05-28|Toshiba Corp|Method and device for noise suppression, speech spectrum smoothing, speech feature extraction, speech recognition and speech model training|

---

US8504377B2|2007-11-21|2013-08-06|Lg Electronics Inc.|Method and an apparatus for processing a signal using length-adjusted window|

---

US8688441B2|2007-11-29|2014-04-01|Motorola Mobility Llc|Method and apparatus to facilitate provision and use of an energy value to determine a spectral envelope shape for out-of-signal bandwidth content|

---

ES2629453T3|2007-12-21|2017-08-09|Iii Holdings 12, Llc|Encoder, decoder and coding procedure|

---

US20100280833A1|2007-12-27|2010-11-04|Panasonic Corporation|Encoding device, decoding device, and method thereof|

---

AT518224T|2008-01-04|2011-08-15|Dolby Int Ab|AUDIO CODERS AND DECODERS|

---

JP5448850B2|2008-01-25|2014-03-19|パナソニック株式会社|Encoding device, decoding device and methods thereof|

---

KR101413968B1|2008-01-29|2014-07-01|삼성전자주식회사|Method and apparatus for encoding audio signal, and method and apparatus for decoding audio signal|

---

US8433582B2|2008-02-01|2013-04-30|Motorola Mobility Llc|Method and apparatus for estimating high-band energy in a bandwidth extension system|

---

US20090201983A1|2008-02-07|2009-08-13|Motorola, Inc.|Method and apparatus for estimating high-band energy in a bandwidth extension system|

---

AU2009220321B2|2008-03-03|2011-09-22|Intellectual Discovery Co., Ltd.|Method and apparatus for processing audio signal|

---

KR101449434B1|2008-03-04|2014-10-13|삼성전자주식회사|Method and apparatus for encoding/decoding multi-channel audio using plurality of variable length code tables|

---

EP2104096B1|2008-03-20|2020-05-06|Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V.|Apparatus and method for converting an audio signal into a parameterized representation, apparatus and method for modifying a parameterized representation, apparatus and method for synthesizing a parameterized representation of an audio signal|

---

KR20090122142A|2008-05-23|2009-11-26|엘지전자 주식회사|A method and apparatus for processing an audio signal|

---

EP2294770B1|2008-06-20|2013-08-07|Rambus, Inc.|Frequency responsive bus coding|

---

MX2011000372A|2008-07-11|2011-05-19|Fraunhofer Ges Forschung|Audio signal synthesizer and audio signal encoder.|

---

KR101706009B1|2008-07-11|2017-02-22|프라운호퍼 게젤샤프트 쭈르 푀르데룽 데어 안겐반텐 포르슝 에. 베.|Audio encoder, audio decoder, method for encoding and decoding an audio signal. audio stream and computer program|

---

JP5203077B2|2008-07-14|2013-06-05|株式会社エヌ・ティ・ティ・ドコモ|Speech coding apparatus and method, speech decoding apparatus and method, and speech bandwidth extension apparatus and method|

---

JP5419876B2|2008-08-08|2014-02-19|パナソニック株式会社|Spectrum smoothing device, coding device, decoding device, communication terminal device, base station device, and spectrum smoothing method|

---

JP2010079275A|2008-08-29|2010-04-08|Sony Corp|Device and method for expanding frequency band, device and method for encoding, device and method for decoding, and program|

---

WO2010028299A1|2008-09-06|2010-03-11|Huawei Technologies Co., Ltd.|Noise-feedback for spectral envelope quantization|

---

US8352279B2|2008-09-06|2013-01-08|Huawei Technologies Co., Ltd.|Efficient temporal envelope coding approach by prediction between low band signal and high band signal|

---

WO2010028292A1|2008-09-06|2010-03-11|Huawei Technologies Co., Ltd.|Adaptive frequency prediction|

---

US8798776B2|2008-09-30|2014-08-05|Dolby International Ab|Transcoding of audio metadata|

