JP2003044096A - Method and device for encoding multi-channel audio signal, recording medium and music distribution system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and a device for encoding multi-channel audio signal, a recording medium and a music distribution system, by which the deterioration of a localization feeling is minimized and the compression efficiency is improved by controlling an IS frequency limit and execution/non-execution for the unit of sfb corresponding to the characteristics and compressibility of an input signal. SOLUTION: A step S4 has a process for processing a value calculated from a step S1 and a form controlled from a step S3, and a step S5 has a process for judging the value calculated from the step S1 and a frequency calculated from the step S4.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a multi-channel audio signal coding method, a multi-channel audio signal coding device, a recording medium and a music distribution system, and mainly to the field of multi-channel audio signal coding. Is.

[0002]

2. Description of the Related Art Conventionally, as a method in the field of such multi-channel audio signal coding, an audio signal coding method (MPEG2 AAC) as shown in FIG. 11 is used.
There is. First, in step S100, an input audio signal (also referred to as “stereo signal” or “multi-channel audio signal”, hereinafter referred to as “input signal”) is analyzed according to human hearing characteristics, and a signal-to-mask ratio (“signal-to-masking”) is analyzed. It is also a psychoacoustic analysis for calculating the value of "ratio: SMR". Step S101 is time / frequency mapping for converting an input signal in the time domain into spectrum data in the frequency domain using a transformation such as MDCT (modified discrete cosine) transformation. Step S102 is TNS processing for controlling the temporal shape of quantization noise. Step S103 is one of the modules for efficiently processing the input signal, and instead of the two channels, for example, the left channel and the right channel (hereinafter referred to as “left and right channels”), the connection energy and directivity of both left and right channels. This is intensity stereo processing for converting into information. In step S104, a signal (hereinafter referred to as "sum signal") that is the sum of input signals of the left and right channels (hereinafter referred to as "left and right channel signals") and a signal that is the difference between the left and right channel signals (hereinafter referred to as "difference signal"). ) Is the sum-difference stereo processing. In step S105, the input signal subjected to the above processing is quantized and coded (hereinafter referred to as "quantization /
It is called "encoding". ) Do. Step S106 is a bitstream generation formation in which the output of step S105 and the control information are multiplexed and shaped into a bitstream.

Intensity stereo processing utilizes the human auditory sense characteristic that the directionality of a high frequency sound is recognized by the intensity difference or the amplitude difference between the left and right channels, and the energy common to the left and right channels with respect to the input signal is used. Replace the envelope with roughly quantified directional information.

[0004]

By the way, the conventional intensity stereo processing cannot be used depending on the conditions of the input signal, that is, the number of channels, the window length, the number of groupings, etc., as shown in the flowchart of FIG. For example, if the input signal is a monaural signal, or if the left and right channels have different window lengths, such as a left channel having a short frame and a right channel having a long frame, the intensity stereo processing step cannot be performed.

Therefore, first, the coding mode is confirmed in step S110 by referring to the coding mode information storing the condition of the input signal. When it becomes possible to perform the intensity stereo process, in step S11 at a frequency at which preset step S112 is started, that is, in a scale factor band equal to or higher than the IS frequency limit.
2 is executed. In step S112, it is not necessary to transmit the spectrum component for each channel, and it is sufficient to transmit one directional information to the energy envelope and scale factor band common to the left and right channels.
In the case of a 2-channel input signal, about 50% of the information amount can be saved in the frequency band in which the intensity stereo is performed. Therefore, when the compression rate is high, the compression efficiency can be increased by increasing the frequency band in which step 112 is performed.

However, in the audio signal encoding method, if the intensity stereo processing is used without controlling,
When intensity stereo processing is performed on signals with very different energy envelopes in different channels, the left and right channel signals cannot be represented by the energy envelope common to the left and right channels and their directional information, and the sense of localization deteriorates. Also, there was a problem that the spatial impression would be collapsed at the center position.

The present invention has been made in order to solve such a problem, and executes or does not execute intensity stereo processing in units of IS frequency limit and scale factor band depending on the characteristics and compression ratio of the input signal (hereinafter simply referred to as " Control) of "execution / non-execution.", Which can minimize deterioration of localization and can obtain high compression efficiency, a multi-channel audio signal encoding device, and a recording medium. And a music distribution system.

[0008]

A multi-channel audio signal encoding method according to the present invention comprises a psychoacoustic analysis step of calculating a value of a signal-to-mask ratio based on an input signal, and spectral data of the other input signal. To a time / frequency mapping step, a TNS processing step of controlling a shape based on the converted spectrum data, an intensity stereo frequency limit calculating step of calculating a frequency based on the controlled shape, and a scale factor. It is one of a module for processing the input signal, and a band-by-band intensity stereo determination step of determining whether to perform intensity stereo processing according to the degree of approximation of the input signals of the left channel and the right channel for each band. An intensity stereo processing step for converting the two channels into connection energy and pointing information for both channels; A sum difference stereo processing step of converting the input signals of the channels into respective sum and difference signals, and a quantization / encoding step of quantizing and encoding the input signal applied in the sum difference stereo processing step. , A bitstream generation forming step of multiplexing the output of this quantization / encoding step and control information and shaping it as a bitstream, wherein the intensity stereo frequency limit calculation step is calculated from the psychoacoustic analysis step. A step of processing the value and the shape controlled from the TNS processing step, and the strength stereo determination process for each band includes the value calculated from the psychoacoustic analysis step and the strength stereo frequency limit calculation step. The method has a step of judging the calculated frequency. With this method, it is possible to minimize deterioration of the sense of localization, which is a drawback of intensity stereo processing, and to obtain high compression efficiency.

Here, in the multi-channel audio signal encoding method of the present invention, the encoding mode information step for setting the encoding mode information necessary for encoding and the intensity stereo can be used from the encoding mode information. A strength stereo pre-processing step of determining whether or not, a compression rate calculation step of calculating a compression rate from the information obtained from the encoding mode information, and a result of comparing the compression rate with a predetermined value. In the intensity stereo frequency limit calculation step, which comprises an output load determination step and an IS frequency limit setting step for setting a frequency for starting intensity stereo processing from the result of the load determination step, wherein the load determination step is the When it is determined that the compression ratio is high, the IS frequency limit setting step has means for performing intensity stereo processing in a wider frequency band, and When the constant-step is judged to be low the compression ratio has a method of the IS frequency limit setting step had the means for performing the intensity stereo processing in a narrower frequency band. With this method, if the compression ratio is low,
By performing the intensity stereo processing in a narrower frequency band, it is possible to minimize the deterioration of localization, which is a drawback of the intensity stereo processing.

In the multi-channel audio signal coding method of the present invention, a coding mode information step for setting coding mode information necessary for coding, and whether strength stereo can be used based on the coding mode information. A strength stereo pre-processing step of determining whether or not, a compression rate calculation step of calculating a compression rate from information obtained from the encoding mode information, and a determination threshold value based on the compression rate calculated in this compression rate calculation step. A determination threshold value calculating step, an approximation degree calculating step for calculating the degree of approximation of the left channel and the right channel based on the determination threshold value calculated in the determination threshold value calculation step, and the result of comparing the determination threshold value and the compression rate. And the IS frequency limit setting step of setting the frequency based on the result of the approximation degree determination processing step. In the calculation step, when the approximation degree determination processing step determines that the degree of approximation is higher than the determination threshold value, the IS frequency limit setting step includes a step of performing intensity stereo processing in a wider frequency band, and When the approximation degree determination processing step determines that the approximation degree is low, the IS frequency limit setting step is a method having means for performing intensity stereo processing in a narrower frequency band. By this method, the degradation of localization due to intensity stereo processing does not occur by adaptively setting the frequency band for performing intensity stereo processing according to the characteristics of the input signal,
It is possible to perform encoding with high compression efficiency.

