WO2013189030A1

WO2013189030A1 - Monophonic or stereo audio coding method

Info

Publication number: WO2013189030A1
Application number: PCT/CN2012/077155
Authority: WO
Inventors: 王磊; 闫建新
Original assignee: 深圳广晟信源技术有限公司
Priority date: 2012-06-19
Filing date: 2012-06-19
Publication date: 2013-12-27
Also published as: CN104170007B; CN104170007A

Abstract

The present invention provides a monophonic or stereo audio coding method. The method comprises: dividing a monophonic or stereo audio signal into a basic layer and at least one enhanced layer; coding the basic layer by using a coding mode of mp3, AAC, SBR, PS and/or DRA; and coding the at least one enhanced layer by using a coding mode of mp3, AAC, SBR, PS, DRA, residual coding, a coding algorithm of part of parameters and/or a coding algorithm of parameters respectively. In the present invention, rough layering is performed on monophonic or stereo audio, where merely 2 or 3 layers are divided; in this manner, compression with a higher efficiency can be ensured in an easy manner, free of all kinds of technical constraints in a fine layering technology. The optimal comprehensive sound quality can be obtained by flexibly controlling the quality of each layer of sound track, and channel coding requirements can be easily met.

Description

Method for encoding mono or stereo

Field of the Invention This invention relates to the field of audio coding processing, and more particularly to a method of encoding mono or stereo.

In layered audio coding, there are already lossy digital audio coding methods and lossless audio coding techniques by fine layering, such as ISO/IEC 14496-3 MPEG-4 BSAC (Bit sliced arithmetic coding) bit slice arithmetic coding, The MPEG-4 BSAC encoding method and the MPEG-4 SLS (Scalable Lossless Coding) lossless enhancement layer method used in AVS (Audio Video coding Standard Workgroup of China) can achieve fine layering of audio for each layer. Coded separately. However, the fine layering method has the disadvantages of low coding efficiency, complicated structure, and high processing logic complexity.

There is also a non-fine layered coding scheme in the prior art: AAC-SSR (Advanced Audio Coding-Scalable Sampling Rate) is provided in both MPEG-4 Part 3 and MPEG-2 Part 7 ), first proposed by Sony, the encoding architecture is similar to its unique ARTAC (Adaptive Transform Acoustic Coding) encoding. The coding scheme first divides the input digital audio signal into four frequency bands through a 4-band polyphase quadrature filter (PQF, PJF, and then performs one 256-point MDCT for each of the four frequency bands.

(512 sample window length) or 8 32 points (64 sample window length) MDCT. The coding scheme can also reduce the data rate by removing the high PQF band, and achieve bitstream layering by reducing the frequency band, thereby obtaining different bit rates and sampling rates. The advantage of this coding scheme is that independent blocks can be selected independently in each frequency band. Or short block MDCT, so the high-frequency can use short block coding to enhance the time resolution; and the low-frequency use long block coding to obtain high frequency resolution. However, due to the aliasing between the four PQF bands, the coding efficiency of the transform domain coefficients of adjacent parts will decrease. Summary of the invention

In order to solve the above technical problem, the present invention provides a method for encoding mono or stereo, comprising: dividing a mono or stereo audio signal into a basic layer and at least one enhancement layer; using mp3, A for the base layer AC, SBR, PS, and/or DRA coding mode coding; encoding at least one enhancement layer using mp3, AAC, SBR, PS, DRA, residual coding, partial parameter coding algorithm, and/or parameter coding algorithm, respectively.

Preferably, the above dividing the mono or stereo audio signal into a base layer and an enhancement layer is: dividing the mono or stereo audio signal into a base layer and an enhancement layer based on the frequency band, and the base layer is mono or The low frequency encoding portion of the stereo; the enhancement layer is a mono or stereo high frequency encoding portion; or the stereo audio signal is divided into a base layer and an enhancement layer based on the channel, and the base layer transmits the left channel or the channel; The layer transmits the right channel or the difference channel; or the stereo audio signal is divided into a base layer and an enhancement layer based on the parametric stereo coding, the base layer transmits a single channel of the left and right channel downmix; the enhancement layer transmits the parameter stereo information; or The mono or stereo audio signal is divided into a base layer and an enhancement layer based on the residual differential layer structure.