---

GB2466201B|2008-12-10|2012-07-11|Skype Ltd|Regeneration of wideband speech|

---

GB0822537D0|2008-12-10|2009-01-14|Skype Ltd|Regeneration of wideband speech|

---

CN101770776B|2008-12-29|2011-06-08|华为技术有限公司|Coding method and device, decoding method and device for instantaneous signal and processing system|

---

SG172976A1|2009-01-16|2011-08-29|Dolby Int Ab|Cross product enhanced harmonic transposition|

---

US8457975B2|2009-01-28|2013-06-04|Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V.|Audio decoder, audio encoder, methods for decoding and encoding an audio signal and computer program|

---

JP4945586B2|2009-02-02|2012-06-06|株式会社東芝|Signal band expander|

---

US8463599B2|2009-02-04|2013-06-11|Motorola Mobility Llc|Bandwidth extension method and apparatus for a modified discrete cosine transform audio coder|

---

JP5564803B2|2009-03-06|2014-08-06|ソニー株式会社|Acoustic device and acoustic processing method|

---

CN101853663B|2009-03-30|2012-05-23|华为技术有限公司|Bit allocation method, encoding device and decoding device|

---

EP2239732A1|2009-04-09|2010-10-13|Fraunhofer-Gesellschaft zur Förderung der Angewandten Forschung e.V.|Apparatus and method for generating a synthesis audio signal and for encoding an audio signal|

---

CO6440537A2|2009-04-09|2012-05-15|Fraunhofer Ges Forschung|APPARATUS AND METHOD TO GENERATE A SYNTHESIS AUDIO SIGNAL AND TO CODIFY AN AUDIO SIGNAL|

---

JP5223786B2|2009-06-10|2013-06-26|富士通株式会社|Voice band extending apparatus, voice band extending method, voice band extending computer program, and telephone|

---

US8515768B2|2009-08-31|2013-08-20|Apple Inc.|Enhanced audio decoder|

---

JP5754899B2|2009-10-07|2015-07-29|ソニー株式会社|Decoding apparatus and method, and program|

---

US8600749B2|2009-12-08|2013-12-03|At&T Intellectual Property I, L.P.|System and method for training adaptation-specific acoustic models for automatic speech recognition|

---

US8447617B2|2009-12-21|2013-05-21|Mindspeed Technologies, Inc.|Method and system for speech bandwidth extension|

---

EP2357649B1|2010-01-21|2012-12-19|Electronics and Telecommunications Research Institute|Method and apparatus for decoding audio signal|

---

JP5375683B2|2010-03-10|2013-12-25|富士通株式会社|Communication apparatus and power correction method|

---

EP2555188B1|2010-03-31|2014-05-14|Fujitsu Limited|Bandwidth extension apparatuses and methods|

---

JP5609737B2|2010-04-13|2014-10-22|ソニー株式会社|Signal processing apparatus and method, encoding apparatus and method, decoding apparatus and method, and program|

---

JP5652658B2|2010-04-13|2015-01-14|ソニー株式会社|Signal processing apparatus and method, encoding apparatus and method, decoding apparatus and method, and program|

---

JP5850216B2|2010-04-13|2016-02-03|ソニー株式会社|Signal processing apparatus and method, encoding apparatus and method, decoding apparatus and method, and program|

---

US8793126B2|2010-04-14|2014-07-29|Huawei Technologies Co., Ltd.|Time/frequency two dimension post-processing|

---

US9047875B2|2010-07-19|2015-06-02|Futurewei Technologies, Inc.|Spectrum flatness control for bandwidth extension|

---

PL3544008T3|2010-07-19|2020-08-24|Dolby International Ab|Processing of audio signals during high frequency reconstruction|

---

US8560330B2|2010-07-19|2013-10-15|Futurewei Technologies, Inc.|Energy envelope perceptual correction for high band coding|

---

JP6075743B2|2010-08-03|2017-02-08|ソニー株式会社|Signal processing apparatus and method, and program|