Further, in the multi-channel audio signal encoding method of the present invention, the scale factor band determination step of comparing the scale factor band with the IS frequency limit, the scale factor band compared with the scale factor band determination step and the scale factor band An approximation degree calculating step of calculating the approximation degree of the left channel and the right channel based on the IS frequency limit, and an approximation degree determining processing step of comparing the magnitude of the approximation degree calculated in the approximation degree calculating step with a preset threshold value. The intensity for each band, which is one of the modules for processing the input signal from the approximation determination processing step, and which has an intensity stereo processing step for converting the two channels into connection energy of both channels and directivity information. In the stereo determination step, the approximation degree determination processing step increases the approximation degree. When it is determined that the intensity stereo processing step has a step of performing intensity stereo processing, and when the approximation degree determination processing step determines that the degree of approximation is low, the scale factor band is obtained without performing the intensity stereo processing. The method has a step of returning to the determination step. By this method, the localization deterioration due to the intensity stereo processing does not occur, and the intensity stereo processing can be executed more in a high frequency band in which the influence of the sound quality is small, so that the compression efficiency can be improved.

Also, the multi-channel audio signal encoding method of the present invention comprises:
A scale factor band determination step of comparing S frequency limits, a determination threshold value calculation step of calculating a determination threshold value based on the scale factor band and the IS frequency limit compared in the scale factor band determination step, and this determination threshold value calculation step In the approximation degree calculating step of calculating the degree of approximation of the left and right channels based on the determination threshold value calculated in, the determination threshold value calculated in the determination threshold value calculating step and the approximation degree calculated in the approximation degree calculating step are compared. An approximation degree determination processing step, and one of modules for processing the input signal from the approximation degree determination processing step, and an intensity stereo processing step of converting the two channels into connection energy of both channels and directional information. In the intensity stereo determination process for each band, wherein the approximation degree determination processing step has an approximation degree higher than the determination threshold. When the intended determination, comprising the step of the intensity stereo processing step performs intensity stereo processing,
In addition, when the approximation degree determination processing step determines that the approximation degree is low, the intensity stereo processing step includes a step of returning to the scale factor band determination step without performing intensity stereo processing. By this method, the localization deterioration due to the intensity stereo processing does not occur, and the intensity stereo processing can be executed more in a high frequency band in which the influence of the sound quality is small, so that the compression efficiency can be improved.

The multi-channel audio signal encoding device of the present invention comprises a psychoacoustic analysis means for calculating a value of a signal-to-mask ratio based on an input signal, and a time / frequency for converting the other input signal into spectrum data. Mapping means, TNS processing means for controlling a shape based on the converted spectrum data, intensity stereo frequency limit calculating means for calculating a frequency based on the controlled shape, and a left channel for each scale factor band. A band-wise intensity stereo determination means for determining whether or not to perform intensity stereo processing according to the degree of approximation of the input signal of the right channel, and one of modules for processing the input signal,
Intensity stereo processing means for converting the two channels into connection energy and directional information of both channels; sum difference stereo processing means for converting the input signals of the channels into respective sum and difference signals; Quantization / encoding means for quantizing and encoding the input signal applied by the stereo processing means, and bitstream generation forming for multiplexing the output of this quantization / encoding means and control information and shaping it as a bitstream. Means, the intensity stereo frequency limit calculation means includes means for processing the value calculated by the psychoacoustic analysis means and the shape controlled by the TNS processing means, and the intensity stereo for each band. The determination means, the value calculated from the psychoacoustic analysis means and the intensity stereo frequency limit calculation means It has a configuration having a means for determining the wave number. With this configuration, it is possible to minimize deterioration of the sense of localization, which is a drawback of the intensity stereo processing, and obtain high compression efficiency.

Here, the multi-channel audio signal coding apparatus of the present invention can use coding mode information means for setting coding mode information necessary for coding, and strength stereo can be used from the coding mode information. Strength stereo pre-processing means for determining whether or not, compression rate calculating means for calculating a compression rate from information obtained from the encoding mode information, and a result of comparing the compression rate with a predetermined value. The intensity stereo frequency limit calculating means comprises: load determining means for outputting; and IS frequency limit setting means for setting a frequency at which intensity stereo processing is started based on the result of the load determining means. When it is determined that the compression ratio is high, the IS frequency limit setting means has means for performing intensity stereo processing in a wider frequency band, and When the constant means determines that lower the compression ratio, it has a configuration in which the IS frequency limit setting means is comprising means for performing intensity stereo processing in a narrower frequency band. With this configuration, when the compression rate is low,
By performing the intensity stereo processing in a narrower frequency band, it is possible to minimize the deterioration of localization, which is a drawback of the intensity stereo processing.

Further, the multi-channel audio signal coding apparatus of the present invention is capable of using coding mode information means for setting coding mode information necessary for coding and strength stereo based on the coding mode information. Intensity stereo pre-processing means for determining whether or not, compression rate calculating means for calculating a compression rate from information obtained from the encoding mode information, and a determination threshold value based on the compression rate calculated by this compression rate calculating means A determination threshold value calculating means for calculating, the approximation degree calculating means for calculating the degree of approximation of the left channel and the right channel based on the determination threshold value calculated by the determination threshold value calculating means, and the result of comparing the determination threshold value and the compression rate. The intensity stereo frequency limit provided with an approximation degree determination processing means for outputting and an IS frequency limit setting means for setting a frequency from the result of the approximation degree determination processing means. Computation means, when the approximation degree determination processing means determines that the approximation degree is higher than the determination threshold value, the IS frequency limit setting means has means for performing intensity stereo processing in a wider frequency band, and The IS frequency limit setting means has means for performing intensity stereo processing in a narrower frequency band when the approximation degree determination processing means determines that the approximation degree is low. With this configuration, the degradation of localization due to the intensity stereo processing does not occur by adaptively setting the frequency band for performing the intensity stereo processing according to the characteristics of the input signal,
It is possible to perform encoding with high compression efficiency.

Further, the multi-channel audio signal encoding device of the present invention comprises an approximation degree calculating means for calculating the approximation degree of the left channel and the right channel based on the scale factor band and the IS frequency limit, and the approximation degree calculating means. It is one of an approximation degree determination processing unit that compares the magnitude of the calculated approximation degree with a preset threshold value, and one of the modules that processes the input signal from the approximation degree determination processing unit. The intensity stereo processing means for each band, which comprises a connection energy of a channel and an intensity stereo processing means for converting to directional information, wherein the intensity stereo processing is performed when the approximation degree determination processing means determines that the approximation degree is high. The means has a means for performing intensity stereo processing, and the approximation degree determination processing means determines that the approximation degree is low. Kiniwa, has a configuration having a means for returning to the scale factor band judging unit without performing the intensity stereo processing. With this configuration, it is possible to improve the compression efficiency without causing the localization deterioration due to the intensity stereo processing.

