JP3237178B2 - Encoding method and decoding method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an encoding method for dividing an input audio signal such as an audio signal and an audio signal into frames, converting the signal into data on a frequency, and performing encoding. The present invention relates to a decoding method for decoding a code string generated by a decoding method.

[0002]

2. Description of the Related Art There are known various encoding methods for compressing an audio signal (including a voice signal and an acoustic signal) by utilizing a statistical property in a time domain and a frequency domain and a characteristic of human perception. ing. As this encoding method,
Coding in the time domain, coding in the frequency domain,
Analysis synthesis coding.

[0003] As an example of high-efficiency encoding of a speech signal or the like, M
BE (Multiband Excitation) coding, SBE (Singleband Excitation) coding, Harmonic coding, SBC
(Sub-band Coding: band division coding), LPC (Linear
Predictive Coding: Linear predictive coding) or DC
T (Discrete Cosine Transform), MDCT (Modified DC
Conventionally, when quantizing various information data such as spectrum amplitude and its parameters (LSP parameter, α parameter, k parameter, etc.) in T), FFT (Fast Fourier Transform), scalar quantization is conventionally performed. There are many.

[0004]

By the way, if the bit rate is reduced to, for example, about 3 to 4 kbps and the quantization efficiency is further improved, the quantization noise (distortion) increases in the scalar quantization. It was difficult to put it to practical use.

Therefore, it is conceivable to employ vector quantization. However, assuming that the number of bits of the output (index) of vector quantization is ^b , the size (size) of the codebook of the vector quantizer increases in proportion to 2 ^b , and the amount of calculation for codebook search also it increases in proportion to 2 ^b. However, if the number of bits b in the output is too small, quantization noise will increase. Therefore, it is possible to reduce the codebook size and the amount of calculation at the time of search while keeping the number of bits b to some extent. desired. Further, if the data converted on the frequency axis is vector-quantized as it is, the coding efficiency may not be sufficiently increased, so that a device for further increasing the compression ratio is required.

The present invention has been made in view of such circumstances, and reduces the size of a codebook of a vector quantizer and the amount of calculation at the time of search without reducing the number of output bits of vector quantization. It is another object of the present invention to provide a coding method capable of further increasing the compression ratio at the time of vector quantization and a decoding method for decoding a code string generated by the coding method.

[0007]

According to the present invention, there is provided an encoding method comprising the steps of dividing an input audio signal into blocks and converting the signal into a frequency axis, and converting the input audio signal into an M-dimensional vector on a frequency axis. A step of obtaining data, a step of dividing the data on the frequency axis of the M-dimensional vector into a plurality of groups, and obtaining a representative value for each group to reduce the dimension to S dimension (S <M); Performing a first vector quantization on the vector data; dequantizing the first vector quantization output data to obtain a corresponding S-dimensional code vector;
Extending the S-dimensional code vector to an original M-dimensional vector; and performing data processing on the data representing the relationship between the expanded M-dimensional vector and the original data on the frequency axis of the M-dimensional vector. And a step of performing vector quantization of 2.

Here, as the data on the frequency axis of the M-dimensional vector, data obtained by nonlinearly compressing the data converted to the frequency axis in the block unit can be used.

[0009]

The difference between blocks of the data to be vector-quantized may be obtained, and the inter-block difference data may be subjected to vector quantization.

According to another encoding method of the present invention, an amplitude of a spectrum obtained by dividing an input audio signal in units of blocks and converting the divided signals into a frequency axis is converted into a dB area and converted into an M signal.
A step of obtaining data on the frequency axis as a dimensional vector, and dividing the data on the frequency axis of the M-dimensional vector into a plurality of groups and obtaining an average value for each group to reduce the dimension to S-dimension (S <M) And the S
Performing a first vector quantization on the average value data of the dimensional vector; dequantizing the first vector quantization output data to obtain a corresponding S-dimensional code vector; Extending the code vector to the original M-dimensional vector, and performing second vector quantization on the difference data between the expanded M-dimensional vector and the original data on the frequency axis of the M-dimensional vector. The above-mentioned problem can be solved.