---

JP2012058358A|2010-09-07|2012-03-22|Sony Corp|Noise suppression apparatus, noise suppression method and program|

---

JP5707842B2|2010-10-15|2015-04-30|ソニー株式会社|Encoding apparatus and method, decoding apparatus and method, and program|

---

WO2012052802A1|2010-10-18|2012-04-26|Nokia Corporation|An audio encoder/decoder apparatus|

---

JP5743137B2|2011-01-14|2015-07-01|ソニー株式会社|Signal processing apparatus and method, and program|

---

JP5704397B2|2011-03-31|2015-04-22|ソニー株式会社|Encoding apparatus and method, and program|

---

JP6024077B2|2011-07-01|2016-11-09|ヤマハ株式会社|Signal transmitting apparatus and signal processing apparatus|

---

JP5975243B2|2011-08-24|2016-08-23|ソニー株式会社|Encoding apparatus and method, and program|

---

JP6037156B2|2011-08-24|2016-11-30|ソニー株式会社|Encoding apparatus and method, and program|

---

JP5942358B2|2011-08-24|2016-06-29|ソニー株式会社|Encoding apparatus and method, decoding apparatus and method, and program|

---

JP5845760B2|2011-09-15|2016-01-20|ソニー株式会社|Audio processing apparatus and method, and program|

---

CN103918030B|2011-09-29|2016-08-17|杜比国际公司|High quality detection in the FM stereo radio signal of telecommunication|

---

JPWO2013154027A1|2012-04-13|2015-12-17|ソニー株式会社|Decoding device and method, audio signal processing device and method, and program|

---

JP5997592B2|2012-04-27|2016-09-28|株式会社Ｎｔｔドコモ|Speech decoder|

---

RU2649944C2|2012-07-02|2018-04-05|Сони Корпорейшн|Decoding device, decoding method, coding device, coding method and program|

---

AU2013284705B2|2012-07-02|2018-11-29|Sony Corporation|Decoding device and method, encoding device and method, and program|

---

TWI517142B|2012-07-02|2016-01-11|Sony Corp|Audio decoding apparatus and method, audio coding apparatus and method, and program|

---

US9437198B2|2012-07-02|2016-09-06|Sony Corporation|Decoding device, decoding method, encoding device, encoding method, and program|

---

JP2014123011A|2012-12-21|2014-07-03|Sony Corp|Noise detector, method, and program|

---

EP3048609A4|2013-09-19|2017-05-03|Sony Corporation|Encoding device and method, decoding device and method, and program|JP3870193B2|2001-11-29|2007-01-17|コーディングテクノロジーズアクチボラゲット|Encoder, decoder, method and computer program used for high frequency reconstruction|

---

JP5754899B2|2009-10-07|2015-07-29|ソニー株式会社|Decoding apparatus and method, and program|

---

JP5652658B2|2010-04-13|2015-01-14|ソニー株式会社|Signal processing apparatus and method, encoding apparatus and method, decoding apparatus and method, and program|

---

JP5850216B2|2010-04-13|2016-02-03|ソニー株式会社|Signal processing apparatus and method, encoding apparatus and method, decoding apparatus and method, and program|

---

JP5609737B2|2010-04-13|2014-10-22|ソニー株式会社|Signal processing apparatus and method, encoding apparatus and method, decoding apparatus and method, and program|

---

JP6075743B2|2010-08-03|2017-02-08|ソニー株式会社|Signal processing apparatus and method, and program|

---

JP5707842B2|2010-10-15|2015-04-30|ソニー株式会社|Encoding apparatus and method, decoding apparatus and method, and program|

---

JP5743137B2|2011-01-14|2015-07-01|ソニー株式会社|Signal processing apparatus and method, and program|

---

JP5704397B2|2011-03-31|2015-04-22|ソニー株式会社|Encoding apparatus and method, and program|

---

EP2523357B1|2011-05-12|2013-09-18|Siemens Aktiengesellschaft|Subsea data communication system and method|