Also, the multi-channel audio signal encoding device of the present invention is characterized in that the scale factor band and the I
Judgment threshold value calculation means for calculating the judgment threshold value based on the S frequency limit, approximation degree calculation means for calculating the approximation degree of the left and right channels based on the judgment threshold value calculated by the judgment threshold value calculation means, and the decision threshold value calculation means Which is one of a module for processing the input signal from the approximation degree determination processing means and the approximation degree determination processing means for comparing the determination threshold value calculated in the above and the approximation degree calculated by the approximation degree calculation means, The intensity stereo determination means for each band, comprising intensity stereo processing means for converting two channels into connection energy of both channels and directivity information, wherein the approximation degree determination processing means has a degree of approximation higher than the determination threshold. When it is determined that the intensity stereo processing means has a means for performing intensity stereo processing, and the approximation degree determination processing means determines that the approximation degree is low. When the has a structure in which the intensity stereo processing means having means for returning to the scale factor band judging unit without performing the intensity stereo processing. With this configuration, the localization deterioration due to the intensity stereo process does not occur, and the intensity stereo process can be further executed in a high frequency band in which the influence of the sound quality is small, so that the compression efficiency can be improved.

The multi-channel audio signal coding apparatus of the present invention has a configuration in which the input signal is quantized and coded by using any one of the above-described multi-channel audio signal coding methods. With this configuration, it is possible to obtain the same effect as that of the above-mentioned multi-channel audio signal encoding method.

The recording medium of the present invention has a configuration in which the steps of the multi-channel audio signal encoding method described in any of the above are stored. With this configuration, it is possible to obtain the same effect as that of the above-mentioned multi-channel audio signal encoding method.

Here, the music distribution system of the present invention is characterized by using this recording medium as a constituent element. With this feature, it is possible to obtain the same effect as that of the above-mentioned multi-channel audio signal encoding method.

Further, the music distribution system of the present invention has a structure in which music is distributed by using any one of the above-described multi-channel audio signal encoding methods. With this configuration, it is possible to obtain the same effect as that of the above-mentioned multi-channel audio signal encoding method.

Further, the music distribution system of the present invention is characterized in that the multi-channel audio signal encoding device according to any one of the above is used as a constituent element. With this feature, it is possible to obtain the same effect as that of the multi-channel audio signal encoding device.

BEST MODE FOR CARRYING OUT THE INVENTION A multi-channel audio signal encoding method according to an embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a flowchart showing a multi-channel audio signal encoding method according to the first embodiment of the present invention.

As shown in the figure, in the multi-channel audio signal coding method, a psychoacoustic analysis step of calculating a value of a signal-to-mask ratio based on a psychoacoustic model utilizing human auditory characteristics based on an input signal. As step S1
And step S2 as a time-frequency mapping step (hereinafter referred to as "time / frequency mapping step") of converting the other input signal in the time domain into spectrum data in the frequency domain, and based on the converted spectrum data. Step S3 as a TNS processing step for controlling the temporal shape of the quantization noise, step S4 as an intensity stereo frequency limit calculating step for calculating a frequency at which intensity stereo is started based on the controlled shape, and intensity stereo. Step S5 as a band-wise intensity stereo determination step of determining whether or not to perform intensity stereo processing according to the degree of approximation of the left and right channel signals for each scale factor band (hereinafter referred to as "sfb") to be processed, It is one of the modules that efficiently processes the input signal from step S5. Step S6 as an intensity stereo processing step for converting the connection energy and directional information of both channels, and step S7 as a sum difference stereo processing step for converting the left and right channel signals into a sum signal and a difference signal, and step S7. Step S8 as a quantizing / encoding step of quantizing and encoding the input signal subjected to in step S8, and step of a bitstream generation forming step of multiplexing the output of step 8 and control information and shaping them as a bitstream. S10 and.

Next, the processing flow of this multi-channel audio signal encoding method will be described.

First, for the input signal, the value of the signal-to-mask ratio is calculated based on the psychoacoustic model in step S1. The other input signal is MD in step S2.
The spectrum data in the time domain is converted into the spectrum data in the frequency domain using a transform such as CT transform.

Next, step S3 controls the temporal shape of the quantization noise based on the converted spectrum data. Then, step S4 calculates an IS frequency limit, which is the frequency at which intensity stereo is started based on the controlled shape. Subsequently, in step S5, it is determined whether or not the intensity stereo process is performed for each sfb at a frequency equal to or higher than the IS frequency limit, according to the degree of approximation of the left and right channel signals. On the other hand, if it is determined in step S5 that the intensity stereo process cannot be performed based on the calculated IS frequency limit (NO), the process proceeds between step S6 and step S7. In step S6, it is determined that the intensity stereo processing can be performed in step S5 (Y
ES), and intensity stereo processing is performed.

Next, a step S7 converts the left and right channel signals into a sum signal and a difference signal, respectively. Then
A step S8 quantizes and encodes the input signal subjected to the above processing. In step S10, the output of step S8 and the control information are multiplexed and shaped as a bit stream.

According to such a multi-channel audio signal encoding method, the frequency at which intensity stereo is started is adaptively changed according to the characteristics and compression rate of the input signal, and sf is changed according to the characteristics and compression rate of the input signal.
Since execution / non-execution is controlled in units of b, it is possible to minimize deterioration of localization, which is a drawback of intensity stereo processing, and obtain high compression efficiency.

FIG. 2 is a flowchart showing step S4 in the second embodiment of the present invention.

As shown in the figure, step S4 includes step S21 as an encoding mode information step for setting encoding mode information necessary for encoding such as bit rate and sampling frequency, and the encoding mode. The compression ratio is calculated from step S22 as a strength stereo pre-processing step of determining whether or not strength stereo can be used from the information, and information such as the bit rate, sampling frequency, and the number of channels obtained from the coding mode information. A step S23 as a compression rate calculation step, a step S24 as a load determination step for outputting the result of comparing the compression rate with a predetermined value, and a frequency for starting the intensity stereo processing from the result of step S24 are set. Step S25 as an IS frequency limit setting step.

Next, the flow of processing in step S4 will be described.

In step S4, intensity stereo processing cannot be used when the left and right channel coding modes (number of channels, window length, grouping, etc.) are different. Therefore, the intensity stereo processing is first performed using the coding mode information S21. It is determined whether it can be used. When the input signals are in different encoding modes such as monaural (when NO),
The process ends (END). If it is determined that the intensity stereo processing is possible (YES), step S23.
Calculates the compression rate from the bit rate, the sampling frequency, and the number of channels obtained in step S21, for example, by the calculation formula shown in (1) below.

[Equation 1]

Where fs is the sampling frequency and br
Is the bit rate and ch is the number of channels.

Next, a step S24 compares the compression rate with a predetermined value set in advance. If the compression rate is higher than the predetermined value, that is, if the compression rate is high, the frequency at which the intensity stereo processing is started in a wider frequency band is lowered in step S25. Conversely, if the compression rate is smaller than the predetermined value, that is, if the compression rate is low,
In step S25, the frequency at which the intensity stereo process is started in a narrower frequency band is increased.