Also, in the decoding method according to the present invention, data on the frequency axis is obtained by using an audio signal as an M-dimensional vector, and the dimension is reduced from M to S (S <M).
The first vector quantization is performed on the data of the dimensional vector to encode an index, and the first vector quantized output data is inversely quantized to obtain a corresponding S-dimensional code vector. The vector is expanded to the original M-dimensional vector, and the second vector quantization is performed on the data representing the relationship between the expanded M-dimensional vector and the original data on the frequency axis of the M-dimensional vector. And a decoding method for receiving and decoding a code string generated by encoding the second vector quantization, wherein the first inverse vector quantization for inversely quantizing the index of the second vector quantization to obtain a corresponding code vector. Process and
A second inverse vector quantization step of inversely quantizing the index of the first vector quantization to obtain a corresponding S-dimensional code vector;
Extending to a vector of dimensions, obtaining data on the frequency axis of an M-dimensional vector from the code vector obtained in the first inverse vector quantization step and the expanded M-dimensional vector, And restoring the audio signal based on the data on the frequency axis of the dimensional vector.

[0013]

According to the encoding method of the present invention, vector quantization having a hierarchically structured codebook is performed such that an M-dimensional vector is reduced to an S-dimensional (S <M) vector and vector-quantized. Is performed, the amount of calculation of the codebook search and the codebook size can be significantly reduced, and the effective application of the error correction code becomes possible. Further, the quality of the quantization can be improved by nonlinearly compressing the data on the frequency axis and performing vector quantization, and the compression efficiency can be further improved by calculating the difference between the blocks. Further, according to the decoding method of the present invention, it is possible to decode the code string generated by the encoding method.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the encoding method according to the present invention will be described below with reference to the drawings. FIG. 1 shows a schematic configuration of an encoding device (encoder) for explaining high-efficiency encoding according to one embodiment of the present invention.

In FIG. 1, a voice signal or an acoustic signal is supplied to an input terminal 11, and this input signal is converted by a frequency axis conversion processing section 12 into spectrum amplitude data on the frequency axis. Inside the frequency axis conversion processing unit 12, for example, a blocking unit 1 for blocking every predetermined number of samples (N samples) of the input signal on the time axis
2a, orthogonal transform unit 12 such as FFT (fast Fourier transform)
b, a data processing unit 12c for obtaining amplitude information representing characteristics of the spectrum envelope is provided. An output from the frequency axis conversion processing unit 12 is sent to a vector quantization unit 15 via a non-linear compression unit 13 which converts the data into, for example, a dB region as needed, and a processing unit 14 which takes a difference between blocks as needed. Sent. This vector quantization unit 1
In step 5, a predetermined number (M samples) of input data is put together to form an M-dimensional vector, and a vector quantization process is performed. In such an M-dimensional vector quantization process, generally, a code vector having the closest distance in an M-dimensional space to an input M-dimensional vector is searched from a codebook, and the search is performed. This is a process of extracting the index of the code vector obtained from the output terminal 16. The vector quantization unit 15 of the embodiment shown in FIG. Is applied.

That is, the vector quantizer 1 shown in FIG.
5, the data of the M-dimensional vector (data on the frequency axis), which is a unit of vector quantization, is
1, the data is divided into a plurality of groups, and a representative value is obtained for each group.
<M). Here, FIG. 2 shows a specific example of each element of the M-dimensional vector X input to the vector quantizer 15, that is, M pieces of amplitude data x (n) on the frequency axis, where 1 ≦ n ≦ M. These M pieces of amplitude data x (n) are put together, for example, every four samples to obtain a representative value, for example, an average value y _i , and as shown in FIG. 3, the average value data y _{1 to} y _S (In this case, S = M / 4) is obtained as an S-dimensional vector Y.

Next, the S-dimensional vector data is subjected to vector quantization by an S-dimensional vector quantizer 22. That is, from among the S-dimensional code vectors in the code book of the S-dimensional vector quantizer 22, the code vector closest to the input S-dimensional vector in the S-dimensional space is searched. The index data is extracted from the output terminal 26, and the searched code vector (a code vector obtained by inverse vector quantization of the output index) is sent to the dimension extension unit 23. FIG. 4 is a diagram illustrating an example in which the S-dimensional vector Y including the _S average value data y _{1 to} y S shown in FIG. shows each element y _VQ1 ~y _VQS the S-dimensional vector Y _VQ as taken out) resulting local decoder outputs the searched code vector during reduction.