---

JP5942358B2|2011-08-24|2016-06-29|ソニー株式会社|Encoding apparatus and method, decoding apparatus and method, and program|

---

JP5975243B2|2011-08-24|2016-08-23|ソニー株式会社|Encoding apparatus and method, and program|

---

JP6037156B2|2011-08-24|2016-11-30|ソニー株式会社|Encoding apparatus and method, and program|

---

CN103035248B|2011-10-08|2015-01-21|华为技术有限公司|Encoding method and device for audio signals|

---

AU2013284705B2|2012-07-02|2018-11-29|Sony Corporation|Decoding device and method, encoding device and method, and program|

---

RU2649944C2|2012-07-02|2018-04-05|Сони Корпорейшн|Decoding device, decoding method, coding device, coding method and program|

---

KR102243688B1|2013-04-05|2021-04-27|돌비 인터네셔널 에이비|Audio encoder and decoder for interleaved waveform coding|

---

US9520140B2|2013-04-10|2016-12-13|Dolby Laboratories Licensing Corporation|Speech dereverberation methods, devices and systems|

---

JP6305694B2|2013-05-31|2018-04-04|クラリオン株式会社|Signal processing apparatus and signal processing method|

---

JP2015050685A|2013-09-03|2015-03-16|ソニー株式会社|Audio signal processor and method and program|

---

EP3048609A4|2013-09-19|2017-05-03|Sony Corporation|Encoding device and method, decoding device and method, and program|

---

CN105761723B|2013-09-26|2019-01-15|华为技术有限公司|A kind of high-frequency excitation signal prediction technique and device|

---

JP6425097B2|2013-11-29|2018-11-21|ソニー株式会社|Frequency band extending apparatus and method, and program|

---

AU2014371411A1|2013-12-27|2016-06-23|Sony Corporation|Decoding device, method, and program|

---

JP2016038435A|2014-08-06|2016-03-22|ソニー株式会社|Encoding device and method, decoding device and method, and program|

---

KR20170041449A|2015-10-07|2017-04-17|주식회사 만도|Radar apparatus for vehicle and method for estimating angle of target using the same|

---

KR20180056032A|2016-11-18|2018-05-28|삼성전자주식회사|Signal processing processor and controlling method thereof|

---

US10896684B2|2017-07-28|2021-01-19|Fujitsu Limited|Audio encoding apparatus and audio encoding method|

---

WO2020014354A1|2018-07-10|2020-01-16|John Rankin|System and method for indexing sound fragments containing speech|

---

JPWO2020179472A1|2019-03-05|2020-09-10|

法律状态:
2019-01-08| B06F| Objections, documents and/or translations needed after an examination request according [chapter 6.6 patent gazette]|

---

2019-09-03| B06U| Preliminary requirement: requests with searches performed by other patent offices: procedure suspended [chapter 6.21 patent gazette]|

---

2020-06-23| B06A| Patent application procedure suspended [chapter 6.1 patent gazette]|

---

2020-10-06| B09A| Decision: intention to grant [chapter 9.1 patent gazette]|

---

2020-12-22| B16A| Patent or certificate of addition of invention granted [chapter 16.1 patent gazette]|Free format text: PRAZO DE VALIDADE: 10 (DEZ) ANOS CONTADOS A PARTIR DE 22/12/2020, OBSERVADAS AS CONDICOES LEGAIS. |

优先权:

---

申请号 | 申请日 | 专利标题

---

JP2009-233814|2009-10-07|

---

JP2009233814|2009-10-07|

---

JP2010092689|2010-04-13|

---

JP2010-092689|2010-04-13|

---

JP2010162259A|JP5754899B2|2009-10-07|2010-07-16|Decoding apparatus and method, and program|

---

JP2010-162259|2010-07-16|

---

PCT/JP2010/066882|WO2011043227A1|2009-10-07|2010-09-29|Frequency band enlarging apparatus and method, encoding apparatus and method, decoding apparatus and method, and program|

---