According to step S4 as described above, when the compression rate is high, the intensity stereo processing is performed in a wider frequency band, so that the amount of information can be saved and the compression efficiency can be improved. Further, according to step S4, when the compression rate is low, by performing the intensity stereo process in a narrower frequency band, it is possible to minimize the degradation of localization, which is a drawback of the intensity stereo process.

FIG. 3 is a flowchart showing step S4 in the third embodiment of the present invention.

As shown in the figure, step S4 is a judgment threshold value calculating step for calculating a judgment threshold value based on the compression rates calculated in step S21, step S22, step S23, and step S23. Step S
31, step S32 as an approximation degree calculating step of calculating the approximation degree of the left and right channels based on the determination threshold value calculated in step S31, and an approximation degree determination processing step of outputting a result of comparing the determination threshold value and the compression rate. Step S33
And step S34 as an IS frequency limit setting step of setting the frequency at which the intensity stereo processing is started based on the result of step S33.

Next, the flow of processing in step S4 will be described.

In step S4, the intensity stereo process cannot be used when the coding modes of the left and right channels are different. Therefore, first in step S21, it is determined in step S22 whether or not intensity stereo can be used. When the input signal is in a different encoding mode such as monaural (NO), the process is ended (END).
If it is determined that the intensity stereo processing is possible (YES), a step S23 determines a compression rate from the bit rate, the sampling frequency, and the number of channels obtained in the step S21, for example, by the calculation formula shown in (1) below. To calculate.

[Equation 1]

Next, in step S31, a judgment threshold value is calculated by comparing the compression rate with a predetermined value that is set in advance. When the compression rate is high, the determination threshold value is set to a value that allows a wide frequency band in which the intensity stereo processing is performed. Further, as the determination threshold, it is possible to set several levels of thresholds depending on the size of the compression rate.

Next, in step S32, the degree of approximation is calculated using the cross-correlation value of the left and right channel signals and the standard deviation of the ratio of the left and right channel signals. Then, step S33.
Is the judgment threshold value calculated in step S31 and step S
The approximations calculated in 32 are compared, and if the approximations are larger than the determination threshold, that is, if the compression rate is high and the left and right channel signals are high, the step S34
In order to perform intensity stereo at a wider frequency, the frequency at which intensity stereo processing is started is low, for example 3k.
Set to Hz. On the other hand, when the degree of approximation is smaller than the determination threshold, that is, when the compression rate is low and the degree of approximation of the left and right channel signals is low, the frequency at which the intensity stereo processing is started is set high in step S34, for example, 6 kHz. When the compression rate is very low and the degree of approximation of the left and right channel signals is very low, the frequency may be set so that the intensity stereo process is not performed.

According to step S4 as described above, by adaptively setting the frequency band in which the intensity stereo processing is performed according to the characteristics of the input signal, localization deterioration due to the intensity stereo processing does not occur, and encoding with high compression efficiency is performed. It can be carried out.

FIG. 4 is a flowchart showing step S5 in the fourth embodiment of the present invention.

As shown in the figure, step S5 is sf
b as a scale factor band determination step of comparing the IS frequency limit with step S41, and step S41 as an approximation degree calculating step of calculating the degree of approximation of the left and right channels based on the sfb and the IS frequency limit compared in step S41.
42, step S43 as an approximation degree determination processing step of comparing the degree of approximation calculated in step S42 with a preset threshold value, and one of the modules for efficiently processing the input signal from step S43. Instead of the two channels, it has a connection energy of both channels and a step S46 as an intensity stereo processing step for converting into directional information.

Next, the flow of processing in step S5 will be described.

First, sfb is set to 0 (zero).
Sfb for performing intensity stereo processing in step S41
And IS frequency limit are compared, and intensity stereo processing is performed.
If fb is greater than the IS frequency limit (if YES)
Proceeds to step S42. Intensity stereo processing
When fb is smaller than IS frequency limit (when NO)
Raises sfb by 1 and returns to step S41. A step S42 calculates the degree of approximation using the cross-correlation value of the left and right channel signals and the standard deviation of the ratio of the left and right channel signals.

Next, in step S43, step S4 is performed.
The degree of approximation calculated in 2 is compared with a preset threshold value, and when the degree of approximation is larger than the preset threshold value, that is, when the left and right channel signals are similar signals (in the case of YES) Advances to step S46 to perform intensity stereo processing. On the other hand, when the degree of approximation is smaller than the preset threshold value, that is, when the left and right channel signals are very different (NO), the intensity stereo process is not performed in sfb. Next, sfb is increased by 1, and the process returns to step S41. The above steps are repeated up to the maximum value of sfb.

According to such step S5, the compression ratio is high, and the intensity stereo process is started in order to increase the compression efficiency by using the intensity stereo process in many sfb.
When b is lowered, it is possible to prevent the intensity stereo processing from being performed in sfb which causes localization deterioration in the sfb in which the intensity stereo processing is performed. Therefore, according to step S5, the degradation of the localization due to the intensity stereo process does not occur, and the intensity stereo process can be further executed in the high frequency band in which the influence of the sound quality is small, so that the compression efficiency can be improved. it can.

FIG. 5 is a flowchart showing step S5 in the fifth embodiment of the present invention.

As shown in the figure, step S5 includes step S51, which is a determination threshold value calculating step for calculating a determination threshold value based on the sfb and IS frequency limit compared with step S41. Step S52 as an approximation degree calculating step of calculating the approximation degree of the left and right channels based on the calculated determination threshold;
Step S53 as an approximation degree determination processing step of comparing the determination threshold value calculated in step 51 with the approximation degree calculated in step S52, and one of the modules that efficiently processes the input signal from step S5. Instead of step S56, there is a connection energy of both channels and step S56 as an intensity stereo processing step of converting into directional information.

Next, the flow of processing in step S5 will be described.

First, sfb is set to 0. Step S
In 41, sfb performing intensity stereo processing is compared with IS frequency limit, and sfb performing intensity stereo processing is IS.
When it is larger than the frequency limit (in the case of YES), the process proceeds to step S51. IS is the sfb that performs intensity stereo processing
When it is smaller than the frequency limit (in the case of NO), sfb is increased by 1, and the process returns to step S41. A step S51 sets a determination threshold value so as to facilitate the intensity stereo processing for the high frequency band having a small influence of the sound quality, and makes a determination so as to limit the intensity stereo processing for the low frequency band having a large influence of the sound quality. Set the threshold. It should be noted that this determination threshold can be adjusted depending on the size of the compression rate.

Next, in step S52, the degree of approximation is calculated using the cross-correlation value of the left and right channel signals and the standard deviation of the ratio of the left and right channel signals. Then, step S53
Is the judgment threshold value calculated in step S51 and step S
If the degree of approximation in each sfb is greater than the determination threshold, that is, if the left and right channel signals are similar signals (in the case of YES), the process proceeds to step S56. Intensity stereo processing is performed. In each sfb, when the degree of approximation is smaller than the determination threshold, that is, when the left and right channel signals are very different (N
In the case of O), intensity stereo processing is not performed in the sfb. Then, sfb is increased by 1, and the process returns to step S41. The above steps are repeated up to the maximum value of sfb.