Preferably, the foregoing base layer and/or at least one enhancement layer are respectively coded by using a bandwidth extension algorithm.

Preferably, the step of separately encoding the base layer and the enhancement layer obtained by dividing the residual difference layer structure comprises: supplementing the base layer low frequency coding part according to the enhancement layer low frequency residual; and modifying the parameter to the base layer by using the enhancement layer bandwidth extension. The bandwidth extension parameters are adjusted.

Preferably, in the case where the audio signal is stereo, the base layer includes encoding the downmixed channel low frequency portion for encoding and bandwidth extension and parametric stereo encoding information; and the enhancement layer transmits the residual encoding of the low frequency portion. Preferably, in the case where the audio signal is stereo, the base layer transmits the low frequency partial coding information of the downmixed mono signal; the enhancement layer transmits the low frequency partial residual coding information and the bandwidth extension and the parameter stereo coding information.

Preferably, the step of encoding the base layer includes: encoding according to a code rate requirement of the base layer, and putting the obtained encoded data into a base layer transmission; comparing the original audio with the restored audio of the base layer decoding to obtain a residual signal. And the step of encoding the enhancement layer is to encode the residual signal as an enhancement layer.

Preferably, the dividing the mono or stereo audio signal into a base layer, the first enhancement layer and the second enhancement layer is: dividing the mono or stereo audio signal into a base layer, a first enhancement layer, and a second enhancement layer, wherein the base layer is a mono or stereo low frequency coding portion; the first enhancement layer is a mono or stereo intermediate frequency coding portion; and the second enhancement layer is a mono or stereo high frequency coding portion.

Preferably, the above-described residual channel layer structure divides the mono or stereo audio signal into a basic layer and at least one enhancement layer; and the step of encoding the base layer includes: encoding according to the code rate requirement of the base layer, The full-band basic quality coded data is placed in the base layer transmission; the original audio is compared with the base layer decoded and recovered audio to obtain a first-stage residual signal; and the first enhancement layer and/or the second enhancement layer are encoded The method includes: encoding the first-level residual signal as the data of the first enhancement layer; removing the signal decoded and restored by the first enhancement layer from the input first-stage residual signal of the first enhancement layer coding, to obtain the second level Residual signal; encoding the second-level residual signal as the data of the second enhancement layer; sequentially obtaining the next-level residual signal according to the residual signal of the previous stage, and encoding the residual signal of the next-level as the next The data of the level enhancement layer is encoded until all enhancement layers are completed.

Preferably, the step of encoding the base layer comprises: performing MDCT transformation on the time domain data x[n] to obtain a spectral coefficient X[k] at the encoding end _; dividing the frequency domain coefficient into a plurality of subbands, belonging to the subband b The spectral coefficient is divided by a quantization step; the quantization step is rounded (nint) to obtain the quantized spectral coefficient

Each quantization step size and spectral coefficient ^X [W is transmitted to the decoder. Preferably, the step of separately encoding the at least one enhancement layer comprises: performing MDCT transformation on the time domain data x[n] to obtain a spectral coefficient X[k] at the encoding end _; dividing the frequency domain coefficient into a plurality of subbands, belonging to The spectral coefficient of subband b is divided by a quantization step; after the quantization step is rounded (nint) to be quantized

; Each spectral coefficients and quantization step size ^XW transmitted to the decoding side; restored by the inverse quantization step size and quantization spectrum ^{f W is} the number of spectral coefficients ^f

^k] = A _b - X[k] . The inverse spectral coefficient W is subtracted from the original spectral coefficient to obtain the residual spectral coefficient EE[k] = X[k] - X[k] . The number is divided into multiple sub-bands, and the spectral coefficient belonging to the sub-band c is divided by a residual spectral coefficient quantization step, and the quantized residual is obtained by nint

Transmitting the residual spectral coefficient quantization step size and the quantized residual spectral coefficient to the decoding end.

