JPH0591062A - Audio signal processing method - Google Patents

Abstract

PURPOSE:To improve an audio quality by sharing bits of a parameter not recorded nor sent for quantization of an audio signal. CONSTITUTION:A scale factor SF of a parameter BF and data of a word length ML for each block relating to the block floating processing are recorded or sent up to a band requiring the parameter BF for each time frame and number of the parameters recorded or sent is recorded or sent.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an audio signal processing method for compressing an audio signal by so-called block floating processing.

[0002]

2. Description of the Related Art Conventionally, as an audio signal processing method, a high-efficiency coding technique of compressing and coding an audio signal includes, for example, input audio signal (digital audio data) at predetermined time intervals (every predetermined time frame). )
There is a method in which a block is divided into a plurality of blocks on the frequency axis, so-called block floating processing is performed for each block, and data in each block is quantized by adaptive bit allocation.

Here, the block floating process is basically to multiply each word in the block by a common value to increase the size, thereby improving the precision in quantization.
Specifically, for example, the largest absolute value (maximum absolute value) among the absolute values of each word in the block is searched for, and a floating coefficient common to all words in the block so that the maximum absolute value is not saturated. As an example, a floating process is performed by using. As a simpler one, there is a floating unit of 6 dB using bit shift.

On the encoder side of a system to which the audio signal processing method for performing the block floating process is applied, usually, as a parameter BF related to the block floating process, for example, a value of a scale factor SF as a floating coefficient, The data of the word length WL indicating the difference between the scale factor SF and the allowable noise level obtained for each block in consideration of the so-called masking effect is quantized block data (hereinafter referred to as main information, for example). ) Is recorded or transmitted to the medium.

The masking effect means a phenomenon in which one sound is masked by another sound and becomes inaudible due to human auditory characteristics. In other words, the above-mentioned masking is a phenomenon in which one signal masks another signal so that it cannot be heard.This masking effect includes the time-axis masking effect of the audio signal on the time axis and the masking effect on the frequency axis. There is a simultaneous masking effect by the signal of. Due to these masking effects, even if there is noise in the masked portion, this noise will not be heard. Therefore, in the actual audio signal, the noise within the masked range is regarded as an acceptable noise.

[0006]

On the encoder side of the system to which the conventional audio signal processing method is applied, the number of parameters BF to be recorded or transmitted is fixed for all time frames. Things are common. FIG. 9 shows a state of recording (or transmitting) data in each time frame on the encoder side of the conventional audio signal processing system. In the case of the example of FIG. 9, it is necessary to record or transmit the value of the parameter BF even for a block to which bits are not actually allocated, and accordingly, the number of bits allocated to audio data is correspondingly increased. In particular, when the compression rate is high (when the bit rate is low), it is difficult to secure sufficient sound quality on the decoder side corresponding to the encoder side.

On the other hand, as shown in FIG. 10, the value of the scale factor SF is not recorded or transmitted to a block (block of word length WL = 0) to which no bit is actually allocated, and A system has also been proposed in which many bits are allocated to audio data according to minutes.
In FIG. 10, the number of scale factors SF recorded is reduced by the number of blocks to which no bits are assigned (blocks with 0 bit assignment).

However, in this case, the data of the word length WL must always be recorded or transmitted. Further, when the scale factor SF is read by the decoder side later, it is necessary to check for each block whether or not the word length WL of the block is 0.

Further, the encoder usually calculates the number of bits required to quantize the audio signal of each block by the calculation for obtaining the masking, and then the total number of bits allocated to the time frame. And the bit allocation to each block is adjusted. However, at this time, if whether or not the scale factor SF of the block is recorded changes depending on the bit allocation of each block as described above, the total number of bits that can be allocated to the main information (audio data) also changes accordingly. However, adjusting the bit allocation becomes very complicated.

Therefore, the present invention has been proposed in view of the above-mentioned circumstances, and it is relatively easy to adjust the bit allocation, and audio quality does not deteriorate even if the bit allocation is adjusted. It is an object of the present invention to provide a signal processing method.