According to such step S5, the compression ratio is high, and the intensity stereo process is started in order to increase the compression efficiency by using the intensity stereo process in many sfb.
When b is lowered, it is possible not to perform the intensity stereo process even if the sfb that causes the localization deterioration exists in the sfb for which the intensity stereo process is performed. Therefore, according to step S5, the degradation of the localization due to the intensity stereo process does not occur, and the intensity stereo process can be further executed in the high frequency band in which the influence of the sound quality is small, so that the compression efficiency can be improved. it can.

The multi-channel audio signal encoding method described in each of the above embodiments can be used as a software program that can be executed by a computer, for example, on any recording medium such as a magnetic disk, an optical disk, a semiconductor memory, or the like. It can be implemented as a method of recording or storing. As the magnetic disk, a floppy (registered trademark) disk,
There is a hard disk, etc., and as an optical disc, CD-
There are ROM, DVD, etc.

Further, although the multi-channel audio signal encoding device (not shown) quantizes and encodes the input signal by using the multi-channel audio signal encoding method described in each embodiment, each of the above-mentioned respective embodiments is performed. It is possible to obtain an effect similar to that of the multi-channel audio signal encoding method in the form.

Further, although not shown, the recording medium has recorded therein the steps of the multi-channel audio signal encoding method described in each of the embodiments, and therefore, the same as the multi-channel audio signal encoding method in each of the above-mentioned embodiments. The effect of can be obtained.

Further, although not shown, the music distribution system has this recording medium as a constituent element, so that it is possible to obtain the same effect as the multi-channel audio signal encoding method in each of the above-mentioned embodiments. Further, in the music distribution system, since music is distributed by using the multi-channel audio signal encoding method, it is possible to obtain the same effect as the multi-channel audio signal encoding method in each of the above embodiments.

In each of the above embodiments, the flow of processing of the multi-channel audio signal encoding method has been described with reference to the flow charts of FIGS. 1 to 5.
A multi-channel audio signal coding apparatus using the multi-channel audio signal coding method will be described with reference to the drawings.

FIG. 6 is a block diagram showing a multi-channel audio signal coding apparatus using the multi-channel audio signal coding method in the above-mentioned first embodiment.

As shown in the figure, the multi-channel audio signal encoding apparatus has a psychoacoustic analysis means 1 for calculating a value of a signal-to-mask ratio based on a psychoacoustic model from an input signal, and another input for the time domain. Time / frequency mapping means 2 for converting a signal into frequency domain spectrum data, TNS processing means 3 for controlling the temporal shape of quantization noise, and intensity stereo frequency limit calculation means for calculating a frequency at which intensity stereo is started. 4, an intensity stereo determination unit 5 for each band that determines whether intensity stereo processing is performed for each sfb according to the degree of approximation of the left and right channel signals, and one of the modules that efficiently processes the input signal.
Intensity stereo processing means 6 for converting connection energy and directional information of both channels instead of one channel
A sum difference stereo processing means 7 for converting the left and right channel signals into a sum signal and a difference signal, and a sum difference stereo processing means 7
A quantizing / encoding means 8 for quantizing and encoding the input signal subjected to 1., and a bitstream generating / forming means 10 for multiplexing the output of the quantizing / encoding means 8 and control information and shaping it as a bitstream. have.

Next, the operation of this multi-channel audio signal encoding device will be described.

First, for the input signal, the psychoacoustic analysis means 1 calculates the value of the signal-to-mask ratio based on the psychoacoustic model. The other input signal is converted by the time / frequency mapping means 2 from the time domain into the spectrum data in the frequency domain using a transformation such as MDCT transformation.

Next, the TNS processing means 3 controls the temporal shape of the quantization noise based on the converted spectrum data. Then, the intensity stereo frequency limit calculation means 4
Calculate the IS frequency limit, which is the frequency at which intensity stereo starts based on the controlled shape. Subsequently, the band-based intensity stereo determination means 5 determines whether or not to perform intensity stereo processing for each sfb at a frequency equal to or higher than the IS frequency limit, according to the degree of approximation of the left and right channel signals of the input signal. The intensity stereo processing means 6 performs the intensity stereo processing when it is determined by the intensity stereo determination means 5 for each band that the intensity stereo processing can be performed. Next, the sum / difference stereo processing means 7 converts the left and right channel signals into a sum signal and a difference signal, respectively. Then, the quantization / encoding processing means 8 quantizes and encodes the input signal processed by the sum difference stereo processing means 7. The bitstream generation / formation unit 10 multiplexes the output of the quantization / encoding processing unit 8 and the control information and shapes them as a bitstream.

According to such a multi-channel audio signal coding apparatus, the frequency at which intensity stereo is started is adaptively changed according to the characteristics and compression rate of the input signal, and sf is changed according to the characteristics and compression rate of the input signal.
Since execution / non-execution is controlled in units of b, it is possible to minimize deterioration of localization, which is a drawback of intensity stereo processing, and obtain high compression efficiency.

FIG. 7 is a block diagram showing the intensity stereo frequency limit calculating means 4 using step S4 in the above-described second embodiment.

As shown in the figure, the intensity stereo frequency limit calculating means 4 has a coding mode information means 21 for setting coding mode information necessary for coding such as a bit rate and a sampling frequency, and the code. Intensity stereo preprocessing means 22 for determining whether or not intensity stereo can be used from the encoding mode information, and compression for calculating a compression rate from information such as the bit rate, sampling frequency, and number of channels obtained from the encoding mode information. Rate calculating means 23,
The load determining means 24 outputs a result of comparing the compression rate with a predetermined value, and the IS frequency limit setting means 25 sets a frequency for starting the intensity stereo processing based on the result of the load determining means 24. There is.

Next, the operation of the intensity stereo frequency limit calculating means 4 will be described.

In the intensity stereo frequency limit calculation means 4, intensity stereo processing cannot be used when the left and right channel encoding modes are different. Therefore, first, the encoding mode information means 21 is used to determine whether or not intensity stereo can be used. Whether or not it is determined. When the input signal is in a different encoding mode such as monaural, the subsequent operation is not performed. When it is determined that the intensity stereo processing can be performed, the compression rate calculation means 2
3 calculates the compression rate from the bit rate, the sampling frequency, and the number of channels obtained from the encoding mode information means 21, for example, by the calculation shown in Formula 1 below.

[Equation 1]

Next, the load judging means 24 compares the compression rate with a predetermined value set in advance. When the compression rate is higher than the predetermined value, that is, when the compression rate is high, the IS frequency limit setting unit 25 lowers the frequency at which the intensity stereo process is started. On the contrary, when the compression rate is smaller than the predetermined value, that is, when the compression rate is low, the IS frequency limit setting means 25 increases the frequency at which the intensity stereo processing is started.

According to the intensity stereo frequency limit calculating means 4 as described above, when the compression rate is high, the intensity stereo processing is performed in a wider frequency band, so that the amount of information can be saved and the compression efficiency can be improved. You can Further, according to the intensity stereo frequency limit calculation means 4, when the compression rate is low, the intensity stereo processing is performed in a narrower frequency band, thereby minimizing the deterioration of localization, which is a drawback of the intensity stereo processing. it can.