The present invention performs coarse layering on mono or stereo, generally only 2 or 3 layers, and is simple to implement to ensure more efficient compression without the various constraints of fine layering technology. The best integrated sound quality can be obtained by flexibly controlling the quality of each channel; it is easy to meet channel coding requirements. DRAWINGS

1 is a schematic diagram of layering a mono or stereo according to an embodiment of the present invention;

2 is a schematic diagram of a coding process of an embodiment of the present invention;

3 is a schematic diagram of layering an audio signal based on a layered structure of a frequency band according to an embodiment of the present invention;

4 is a schematic diagram of layering an audio signal based on a layered structure of a channel according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of layering an audio signal based on a layered structure of parametric stereo coding according to an embodiment of the present invention; FIG. 6 is a schematic diagram of a layered structure according to an embodiment of the present invention;

FIG. 7 is a schematic diagram of layering an audio signal based on a hierarchical structure of residuals according to an embodiment of the present invention; FIG.

FIG. 8 is a schematic diagram of a two-layer structure based on a residual difference layer when a base layer has a bandwidth extension algorithm according to an embodiment of the present invention; FIG.

FIG. 9 is a schematic diagram of a two-layer structure based on a residual difference layer when the enhancement layer has a bandwidth extension algorithm according to an embodiment of the present invention; FIG.

FIG. 10 is a schematic diagram of a two-layer architecture based on a residual differential layer with bandwidth extension and bandwidth extension correction in an enhancement layer according to an embodiment of the present invention; FIG.

FIG. 11 is a schematic structural diagram of layering a stereo audio signal according to an embodiment of the present invention; FIG. 12 is a schematic structural diagram of layering a stereo audio signal according to an embodiment of the present invention; FIG. A schematic diagram of an audio layered multilayer structure;

FIG. 14 is a schematic diagram of another audio layered multi-layer structure according to an embodiment of the present invention; FIG.

FIG. 15 is a schematic diagram of an audio layered structure according to an embodiment of the present invention; FIG.

16 is a simplified schematic diagram of a dra algorithm according to an embodiment of the present invention;

FIG. 17 is a schematic diagram of a DRA kernel residual coding algorithm according to an embodiment of the present invention; FIG.

FIG. 18 is a schematic diagram of a layered structure of stereo audio according to an embodiment of the present invention. detailed description

The technical contents, structural features, objects and effects achieved by the present invention will be described in detail below with reference to the embodiments.

Please refer to the schematic diagram of layering mono or stereo shown in FIG. 1 and the coding flow diagram shown in FIG. 2. The method for encoding mono or stereo in this embodiment includes:

Step S1, dividing the mono or stereo audio signal into a basic layer and at least one enhancement layer; Step S2, encoding the basic layer by using mp3, AAC, SBR, PS, and/or DRA coding modes; Step S3: encoding, by using at least one enhancement layer, mp3, AAC, SBR, PS, DRA, residual coding, partial parameter coding algorithm, and/or parameter coding algorithm. Based on the above embodiments, the present invention provides a series of different layering schemes.

Referring to the schematic diagram of layering the audio signal according to the frequency band-based hierarchical structure shown in FIG. 3, the present invention divides the mono or stereo audio signal into a basic layer and an enhancement layer based on the frequency band, sequentially from low frequency to high frequency. The audio coding information of each frequency band is placed in the base layer and the enhancement layer. The base layer is the low frequency encoding portion of the mono or stereo sound; the enhancement layer is the mono or stereo high frequency encoding portion.

Under this layered scheme, the high frequency partial coding can participate in the same algorithm as the low frequency part, or use a parameter method such as a bandwidth extension algorithm. The basic layer generally adopts normal coding algorithms such as mp3, AAC or DRA, etc. The enhancement layer can still use normal coding algorithms, partial parameter coding algorithms such as intensity stereo, parameter coding algorithms such as bandwidth extension. The advantage of the band stratification scheme is to guarantee the quality of the low frequencies. Referring to the channel-based hierarchical structure shown in FIG. 4, the audio signal is layered. The present invention divides the stereo audio signal into a basic layer and an enhancement layer based on the channel, and the base layer transmits the left channel or the harmony. The enhancement layer transmits the right channel or the difference channel.