[0011]

The audio signal processing method of the present invention has been proposed in order to achieve the above-mentioned object, and an input audio signal is divided into a plurality of blocks on the frequency axis at every predetermined time frame. Block, each block is subjected to block floating processing, data of each block is quantized by adaptive bit allocation, and the quantized audio signal is recorded or transmitted, and is related to the block floating processing. An audio signal processing method for recording or transmitting a parameter, comprising recording or transmitting a parameter for each block related to the block floating process up to a required band of the parameter for each time frame, and recording or transmitting the parameter. The number of parameters to be recorded is also recorded or transmitted.

That is, as the band of the audio signal actually recorded or transmitted for each time frame, for example, high frequency recording or transmission is not performed because it is difficult to be perceived (difficult to hear). In this case, the parameters (scale factor, word length value) of the high frequency block of the time frame are not recorded or transmitted, and those bits are perceived as important low-frequency main information (of the block). Quantized data). Therefore, the data to be recorded or transmitted becomes the low frequency parameter and the main information. In this case, the number of parameters recorded or transmitted for each time frame is also recorded (or transmitted).
I am trying.

[0013]

According to the audio signal processing method of the present invention, the parameters for each block are recorded or transmitted for each time frame up to the band in which this parameter is required, that is, the parameters in the band not requiring the parameter are recorded or transmitted. Instead, the bits are allocated to the code of the audio data.

[0014]

Embodiments of the present invention will be described below with reference to the drawings.

In the audio signal processing method of this embodiment, the time-series input audio signal TS is converted into a frequency axis signal (spectrum signal SP) for each predetermined time frame, and the spectrum signal SP is divided into blocks for each band. The data is divided into blocks, a block floating process is performed for each block, the data of each block is quantized by adaptive bit allocation, the quantized audio signal is recorded or transmitted, and the parameters related to the block floating process. An audio signal processing method for recording or transmitting a floating coefficient (scale factor SF) as BF and data of a word length WL, as shown in FIG. 1, a parameter BF for each block related to the block floating process. The parameter BF is required for each of the above time frames. Thereby recorded or transmitted to a band, the number N of parameters BF being the recording or transmission is also obtained by such recording or transmission.

FIG. 1 shows one time frame and the one
It shows blocks of multiple frequency bands in one time frame. The main information in FIG. 1 is quantized block data (quantized block data of the spectrum signal SP for each block obtained by frequency-dividing the audio signal TS).

The effectiveness of the embodiment of the present invention is based on the following facts.

That is, in a high band of 10 kHz or higher, for example, the so-called minimum audible limit, which will be described later, based on human auditory characteristics is high, and since the masking effect by the low-frequency signal works effectively, quantization noise is less than that. Even if it becomes larger than the band (for example, 10 kHz or less), deterioration of sound quality is hard to be perceived by the ear. In particular, in particular, even if the signal component in the band of 15 kHz or less is deleted (0 bit is allocated), there are many cases in which the difference is not audible.

Therefore, in the present embodiment, as shown in FIG. 1, the scale factor SF as a floating coefficient which is a parameter BF of each block related to the block floating processing and the number of allocated bits at the time of quantization are set. The corresponding data of the word length WL is
The parameter BF is recorded or transmitted for each time frame up to a required band (that is, only a low frequency band which is important for hearing). As a result, many bits can be assigned to a low-frequency code that is important for hearing and cannot be omitted, and as a result, sound quality is improved.

At the minimum audible limit, if the noise absolute level is below this minimum audible limit, the noise cannot be heard. Even if the coding is the same, the minimum audible limit varies depending on, for example, a difference in reproduction volume during reproduction. However, in a realistic digital system, for example, a method of inputting music to a 16-bit dynamic range is very different. Therefore, if the quantization noise in the most audible frequency band around 4 kHz is not heard, it is considered that the quantization noise below the level of the minimum audible curve is not heard in other frequency bands.