FIG. 8 is a block diagram showing the intensity stereo frequency limit calculating means 4 using step S4 in the above-mentioned third embodiment.

As shown in the figure, the intensity stereo frequency limit calculation means 4 includes the encoding mode information means 21, the intensity stereo preprocessing means 22, the compression rate calculation means 23, and the compression rate calculation means 23. A judgment value calculation means 31 for calculating a judgment threshold value based on the compression ratio calculated in step 1, an approximation degree calculation means 32 for calculating an approximation degree for the left and right channels based on the judgment threshold value calculated by the judgment value calculation means 31, and a judgment threshold value And an approximation degree determination processing unit 33 that outputs a result of comparing the compression rate with
The IS frequency limit setting means 25 for setting the frequency at which the intensity stereo processing is started based on the result of the similarity determination processing means 33.
And have.

Next, the operation of the intensity stereo frequency limit calculating means 4 will be described.

In the intensity stereo frequency limit calculation means 4, intensity stereo processing cannot be used when the left and right channel encoding modes are different. Therefore, first, the encoding mode information means 21 is used to determine whether or not intensity stereo can be used. The determination is made by the intensity stereo preprocessing means 22. When the input signal is in a different encoding mode such as monaural, the subsequent operation is not performed. When it is determined that the intensity stereo processing is possible, the compression rate calculation means 23 uses, for example, the following (1) from the bit rate, the sampling frequency, and the number of channels obtained from the coding mode information means 21.
The compression ratio is calculated by the calculation shown in.

[Equation 1]

Next, the judgment threshold value calculation means 31 calculates the judgment threshold value by comparing the compression rate with a predetermined value set in advance to determine the magnitude. When the compression rate is high, the determination threshold value is set to a value that allows a wide frequency band in which the intensity stereo processing is performed. Further, as the determination threshold, it is possible to set several levels of thresholds depending on the size of the compression rate.

Next, the approximation degree calculating means 32 calculates the approximation degree using the cross-correlation value of the left and right channel signals and the standard deviation of the ratio of the left and right channel signals. Subsequently, the approximation degree determination processing unit 33 compares the determination threshold value calculated by the determination threshold value calculation unit 31 and the approximation degree calculated by the approximation degree calculation unit 32, and when the approximation degree is larger than the determination threshold value, That is, when the compression rate is high and the closeness of the left and right channel signals is high, the frequency at which the intensity stereo processing is started is set low, for example, 3 kHz in the IS frequency limit setting means 34 so that intensity stereo can be performed at a wider frequency. .
On the other hand, when the degree of approximation is smaller than the determination threshold, that is, when the compression rate is low and the degree of approximation of the left and right channel signals is low, the frequency at which the intensity stereo processing is started in the IS frequency limit setting means 34 is high, for example, 6 kHz. Set. When the compression rate is very low and the degree of approximation of the left and right channel signals is very low, the frequency may be set so that the intensity stereo process is not performed.

According to the intensity stereo frequency limit calculating means 4 as described above, the frequency band for performing the intensity stereo processing is adaptively set according to the characteristics of the input signal, so that the localization deterioration due to the intensity stereo processing does not occur, and the compression efficiency is reduced. It is possible to perform high encoding.

FIG. 9 is a band-wise intensity stereo determination means 5 using step S5 in the above-described fourth embodiment.
It is a block diagram showing.

As shown in the figure, the intensity stereo determination means 5 for each band includes an approximation degree calculation means 32 and an approximation degree calculation means 3.
It is one of the modules for efficiently processing the input signal from the approximation degree determination processing means 43 and the approximation degree determination processing means 43 for comparing the magnitude of the approximation degree calculated in 2 with a preset threshold value. Instead of channels, it has connection energy of both channels and intensity stereo processing means 46 for converting it into directional information.

Next, the intensity stereo determination means 5 for each band
The operation of will be described.

First, in sfb for performing intensity stereo processing in the approximation degree calculating means 32, the approximation degree is calculated using the cross-correlation value of the left and right channel signals and the standard deviation of the ratio of the left and right channel signals. Next, the approximation degree determination processing means 43
The degree of approximation calculated by the degree-of-approximation calculating unit 32 is compared with the preset threshold value for each sfb, and when the degree of approximation is larger than the preset threshold value, that is, a signal in which the left and right channel signals are similar to each other. If so, the intensity stereo permission flag is set to the sfb.

On the other hand, when the degree of approximation is smaller than the preset threshold value, that is, when the left and right channel signals are very different, the intensity stereo permission flag is set to s.
I can't stand on fb. The intensity stereo processing means 46 performs the intensity stereo processing at sfb for which the intensity stereo permission flag is set in the approximation degree determination processing means 43.

Such band-wise intensity stereo determination means 5
According to this, when the compression ratio is high and the sfb that starts the intensity stereo process is lowered in order to increase the compression efficiency by using the intensity stereo process with many sfb, the localization in the sfb that performs the intensity stereo process is performed. It is possible to prevent the intensity stereo processing from being performed in the sfb that causes deterioration.
Therefore, according to the intensity stereo determination means 5 for each band, the localization deterioration due to the intensity stereo processing does not occur, and the compression efficiency can be improved.

FIG. 10 is a block diagram showing the band-by-band intensity stereo determination means 5 using step S5 in the above-described fifth embodiment.

As shown in the figure, the intensity stereo determination means for each band 5 is a determination threshold value calculation means 51 for calculating a determination threshold value based on sfb and the IS frequency limit, and a determination threshold value calculation means 5.
The approximation degree calculation means 52 for calculating the approximation degree of the left and right channel signals calculated in 1 and the approximation degree determination processing means 53 for comparing the determination threshold value calculated by the determination threshold value calculation means 51 with the approximation degree calculated by the approximation degree calculation means 52. And the approximation degree determination processing means 5
It is one of the modules for efficiently processing the input signal from 3 and has intensity stereo processing means 56 for converting connection energy of both channels and directional information instead of two channels.

Next, the intensity stereo determination means 5 for each band
The operation of will be described.

First, in the judgment threshold value calculation means 51, a judgment threshold value is set for a high frequency band having little influence on sound quality, and a judgment threshold value is set so as to facilitate intensity stereo processing, and an intensity is set for a low frequency band having a large influence on sound quality. A judgment threshold value is set so as to limit stereo processing. It should be noted that this determination threshold can be adjusted depending on the size of the compression rate.

Next, in sfb for performing intensity stereo processing by the approximation degree calculation means 52, the approximation degree is calculated using the cross correlation value of the left and right channel signals and the standard deviation of the ratio of the left and right channel signals. Then, the similarity determination processing means 53
The determination threshold value calculated by the determination threshold value calculation means 51 and the approximation degree calculated by the approximation degree calculation means 52 are compared for each sfb, and when the approximation degree is larger than the determination threshold value in each sfb, that is, the left and right channel signals are similar. If the signal is a signal that has been set, the intensity stereo permission flag is set to the sfb.
When the degree of approximation is smaller than the determination threshold in each sfb,
That is, when the left and right channel signals are very different, the intensity stereo permission flag is not set to the sfb. The intensity stereo processing means 56 performs the intensity stereo processing at sfb in which the intensity stereo permission flag is set by the approximation degree determination processing means 53.