Under this layered scheme, the bandwidth extension algorithm can be selected for any single channel, such as the left channel or the channel, to improve subjective sound quality at low bit rates and to ensure a broadband quality. Referring to the schematic diagram of layered structure based on parametric stereo coding shown in FIG. 5, the present invention divides a stereo audio signal into a basic layer and an enhancement layer based on parametric stereo coding, and the base layer transmits left and right channels. Mixed single channel; enhancement layer transmits parametric stereo information.

Referring to the hierarchical structure diagram shown in FIG. 6, each layer is coded under the layering scheme, and the low-band portion of the base layer may select a single channel after the left-right channel downmixing using the bandwidth extension algorithm; the enhancement layer transmission The parameter is stereo information, and the high frequency portion of the downmix channel encoded by the transmission bandwidth extension algorithm can also be selected. The layering scheme and the coding scheme can achieve higher quality at a low bit rate. Referring to the residual-based hierarchical structure shown in FIG. 7, the audio signal is layered. The present invention divides the mono or stereo audio signal into a basic layer and an enhancement layer based on the residual differential layer structure.

Under the hierarchical structure, the steps of encoding the base layer and the enhancement layer include:

Step S21: Encoding according to a code rate requirement of the base layer, and putting the obtained coded data into a basic layer for transmission;

Step S22: Compare the original audio with the restored audio of the base layer decoding to obtain a residual signal. Step S3: The step of encoding the enhancement layer is to encode the residual signal as an enhancement layer.

Specifically, when encoding mono or stereo audio, the normal encoding is first performed according to the code rate requirement of the first layer, and the encoded data is transmitted in the base layer; then the original audio and the base layer are decoded and restored. The audio comparison acquires the residual signal (either in the time domain or in the transform domain) and continues to encode the residual signal as an enhancement layer.

Further, the audio signal can be layered in a plurality of hierarchical structures. For example, the base layer shown in FIG. 8 has a two-layer structure diagram based on the residual difference layer when the bandwidth layer has a bandwidth extension algorithm; FIG. 9 is a schematic diagram of a two-layer structure based on the residual difference layer when the enhancement layer has a bandwidth extension algorithm; The basic layer has a two-layer structure diagram based on the residual differential layer with bandwidth extension and enhancement layer bandwidth extension correction. In the structure shown in FIG. 10, the base layer low frequency coding portion is supplemented according to the enhancement layer low frequency residual, a more accurate low frequency portion is obtained, and the base layer bandwidth extension parameter is adjusted by the enhancement layer bandwidth extension correction parameter to better Restore the high frequency portion of each channel. Referring also to the layering of the stereo audio signal shown in Fig. 11, the base layer includes the channel low frequency partial coding of the downmix and the bandwidth extension and parametric stereo coding information, and the enhancement layer transmits the residual coding of the low frequency portion. Referring to another layered stereo audio signal shown in FIG. 12, the base layer transmits the low frequency partial coding information of the downmixed mono signal, and the enhancement layer transmits the low frequency partial residual coding information and the bandwidth extension and the parameter stereo coding. information.

With the residual differential layer structure of the embodiment, the layered structure of the audio signal is simple, and the coding efficiency is improved. The present invention also proposes that, in addition to a two-layer structure of a base layer and a reinforcement layer, the audio signal can be divided into a multilayer structure of a base layer and a plurality of enhancement layers.

Referring to FIG. 13, a schematic diagram of an audio layered multi-layer structure, which divides a mono or stereo audio signal into a base layer, a first enhancement layer, and a second enhancement layer, wherein the base layer is mono or stereo. The low frequency encoding portion; the first enhancement layer is a mono or stereo intermediate frequency encoding portion; and the second enhancement layer is a mono or stereo high frequency encoding portion.

Referring to another audio layered multi-layer structure diagram of FIG. 14, the present invention may further divide a mono or stereo audio signal into a base layer and at least one enhancement layer based on the residual differential layer structure.

In the multi-layer structure, the step S2 of encoding the base layer includes:

Step S21: Encoding according to the code rate requirement of the base layer, and putting the obtained full-band basic quality coded data into the base layer transmission;

Step S22: Compare the original audio with the audio restored by the base layer decoding to obtain a first-level residual signal.