Further, in the above-described manner, the number of the above parameters BF changes for each time frame. Therefore, in the present embodiment, the number N of the parameter BF is recorded or transmitted for each time frame. There is. in this case,
The data amount of the number N is small. For example, parameter B
If the number of Fs is about several tens, 7 bits is sufficient, and the bandwidth that can be taken in the time frame is 2
In the case of types (two blocks), recording or transmission is possible with 1 bit.

As described above, in the present embodiment, the band of the audio signal actually recorded or transmitted for each time frame is, for example, a high frequency band because it is hard to be perceived (difficult to hear). When recording or transmission is not performed, bits corresponding to parameter BF (scale factor SF, value of word length WL) of the high frequency block of the time frame are allocated for low frequency main information, and The parameter BF and the main information are recorded (or transmitted) on a medium, for example. Further, the number N of the parameters BF for each time frame is also recorded (or transmitted).

If the signal level in the low frequency range is low and the signal level in the high frequency range is very high, deleting the high frequency signal makes it easier to perceive the difference in sound quality deterioration. .. Therefore, in this case, as shown in FIG. 2, the high frequency signal and the parameter BF are also recorded and encoded.

FIG. 3 shows a concrete configuration of an encoder of an audio signal processing system to which the audio signal processing method of this embodiment is applied.

In FIG. 3, a time series audio signal (waveform data) is divided by a band division filter, and this is converted into a frequency axis signal by a modified discrete cosine transform (MDCT) process. (Data) SP is encoded.

That is, the encoder shown in FIG.
The time-sequential input audio signal (digital audio data) TS supplied to the input terminal 1 is divided into a plurality of blocks on the frequency axis for each predetermined time frame, and each block is subjected to block floating processing and Data for each block is encoded by adaptive bit allocation, and parameters for each block related to the block floating process (the scale factor SF as a floating coefficient and the word length corresponding to the bit allocation number) The data recording circuit 19 records the (WL) BF for each time frame up to the required band and also records the number of the recorded parameters. The data recorded in the data recording circuit 19 is transmitted to the decoder side described later.

Here, in the encoder of this specific example, the band of each block obtained by dividing the time frame into a plurality of frequency bands is made variable, and when the band is narrow, as in the above-described embodiment. Then, the number of parameters BF recorded in the data recording circuit 19 is reduced, and the reduced bits are used for the main information (for the quantization of the spectrum signal SP for each block obtained by frequency-dividing the audio signal TS described later). I'm trying to allocate it.

In order to do this, the time-series audio signal (waveform data) TS supplied to the input terminal 1 for each predetermined time frame is the time for converting the supplied time-series signal into a spectrum signal. / Frequency conversion circuit 11
Is converted into a spectrum signal SP by. The time / frequency conversion circuit 11 divides the time-series audio signal TS into predetermined time frames and divides each time frame into a plurality of frequency bands, as shown in FIGS. 1 and 2, for example. It is divided into blocks and the band of each block is variable.

Further, when the band of the block is made variable by the time / frequency conversion circuit 11, each band obtained by dividing the time frame by, for example, band division in consideration of human auditory characteristics is used as the block width. It is designed to be a block. That is, the spectrum signal SP is divided into a plurality of bands with a bandwidth which is generally called a critical band and becomes wider in a higher frequency band. It is divided into three bands: the range, the middle range, and the low range. The critical band is a frequency band divided in consideration of human auditory characteristics, and when the pure tone is masked by a narrow band noise of the same strength in the vicinity of the frequency of a pure tone, the noise of that pure tone is masked. It is the bandwidth that you have. The division in the critical band is, for example, 0 to 20k.
It is also possible to divide the entire frequency band of Hz into 25 critical bands, for example.

The spectrum signal SP from the time / frequency conversion circuit 11 is sent to the spectrum signal quantization circuit 15 and quantized. That is, the spectrum signal quantization circuit 15 normalizes (normalizes) the supplied spectrum signal SP of each block by the block floating process, and then quantizes it with an adaptive allocation bit number in consideration of a so-called masking effect.