Such band-wise intensity stereo determination means 5
According to this, when the compression ratio is high and the sfb that starts the intensity stereo process is lowered in order to increase the compression efficiency by using the intensity stereo process in many sfb, the localization in the sfb that performs the intensity stereo process is performed. It is possible to prevent the intensity stereo processing from being performed even if the sfb that causes deterioration exists. Therefore, the intensity stereo determination unit 5 for each band does not cause localization deterioration due to the intensity stereo process, and the intensity stereo process can be performed more in a high frequency band with little influence of sound quality.
The compression efficiency can be increased.

Although not shown, the audio distribution system has the multi-channel audio signal coding device and the recording medium described in each of the above-described embodiments as its constituent elements. The same effect can be obtained.

[0092]

As described above, according to the present invention, the psychoacoustic analysis step of calculating the value of the signal-to-mask ratio based on the input signal, and the time / frequency mapping for converting the other input signal into spectrum data. A TNS processing step of controlling a shape based on the converted spectral data, an intensity stereo frequency limit calculating step of calculating a frequency based on the controlled shape, and a left channel and a right channel for each scale factor band. The intensity stereo determination process for each band for determining whether or not to perform intensity stereo processing according to the degree of approximation of the input signal of a channel, and one of the modules that processes the input signal. Intensity stereo processing step for converting the connection energy and directivity information of each channel, and summing each of the input signals of the channels. And a difference stereo processing step of converting into a difference signal, a quantization / encoding step of quantizing and encoding the input signal applied in the sum difference stereo processing step, and a quantization / encoding step of this quantization / encoding step. A bitstream generation / forming step of multiplexing output and control information and shaping as a bitstream, wherein the intensity stereo frequency limit calculation step is controlled from the value calculated from the psychoacoustic analysis step and the TNS processing step. And a step of processing the shape, and the band-wise intensity stereo determination step,
By having the step of judging the value calculated from the psychoacoustic analysis step and the frequency calculated from the intensity stereo frequency limit calculation step, the IS frequency limit and scale factor depending on the characteristics and compression ratio of the input signal. A multi-channel audio signal encoding method and a multi-channel audio signal encoding method, which have an excellent effect that deterioration of localization is minimized and high compression efficiency can be obtained by controlling execution / non-execution in band units. An apparatus, a recording medium, and a music distribution system can be provided.

[Brief description of drawings]

FIG. 1 is a flowchart showing a multi-channel audio signal encoding method according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing step S4 in the second embodiment according to the present invention.

FIG. 3 is a flowchart showing step S4 in the third embodiment of the present invention.

FIG. 4 is a flowchart showing step S5 in the fourth embodiment of the present invention.

FIG. 5 is a flowchart showing step S5 in the fifth embodiment of the present invention.

FIG. 6 is a block diagram showing a multi-channel audio signal coding apparatus using the multi-channel audio signal coding method according to the first embodiment.

FIG. 7 is a block diagram showing intensity stereo frequency limit calculation means 4 using step S4 in the second embodiment.

FIG. 8 is a block diagram showing intensity stereo frequency limit calculation means 4 using step S4 in the third embodiment.

FIG. 9 is a block diagram showing band-wise intensity stereo determination means 5 using step S5 in the fourth embodiment.

FIG. 10 is a block diagram showing band-wise intensity stereo determination means 5 using step S5 in the fifth embodiment described above.

FIG. 11 is a block diagram showing a conventional audio signal encoding method.

FIG. 12 is a flowchart showing conventional intensity stereo processing.

[Explanation of symbols]

S1 auditory psychological analysis (auditory psychological analysis process) S2 time / frequency mapping (time / frequency mapping process) S3 TNS processing (TNS processing process) S4 intensity stereo frequency limit calculation (intensity stereo frequency limit calculation process) S5 intensity stereo determination for each band (Intensity stereo determination process for each band) S6 Intensity stereo processing (intensity stereo processing step) S7 Sum difference stereo processing (sum difference stereo processing step) S8 Quantization / encoding (quantization / encoding step) S10 Bitstream generation formation ( Bitstream generation process)

Claims (15)