The step S3 of encoding the first enhancement layer and/or the second enhancement layer includes:

Step S31, encoding the first-level residual signal as the data of the first enhancement layer; Step S32, removing the signal decoded and restored by the first enhancement layer from the input first-stage residual signal of the first enhancement layer coding, Obtaining a second level residual signal;

Step S33, encoding the second-level residual signal as the data of the second enhancement layer; Step S34, sequentially obtaining the next-level residual signal according to the residual signal of the previous stage, and encoding the residual signal of the next stage as The data of the next level of enhancement layer is encoded until all enhancement layers are completed.

The present invention can implement Layer 2, Layer 3 or Layer 4 and above layering and encoding for audio signals, generally no more than four layers to simplify the layering and encoding process. A specific example of the present invention is given here. Referring to FIG. 15, a schematic diagram of an audio hierarchy is shown, wherein the DRA core coding module is a standard algorithm for implementing DRA according to the standard GB/T 22726-2008. In the present invention, mono and stereo DRA coding is specifically referred to. The simple diagram of the dra algorithm is shown in Figure 16. Shown. In order to clearly describe this patent, the decoding end is also briefly described, wherein the decoding end module is shown in the dashed block diagram of FIG.

The steps of implementing coding of the base layer in this embodiment are as follows:

Step S211, at the encoding end, performing MDCT transformation on the time domain data x[n] to obtain a spectral coefficient X[k] _{; in} step S212, dividing the frequency domain coefficient into a plurality of subbands, and dividing the spectral coefficient belonging to the subband b by one Quantization step size;

Step S213, obtaining a quantized spectral coefficient for rounding (Nint) [W

Step S214, each quantization step size and spectral coefficient ^X [W are transmitted to the decoding end by various means ₍ the steps of decoding the base layer at the decoding end are:

Step S4, using the quantization step size and the spectral coefficient W transmitted in step S214 to restore the inverse quantized spectral coefficient f[W

X[k] = A _b - X[k] Step S51, the inverse quantized spectral coefficient ^fc 3⁄4 _{IMDCT is} obtained by inversely quantized time domain data. The above SBR coding module is in accordance with the standard "ISO/IEC 14496-3:2001/Amd.l:2003,

Bandwidth Extension is implemented. In this patent, placing SBR in the base layer can achieve higher quality at a lower code rate. Since the implementation of SBR is not related to this patent, and the SBR coding module is optional, this patent The present invention further provides an example of separately encoding at least one enhancement layer based on the above coding of the base layer. The DRA core residual coding module used in this embodiment is an intermediate module as shown in FIG. 16. The schematic diagram of the DRA kernel residual coding algorithm shown in FIG. 17 shows that the base layer and the coding end of FIG. 18 are completely identical, that is, fully compatible. The implementation of the base layer is as above. The implementation steps of at least one enhancement layer coding in this embodiment as follows:

The following steps of adding the following enhancement layer in the base layer step 3 include:

Step S311, at the encoding end, performing MDCT transform on the time domain data x[n] to obtain a spectral coefficient X[k] _{; in} step S312, dividing the frequency domain coefficient into a plurality of subbands, and dividing the spectral coefficient belonging to the subband b by one Water - : step size; step S313, rounding the quantized length (nint) to obtain the quantized spectral coefficient W

Step S314, transmitting each quantization step and spectral coefficient ^] to the decoding end; Step S315, restoring the inverse quantized spectral coefficient WX[k] = A _b -X[k] with the quantization length and the spectral coefficient ^] Step S316, using the original spectral coefficient ^X [W minus the inverse quantized spectral coefficient], to obtain the residual spectral coefficient E[W

E[k] = X[k]-X[k], step S317, dividing the residual spectral coefficient into a plurality of sub-bands, dividing the spectral coefficient belonging to the sub-band c by a residual spectral coefficient quantization step, and rounding (nint) the quantized residual spectral coefficient

Step S318, transmitting the residual spectral coefficient quantization step size ^ and the quantized residual spectral coefficient to the solution

The process of decoding the at least one enhancement layer at the decoding end is as follows:

Step S41, using the quantization step size and spectral coefficient passed in step S214] to recover the inverse quantized spectral coefficient f[W X[k] = A _b - X[k] , step S42, using the residual spectral coefficient quantization step size and the quantized residual spectral coefficient passed in step S34 to restore the inverse quantized residual spectral coefficient

E[k] = A _e - E[k] . Step S43, adding the inverse quantized spectral coefficient obtained in step S41 and the inverse quantized residual spectral coefficient obtained in step S42 to obtain an enhanced inverse quantized spectral coefficient ^[ ^ ^]

X _a [k] = X[k] - E[k] , step S52, inversely quantized spectral coefficient ^f for enhancement. ^W does IMDCT to get inverse quantized time domain data x[n]

The present invention further proposes that the total coding rate is 48 kbps, and the audio signal is divided into two layers by a residual differential layer structure, and each layer is 24 kbps as an example to describe the implementation steps of separately coding the base layer and the at least one enhancement layer in this embodiment.

Step S201: Encoding the base layer with a coding rate of 24 kbps at a coding bandwidth of 48 kbps, and obtaining a quantization step size of the 24 kbps code rate and a quantized spectral coefficient and an sbr code stream;

Step S301, multiplying the quantized spectral coefficients by the quantized step size at the encoding end to obtain an inverse quantized spectral coefficient at a coding rate of 24 kbps.

Step S302, subtracting the inverse quantized spectral coefficient ^f W from the original spectral coefficient ^x W to obtain a residual signal spectral coefficient E[k]. Step S303, using a 24 kbps code rate for the residual signal spectral coefficient ^£ [W, quantization, quantization method Consistent or similar to the quantization, the quantized step size ^Δ Ρ quantized residual spectral coefficients of the quantized residual signal are obtained and transmitted to the decoding end. The invention also proposes that if only stereo coding is performed, except for the above implementation, The encoding of the base layer and the at least one enhancement layer can be implemented with the next embodiment. An advantage of this embodiment over the previous embodiment is that higher quality can be obtained when the stereo total code rate is low.

As shown in FIG. 18, a stereo audio layering structure diagram, in this embodiment, the two stereo channels are downmixed into one channel and encoded by PS, wherein the PS code is in accordance with the standard ISO/IEC 14496-3:2001/Amd. 2:2004: "Parametric Coding for High Quality Audio" is implemented. The DRA downmix channel coding is the same as the base layer coding principle and the procedure in FIG. 16; and the coding principle of the enhancement layer in this embodiment is the same as the DRA downmix channel residual coding, and therefore will not be described again.

The method for encoding mono or stereo according to the present invention achieves the above objects and effects by the method disclosed above, but the above disclosure is only a preferred embodiment of the present invention, and the right to the present invention cannot be limited thereby. Scope, other equivalent modifications or variations of the invention are intended to be included within the scope of the appended claims.

Claims

claims

1. A method for encoding mono or stereo, characterized in that it includes: dividing the mono or stereo audio signal into a basic layer and at least one enhancement layer;

The base layer is encoded using mp3, AAC, SBR, PS and/or DRA coding methods; the at least one enhancement layer is encoded using mp3, AAC, SBR, PS, DRA, residual coding, partial parameter coding algorithm and /or parameter encoding algorithm encoding.

2. The method for encoding mono or stereo according to claim 1, characterized in that dividing the mono or stereo audio signal into a basic layer and an enhancement layer is:

The mono or stereo audio signal is divided into a basic layer and an enhancement layer based on the frequency band. The basic layer is the low-frequency encoding part of the mono or stereo; the enhancement layer is the high-frequency encoding part of the mono or stereo. ; or

Divide the stereo audio signal into a base layer and an enhancement layer based on the channel, the base layer transmits the left channel or the harmony channel; the enhancement layer transmits the right channel or the difference channel; or

The stereo audio signal is divided into a basic layer and an enhancement layer based on parametric stereo coding. The basic layer transmits a single channel of left and right channel downmix; the enhancement layer transmits parametric stereo information; or

Based on the residual layered structure, the mono or stereo audio signal is divided into a base layer and an enhancement layer.

3. The method for encoding mono or stereo according to claim 2, characterized in that the base layer and/or at least one enhancement layer are encoded using a bandwidth extension algorithm respectively.