Here, the spectrum signal quantization circuit 1 is used.
The floating factor (scale factor SF) for performing the block floating process in 5 is supplied from the scale factor calculation circuit 13. That is, the spectrum signal SP is supplied to the scale factor calculation circuit 13, and the scale factor calculation circuit 13 outputs, for example, a peak or an average value of the spectrum signal SP for each block in the plurality of frequency bands for each time frame. Is multiplied by a predetermined coefficient to output a floating coefficient (scale factor SF).

The spectrum signal quantizing circuit 15 is also used.
In order to quantize the adaptive number of allocated bits, the spectrum signal SP is masked by the masking calculation circuit 17
Has also been sent to. In the masking calculation circuit 17,
As will be described later, masking information MSKI for each block according to human auditory characteristics and / or masking information MSKI of the target block due to a masking effect from another block adjacent to the target block are obtained. The masking information MSKI from the masking calculation circuit 17 is sent to the bit allocation calculation circuit 14, and the bit allocation calculation circuit 14 uses the word length as bit allocation information for each block based on the masking information MSKI. WL data is required. That is, the spectrum signal quantization circuit 15 performs adaptive quantization for each block of the supplied spectrum signal SP based on the information of the word length WL.

Here, the processing in the masking calculation circuit 17 and the bit allocation calculation circuit 14 is specifically as follows.

That is, in the spectrum signal SP sent to the masking calculation circuit 17, the energy is first calculated for each block. When calculating the energy for each block, for example, the energy for each critical band (critical band) is obtained by, for example, calculating the sum of the amplitude values in the band. Instead of the energy for each band, the peak value, the average value, etc. of the amplitude value may be used. The spectrum of the sum total value of each band obtained by this energy calculation is generally called a Bark spectrum.

Next, in the masking calculation circuit 17,
In order to consider the influence of the above-described Bark spectrum on so-called masking, a convolution process is performed such that the Bark spectrum is multiplied by a predetermined weighting function and added. As a configuration for performing this convolution processing, for example, a plurality of delay elements that sequentially delay input data and a plurality of multipliers that multiply outputs from these delay elements by a filter coefficient (weighting function) (for example, corresponding to each band) 25 multipliers) and a sum adder that sums the outputs of the respective multipliers.

After the above convolution processing is performed, inverse convolution processing is performed to obtain a masking threshold. That is, this masking threshold has an allowable noise spectrum. Here, the masking threshold is subtracted from the Bark spectrum to obtain a level at which the Bark spectrum is masked by the masking threshold. This masking level is sent to the bit allocation calculation circuit 14 as the masking information MSKI.

When obtaining the masking information MSKI, for example, the data showing the minimum audible limit, which is a human auditory characteristic, and the masking threshold can be combined. In this minimum hearing limit,
As described above, if the absolute noise level is below this minimum audible limit, the noise will not be heard.

The bit allocation calculation circuit 14 is, for example, a ROM in which information on the number of allocated bits is stored in advance.
Etc. are provided, and the allocated bit number information for each band is obtained from the ROM or the like according to the level of the difference between the masking level of the masking information MSKI and the energy of each band. Further, based on the information on the number of allocated bits for each band, the data of the word length WL corresponding to the number of allocated bits for each block of the high band, the middle band, and the low band is obtained.

Further, the bit allocation calculation circuit 14 can correct the allowable noise level based on the masking information MSKI, for example, based on the information of the equal loudness curve. Here, the equal loudness curve is a characteristic curve relating to human auditory characteristics,
For example, it is a curve obtained by finding the sound pressure of a sound at each frequency that sounds the same as a pure tone of 1 kHz, and is also called a loudness isosensitivity curve. It should be noted that this equal loudness curve is a curve that is substantially the same as the above-mentioned minimum audible limit curve. Therefore, in this equal loudness curve, for example, in the vicinity of 4 kHz, even if the sound pressure is reduced by 8 to 10 dB from 1 kHz, it sounds as loud as 1 kHz, and conversely, in the vicinity of 50 kHz, it is about 15 dB higher than that at 1 kHz. Without it, it doesn't sound the same. Therefore, it is understood that the noise (allowable noise level) exceeding the level of the minimum audible curve should have the frequency characteristic given by the curve corresponding to the equal loudness curve. From this, it is understood that correcting the permissible noise level in consideration of the equal loudness curve is suitable for human auditory characteristics.