[Claims] 1. A psychoacoustic analysis step of calculating a value of a signal-to-mask ratio based on an input signal, a time-frequency mapping step of converting the other input signal into spectrum data, and the converted spectrum data. A TNS processing step for controlling the shape based on the above, an intensity stereo frequency limit calculating step for calculating a frequency based on the controlled shape, and an approximation degree of the input signals of the left channel and the right channel for each scale factor band. Depending on the intensity stereo determination process for each band to determine whether to perform the intensity stereo processing according to, is one of the modules for processing the input signal,
An intensity stereo processing step of converting the two channels into connection energy and directional information of both channels; a sum difference stereo processing step of converting the input signals of the channels into respective sum and difference signals; Quantization / encoding step of quantizing and encoding the input signal applied in the stereo processing step, and bitstream generation forming of multiplexing the output of this quantization / encoding step and control information and shaping as a bitstream And the step of processing the intensity stereo frequency limit calculation step, the step of processing the value calculated from the psychoacoustic analysis step and the shape controlled from the TNS processing step, and the intensity stereo for each band. The determination step, the value calculated from the psychoacoustic analysis step and the strength stereo frequency limit calculated from the calculated Multichannel audio signal encoding method characterized by having a step of determining a wavenumber. 2. A coding mode information step for setting coding mode information necessary for coding, and a strength stereo preprocessing step for judging whether or not strength stereo can be used from the coding mode information. , A compression rate calculation step of calculating a compression rate from information obtained from the encoding mode information, a load determination step of outputting a result of comparing the compression rate with a predetermined value, and a result of the load determination step In the intensity stereo frequency limit calculating step including an IS frequency limit setting step of setting a frequency for starting the intensity stereo processing, the IS frequency limit is set when the load determining step determines that the compression rate is high. If the setting step has means for performing intensity stereo processing in a wider frequency band, and the load determination step determines that the compression rate is low, the IS frequency 2. The multi-channel audio signal coding method according to claim 1, wherein the limit setting step has means for performing intensity stereo processing in a narrower frequency band. 3. A coding mode information step for setting coding mode information necessary for coding, and a strength stereo pre-processing step for judging whether or not strength stereo can be used from the coding mode information. A compression rate calculation step of calculating a compression rate from information obtained from the encoding mode information, a determination threshold value calculation step of calculating a determination threshold value based on the compression rate calculated in the compression rate calculation step, and a determination threshold value calculation step An approximation degree calculating step of calculating the approximation degree of the left channel and the right channel based on the determination threshold value calculated in the step, an approximation degree determination processing step of outputting a result of comparing the determination threshold value and the compression rate, and this approximation The intensity stereo frequency limit calculation step including an IS frequency limit setting step of setting a frequency based on the result of the degree determination processing step, wherein the approximation degree determination processing step is the determination threshold value. When determining that the degree of approximation is high, the IS frequency limit setting step includes a step of performing intensity stereo processing in a wider frequency band, and when the degree of similarity determination processing determines that the degree of approximation is low. 2. The multi-channel audio signal encoding method according to claim 1, wherein said IS frequency limit setting step has means for performing intensity stereo processing in a narrower frequency band. 4. A scale factor band determination step of comparing the scale factor band with an IS frequency limit,
An approximation degree calculating step of calculating an approximation degree of the left channel and the right channel based on the scale factor band and the IS frequency limit compared in the scale factor band determination step, and an approximation degree calculated in the approximation degree calculating step and preset. Is a module for processing the input signal from the approximation degree determination processing step and the approximation degree determination processing step for comparing the magnitude of the two channels with the connection energy of both channels and directivity information. In the intensity stereo determination step for each band, including the intensity stereo processing step for converting to, when the approximation degree determination processing step determines that the approximation degree is high, the intensity stereo processing step performs intensity stereo processing. If there is a step and the approximation degree determination processing step determines that the approximation degree is low, the intensity stereo Multichannel audio signal encoding method according to claim 1, characterized in that comprising the step of returning to the scale factor band determination step without performing the process. 5. A scale factor band determination step of comparing the scale factor band with an IS frequency limit,
A judgment threshold value calculating step for calculating a judgment threshold value based on the scale factor band and the IS frequency limit compared in the scale factor band judging step, and approximation of the left and right channels based on the judgment threshold value calculated in the judgment threshold value calculating step. From an approximation degree calculation processing step of calculating a degree, an approximation degree determination processing step of comparing the determination threshold value calculated in the determination threshold value calculation step and the approximation degree calculated in the approximation degree calculation step, and an approximation degree determination processing step Which is one of the modules for processing the input signal of, and the intensity stereo determination step for each band, which comprises an intensity stereo processing step of converting the two channels into connection energy of both channels and directional information, When the approximation degree determination processing step determines that the degree of approximation is higher than the determination threshold value, the intensity stereo processing step is performed. And a step of performing a rheo process, and when the approximation degree determination process step determines that the approximation degree is low, the intensity stereo process step returns to the scale factor band determination step without performing intensity stereo process. The multi-channel audio signal encoding method according to claim 1, wherein the multi-channel audio signal encoding method comprises: 6. A psychoacoustic analysis means for calculating a value of a signal-to-mask ratio based on an input signal, a time / frequency mapping means for converting the other input signal into spectrum data, and the converted spectrum data. TNS processing means for controlling the shape based on the above, intensity stereo frequency limit calculating means for calculating the frequency based on the controlled shape, and approximation degree of the input signal of the left channel and the right channel for each scale factor band. In accordance with each band intensity stereo determination means for determining whether to perform the intensity stereo processing according to, one of the module for processing the input signal,
Intensity stereo processing means for converting the two channels into connection energy and directional information of both channels; sum difference stereo processing means for converting the input signals of the channels into respective sum and difference signals; Quantization / encoding means for quantizing and encoding the input signal applied by the stereo processing means, and bitstream generation forming for multiplexing the output of this quantization / encoding means and control information and shaping it as a bitstream. Means, the intensity stereo frequency limit calculation means includes means for processing the value calculated by the psychoacoustic analysis means and the shape controlled by the TNS processing means, and the intensity stereo for each band. The determination means, the value calculated from the psychoacoustic analysis means and the intensity stereo frequency limit calculation means Multichannel audio signal encoding apparatus characterized by having means for determining the wave number. 7. Encoding mode information means for setting encoding mode information necessary for encoding, and intensity stereo preprocessing means for determining whether or not intensity stereo can be used from the encoding mode information. , A compression rate calculating means for calculating a compression rate from information obtained from the encoding mode information, a load determining means for outputting a result of comparison between the compression rate and a predetermined value, and a result of the load determining means In the intensity stereo frequency limit calculating means including an IS frequency limit setting means for setting a frequency for starting the intensity stereo processing, the IS frequency limit is set when the load determining means determines that the compression rate is high. If the setting means has means for performing intensity stereo processing in a wider frequency band, and the load determination means determines that the compression rate is low, the IS frequency 7. The multi-channel audio signal coding apparatus according to claim 6, wherein the limit setting means has means for performing intensity stereo processing in a narrower frequency band. 8. Encoding mode information means for setting encoding mode information necessary for encoding, and intensity stereo pre-processing means for determining whether or not intensity stereo can be used from the encoding mode information. A compression rate calculation means for calculating a compression rate from information obtained from the encoding mode information, a determination threshold value calculation means for calculating a determination threshold value based on the compression rate calculated by the compression rate calculation means, and a determination threshold value calculation An approximation degree calculating means for calculating the approximation degree of the left channel and the right channel based on the determination threshold value calculated by the means, an approximation degree determination processing means for outputting a result of comparing the determination threshold value and the compression rate, and this approximation The intensity stereo frequency limit calculation means including an IS frequency limit setting means for setting a frequency based on the result of the degree judgment processing means, wherein the approximation degree judgment processing means is the judgment threshold value. When determining that the degree of approximation is high, the IS frequency limit setting means has means for performing intensity stereo processing in a wider frequency band, and when the degree of similarity determination processing means determines that the degree of approximation is low. 7. The multi-channel audio signal encoding device according to claim 6, wherein said IS frequency limit setting means has means for performing intensity stereo processing in a narrower frequency band. 9. An approximation degree calculating means for calculating the approximation degree of the left channel and the right channel based on the scale factor band and the IS frequency limit, and the approximation degree calculated by the approximation degree calculating means and a preset threshold value. One of the modules for processing the input signals from the similarity determination processing means for comparing the magnitude and the similarity determination processing means, and an intensity stereo for converting the two channels into connection energy of both channels and directional information. In the intensity stereo determination means for each band including a processing means, when the approximation degree determination processing means determines that the degree of approximation is high, the intensity stereo processing means has means for performing intensity stereo processing, Further, when the approximation degree determination processing means determines that the approximation degree is low, the scale stereo processing is performed without performing the intensity stereo processing. 7. The multi-channel audio signal encoding device according to claim 6, further comprising means for returning to the band determination means. 10. A determination threshold value calculating means for calculating a determination threshold value based on the scale factor band and the IS frequency limit, and an approximation for calculating a degree of approximation of the left and right channels based on the determination threshold value calculated by the determination threshold value calculating means. A degree calculation means, an approximation degree determination processing means for comparing the determination threshold value calculated by the determination threshold value calculation means and the approximation degree calculated by the approximation degree calculation means, and the input signal from the approximation degree determination processing means. The intensity stereo determination means for each band, which is one of the modules for processing, and comprises intensity stereo processing means for converting the two channels into connection energy of both channels and directivity information, wherein the approximation degree determination processing is performed. When the means determines that the degree of approximation is higher than the determination threshold, the intensity stereo processing means has means for performing intensity stereo processing, and 7. The intensity stereo processing means includes means for returning to the scale factor band determination means without performing intensity stereo processing when the approximation degree determination processing means determines that the approximation degree is low. The multi-channel audio signal encoding device according to. 11. A multi-channel audio signal coding apparatus, wherein the input signal is quantized and coded by using the multi-channel audio signal coding method according to any one of claims 1 to 5. . 12. A recording medium in which the steps of the multi-channel audio signal encoding method according to claim 1 are stored. 13. A music distribution system comprising the recording medium according to claim 12 as a constituent element. 14. A music distribution system, wherein music is distributed by using the multi-channel audio signal encoding method according to any one of claims 1 to 5. 15. A music distribution system comprising the multi-channel audio signal encoding device according to any one of claims 6 to 10 as a constituent element.