4. The method for encoding mono or stereo according to claim 2, characterized in that the step of separately encoding the basic layer and an enhancement layer obtained by dividing the residual layered structure includes:

Supplement the low-frequency coding part of the base layer based on the low-frequency residual of the enhancement layer;

The base layer bandwidth expansion parameter is adjusted through the enhancement layer bandwidth expansion correction parameter.

5. The method for encoding mono or stereo according to claim 2, characterized in that:

When the audio signal is stereo, the basic layer includes coding of the low-frequency part of the downmixed channel, bandwidth extension and parametric stereo coding information;

The enhancement layer transmits the residual coding of the low-frequency part.

6. The method for encoding mono or stereo according to claim 2, characterized in that:

When the audio signal is stereo, the base layer transmits the low-frequency part encoding information of the downmixed mono signal;

The enhancement layer transmits low-frequency part residual coding information and bandwidth extension and parametric stereo coding information.

7. The method for encoding mono or stereo according to claim 2, characterized in that the step of encoding the base layer includes:

Encode according to the code rate requirements of the base layer, and put the resulting encoded data into the base layer for transmission; compare the original audio with the audio after decoding and recovery of the base layer to obtain the residual signal; and

The step of encoding the enhancement layer is encoding the residual signal as an enhancement layer.

8. The method for encoding mono or stereo according to claim 1, characterized in that: dividing the mono or stereo audio signal into a basic layer, a first enhancement layer and a second enhancement layer is :

The mono or stereo audio signal is divided into a base layer, a first enhancement layer and a second enhancement layer based on the frequency band, where the base layer is the low-frequency coding part of the mono or stereo; the first enhancement layer is the mono or a stereo mid-frequency coding part; the second enhancement layer is a mono or stereo high-frequency coding part.

9. The method for encoding mono or stereo according to claim 1, characterized in that the mono or stereo audio signal is divided into a basic layer and at least an incremental layer based on the residual layered structure. strong layer; while

The steps of encoding the base layer include:

Encode according to the code rate requirements of the base layer, and put the obtained full-band basic quality encoded data into the base layer for transmission;

Compare the original audio with the audio recovered after base layer decoding to obtain the first-level residual signal; and the step of encoding the first enhancement layer and/or the second enhancement layer includes:

Encoding the first-level residual signal as data of the first enhancement layer;

Remove the signal recovered by decoding the first enhancement layer from the first-level residual signal input by the first enhancement layer encoding to obtain the second-level residual signal;

Encode the second-level residual signal as the data of the second enhancement layer;

The next-level residual signal is obtained in turn based on the upper-level residual signal, and the next-level residual signal is encoded as the data of the next-level enhancement layer until all enhancement layers are encoded.

10. The method for encoding mono or stereo according to any one of claims 1 to 9, characterized in that the step of encoding the base layer includes:

At the encoding end, perform MDCT transformation on the time domain data x[n] to obtain the spectral coefficients X[k] _;

Divide the frequency domain coefficients into multiple sub-bands, and divide the spectral coefficients belonging to sub-band b by a quantization step;

Round the quantization step size (nint) to obtain the quantized spectral coefficient ^f

Each quantization step size and spectral coefficient W are transmitted to the decoding end.

11. The method for encoding mono or stereo according to claim 10, characterized in that the step of encoding the at least one enhancement layer separately includes:

Divide the frequency domain coefficients into multiple sub-bands, and divide the spectral coefficients belonging to sub-band b by a quantization step size Δ Round the quantization step size (nint) to obtain the quantized spectral coefficient W

Each quantization step size and spectral coefficient are transmitted to the decoding end;

Use the quantization step size and spectral coefficient to restore the inverse quantized spectral coefficient ^f

X[k] = A _b -X[k], subtract the inverse quantized spectral coefficient ^ ^] from the original spectral coefficient ^XW to obtain the residual spectral coefficient ^£W E[k] = X[k]-X[k] . Divide the residual spectral coefficient ^£ [W into multiple sub-bands, and divide the spectral coefficient belonging to sub-band c by a residual

;Page quantification step size ^Δ «, rounded (nint) to obtain the quantized residual spectrum coefficient