Further, in the masking calculation circuit 17,
By performing the masking calculation, when the spectrum signal quantization circuit 15 quantizes the spectrum signal SP and records it in the data recording circuit 19, which band is the highest band HB to be recorded is also obtained. There is. That is, by taking the masking effect and the like into consideration, the band in which the sound quality on hearing is not affected even if it is not recorded is separated from the band in which the hearing is adversely affected if it is not recorded, and the information of the band HB to be recorded is divided. , To the parameter number calculation circuit 18 which calculates the number of parameters BF of the block floating process.

In the parameter number calculation circuit 18, based on the information of the band HB to be recorded, of the parameters BF of each block, a band equal to or less than the band HB to be recorded (that is, a low band to be recorded). The number N of the parameters BF is calculated. Alternatively, the number of parameters BF in a band equal to or higher than the band HB to be recorded (that is, a high band that may not be recorded) may be calculated.

The information on the number N is sent to the bit allocation calculation circuit 14. Therefore, in the bit allocation calculation circuit 14, the bit allocation number calculation process as described above and the band HB or more to be recorded on the basis of the information of the number N (that is, it is not necessary to record, for example, Area), bits are not allocated. That is, in the bit allocation calculation circuit 14, if the number N is obtained,
Since the total number of bits that can be assigned to the main information is obtained, the extra number of bits is assigned to the spectrum signal SP in the band up to the highest band HB to be recorded according to the number N.

The data recording circuit 19 records the number N of the parameters BF, the N parameters BF, and the quantized spectrum signal QSP.

The coded data CDT from the decoder recording circuit 19 is output via the output terminal 2.

FIG. 4 shows a specific configuration of the time / frequency conversion circuit 11 of the encoder shown in FIG.

In the configuration of FIG. 4, for example, a signal is compressed by combining a band division filter such as a QMF filter and a modified discrete cosine transform (MDCT). The above QMF filter is a 1976R.E Crochiere, Digit
al coding of speech in subbands, Bell Syst. Tech.
J. Vol.55, No.8 1976. Also, ICAS
SP 83, BOSTON, Polyphase Quadrature filter -A new s
The ubband coding technique, Joseph H. Rothweiler, describes an equal bandwidth filter partitioning technique.
Regarding the above MDCT, ICASSP 1987 Subband /
Transform Coding Using Filter Bank Designs Based on
Time Domain Aliasing Cancellation, JPPrincen,
ABBradley, Univ. Of Surrey Royal Melbourne Inst.
of Tech. Instead of the MDCT, for example, a fast Fourier transform (FFT), a discrete cosine transform (DCT) or the like may be performed to convert the time axis into the frequency axis.

In the configuration of FIG. 4, the bandwidth of the input audio signal TS such as a time-series PCM signal becomes wider as it goes higher, based on the so-called critical band in consideration of human auditory characteristics. The frequency is divided into. In this specific example, considering the above critical band, the band is roughly divided into three bands of a high band, a middle band, and a low band. The band division may be a block in which the critical band width is further subdivided in the critical band unit or in the high band.

That is, in FIG. 4, the input audio signal TS, which is an audio PCM signal of 0 to 20 kHz, for example, is supplied to the input terminal 1. This input audio signal TS is, for example, 0 to 10 kHz by a band division filter 71 such as a so-called QMF filter.
Band and 10 kHz to 20 kHz band (high frequency band), and signals in the 0 to 10 kHz band are also called QMF.
For example, 0 to 5 depending on the band division filter 72 such as a filter.
It is divided into a kHz band (low band) and a 5 kHz to 10 kHz band (middle band). The high-frequency (10 kHz to 20 kHz band) signal from the band-division filter 71 is sent to the MDCT circuit 73, which is an example of an orthogonal transformation circuit, and the medium-frequency (5 kHz to 10 kHz band) signal from the band-division filter 72 is MDCT. It is sent to the circuit 74, and the band division filter 7
The signals in the low frequency band (0 to 5 kHz band) from 2 are sent to the MDCT circuit 75 to be respectively MDCT processed. These MDCT-processed high-frequency signals are output to the terminal 76.
, The mid-range signal is output via a terminal 77, and the low-range signal is output via a terminal 78.

Here, each MDCT circuit 13, 14, 15
In the concrete example, the block size of (1) is such that the frequency band is widened toward the higher frequency side and the time resolution is increased (block length is shortened). That is, the number of samples in one block is, for example, 256 samples for a signal in the low frequency band of 0 to 5 kHz and a signal in the middle frequency band of 5 kHz to 10 kHz, and for a signal in the high frequency band of 10 kHz to 20 kHz. Divides one block into blocks each having a length of ½ of each of the blocks on the low band and the middle band. In this way, the number of orthogonal transform block samples in each band is the same. Further, each band can be further adaptively divided into ½, ¼, etc. blocks assuming a case where a signal temporal change is large.

FIG. 5 shows the configuration of a decoder corresponding to the encoder of FIG.

That is, in FIG. 5, the input terminal 5
1 is supplied with the coded data CDT. The data CDT includes a quantized spectrum signal reading circuit 54 for reading (retrieving) a quantized spectral signal from the data CDT, a parameter number reading circuit 52 for reading (retrieving) the number N of the block floating parameters BF, and The data of the block floating parameter BF is sent to the parameter reading circuit 53 which reads (takes out) the data.
The quantized spectrum signal QSP from the quantized spectrum signal reading circuit 54, the number N of data from the parameter number reading circuit 52, the parameter BF from the parameter reading circuit 53, that is, the scale factor SF and the word length WL. The data of 1 is sent to the spectrum signal restoration circuit 55. The spectrum signal restoration circuit 55 performs a decoding process using the supplied signal. The spectrum signal RSP decoded by the spectrum signal restoration circuit 55 is the frequency / time conversion circuit 5
6 as a time series audio signal RTS and output terminal 5
It is output from 7.

FIG. 6 shows a specific structure of the frequency / time conversion circuit 56 having the structure of the decoder shown in FIG.

In FIG. 6, the spectrum signal RSP of each block is given to the terminals 61, 62 and 63, and the IMDCT (inverse MDCT) circuits 64, 65 and 66 convert the signals on the frequency axis on the time axis. Converted to a signal. The signals on the time axis of these subbands are IQMF (inverse QM
F) The signals are decoded into full band signals by the circuits 67 and 78 and taken out from the terminal 69.

FIG. 7 shows a flowchart of signal processing in the encoder of FIG.

That is, in the flowchart of FIG. 7, in step S1, the time / frequency conversion circuit 1 is operated.
The time series audio signal TS input by 1 is converted into a spectrum signal (spectrum data) SP. After step S1, the process proceeds to step S2, and the scale factor calculation circuit 13 performs the calculation for obtaining the scale factor SF. In step S3, the masking calculation circuit 17 performs the masking calculation process, and in step S4, the parameter number calculation circuit 18 determines the band to which the bit is to be allocated and the number N of the parameter BF of the block floating process. In step S5, the bit allocation calculation circuit 14 calculates the bit allocation and the word length WL.
Is calculated. In step S6, the spectrum signal quantization circuit 15 quantizes the spectrum signal (spectrum data). Finally step S7
Then, the data recording circuit 19 records the number N of data, records the parameter BF of the block floating process (records the parameter BF of the band to be recorded), and the quantized spectrum signal (spectrum data). Recording of QSP data is performed.

FIG. 8 shows a flowchart of processing in the configuration of the decoder shown in FIG.

That is, in the flowchart of FIG. 8, first, in step S11, the parameter number reading circuit 52 first reads the number N of the parameters BF, and then in step S12, the parameter reading circuit 53 reads the number N. Parameter B only
Read F data. Subsequently, in step S13, the quantized spectrum signal reading circuit 54 reads the quantized spectrum signal QSP according to the word length WL of the parameter BF. In step S14, the spectrum signal restoration circuit 55 causes the scale factor SF and the word length WL.
Based on the above, the read quantized spectrum signal QSP is restored as an approximate value of the original spectrum signal SP (the spectrum signal RSP). Finally, in step S15, the frequency / time conversion circuit 57
The audio signal RTS is restored by performing a process (inverse MDCT) opposite to the MDCT on the spectrum signal RSP and passing the band synthesis filter.

Although the above-described embodiment of the present invention has described the system for encoding the signal obtained by converting the time-series input audio signal TS into the spectrum signal SP, the present invention describes the time-series signal. It can also be applied to a system in which sub-bands are divided and encoded (so-called sub-band coding).

[0059]

As described above, in the audio signal processing method according to the present invention, the parameters for each block related to the block floating process are recorded or transmitted for each time frame up to the required band, and Since the number of parameters to be recorded or transmitted is also recorded or transmitted, adjustment of bit allocation is relatively easy, and even if adjustment of bit allocation is performed, sound quality is not deteriorated. In other words, because it is hard to be perceived in practice, for example, when a high frequency signal is not recorded, the high frequency parameter bits can also be allocated for quantization of the low frequency audio signal, improving the sound quality. You can let me do it. Further, even when a high level signal is included in the high frequency band, recording can be performed without narrowing the band. Furthermore, these processes can be realized with a simple configuration.

[Brief description of drawings]

FIG. 1 is an example of data recording or transmission in a method according to an embodiment of the present invention.
It is a figure for demonstrating (when a high-frequency signal is not recorded or transmitted).

FIG. 2 is an example of data recording or transmission in the method of the embodiment of the present invention.
It is a figure for demonstrating (when a high-frequency signal is also recorded or transmitted).

FIG. 3 is a block circuit diagram showing a schematic configuration of an encoder.

FIG. 4 is a block circuit diagram showing a specific configuration of the time / frequency conversion circuit having the configuration of FIG.

FIG. 5 is a block circuit diagram showing a schematic configuration of a decoder side.

6 is a block circuit diagram showing a specific configuration of the frequency / time conversion circuit having the configuration of FIG.

FIG. 7 is a flowchart of processing performed by the encoder.

FIG. 8 is a flowchart of processing in a decoder.

FIG. 9 is a diagram for explaining an example of recording or transmission data in a conventional system (in the case of a system in which the number of parameters is constant).

FIG. 10 is an example of recording or transmission data in a conventional system.
It is a figure for demonstrating (in the case of a system with a variable number of scale factors).

[Explanation of symbols]

 WL: Word length SF: Scale factor 11: Time / frequency conversion circuit 13: Scale factor calculation circuit 14: Bit allocation Case calculation circuit 15 ... Spectrum signal quantization circuit 17 ... Masking calculation circuit 18 ... Parameter number calculation circuit 19 ... Data recording circuit

Claims (1)

[Claims] 1. An input audio signal is divided into a plurality of blocks on a frequency axis for each predetermined time frame, block floating processing is performed for each block, and data for each block is quantized by adaptive bit allocation. An audio signal processing method for recording or transmitting the quantized audio signal and recording or transmitting a parameter related to the block floating process, wherein the parameters for each block related to the block floating process are An audio signal processing method, wherein the parameter is recorded or transmitted up to a required band for each time frame, and the number of parameters recorded or transmitted is also recorded or transmitted.