Next, the dimension extension unit 23 extends the S-dimensional code vector to the original M-dimensional vector. FIG. 5 shows an example of each element of the expanded M-dimensional vector.
As apparent from FIG. 5, by increasing each element y _VQ1 ~y _VQS of the inverse vector quantized S-dimensional vector Y _VQ in each respective original 4 samples, 4
An M-dimensional vector consisting of S = M elements is obtained. The second vector quantization is performed on the data representing the relationship between the expanded M-dimensional vector and the original data on the frequency axis of the M-dimensional vector.

In the embodiment shown in FIG. 1, the extended M-dimensional vector data from the dimension extension unit 23 is sent to a subtractor 24,
By subtracting the original M-dimensional vector from the data on the frequency axis, S vector data representing the relationship between the S-dimensionally expanded M-dimensional vector and the original M-dimensional vector is obtained. FIG. 6 shows M pieces of amplitude data x on the frequency axis which are each element of the M-dimensional vector X shown in FIG.
from (n), shows the M data r ₁ ~r _M obtained by subtracting each element of the extended M-dimensional vector shown in FIG. 5,
The four samples of the _M data r _{1 to} r M are set as a set (vector) and S four-dimensional vectors R _{1 to} R _M are set.
_S is obtained.

The S obtained from the subtractor 24 in this manner
Each of the vectors is subjected to vector quantization by the S vector quantizers 25 _{1 to} 25 _S of the vector quantizer group 25. Each vector quantizer 2
Index output from 5 ₁ to 25 _S are respectively taken from the output terminal 27 ₁ ~ 27 _S. FIG. 7 shows a four-dimensional vector quantizer 25 _{1 to} 25 _S using four-dimensional vector quantizers, and the four-dimensional vectors R _{1 to} R _S shown in FIG. R _VQ1 ~R each element of the _VQS r _VQ1 ~r
_VQ4, r _VQ5 ~r _VQ8, shows the ... ~r _VQM.

By performing such hierarchically structured two-stage vector quantization, the amount of calculation for a codebook search can be reduced, and the amount of memory (eg, ROM capacity) for a codebook can be reduced. The output terminal 2
For example, by applying error correction coding to the more important index of the upper layer obtained from No. 6 to protect the index more importantly, it is possible to effectively apply the error correction code. Note that the hierarchical structure of the vector quantization unit 15 is not limited to two levels, and may have a multilayer structure of three or more levels.

It is not necessary that all the components shown in FIG. 1 be constituted by hardware, but may be realized by software using, for example, a so-called DSP (digital signal processor). Also, the vector quantization unit 15
The adder 28 in the first and second vector quantizers 2
To obtain M pieces of quantized data by adding each element of the data quantized by 2, 25, respectively,
Each of the M pieces of data expanded to M dimensions from the dimension expansion unit 23 and the M pieces of element data of each of the S code vectors from the vector quantizers 25 _{1 to} 25 _S are added, M pieces of data are extracted from the output terminal 29. The adder 28 is used for calculating a difference between blocks (or a difference between frames) to be described later, and may be omitted when such a difference between blocks is not calculated.

Next, FIG. 8 shows a schematic configuration of an encoding device (encoder) for explaining a high-efficiency encoding method as a second embodiment of the present invention. In FIG. 8, an audio signal such as an audio signal or an acoustic signal supplied to an input terminal 11 is divided by an N-sample block unit in a frequency axis conversion processing unit 12 and converted into a frequency axis. To the non-linear compression unit 13 to perform non-linear compression such as conversion into dB area data, for example.
The obtained M pieces of non-linear compressed data are combined into an M-dimensional vector, subjected to vector quantization by the vector quantization unit 15, and extracted from the output terminal 16. The vector quantization unit 15 may have a two-stage hierarchical structure as shown in FIG. 1 above, may have a three-stage hierarchical structure, or may have a normal hierarchical structure without a hierarchical structure. A (one-stage) vector quantization may be performed. The non-linear compression unit 13 may use, for example, pseudo logarithmic compression such as so-called μ-law or A-law in addition to log compression (logarithmic compression) for converting the data into dB area data. Thus, efficient encoding is realized by logarithmically converting the amplitude, compressing the result, and linearly encoding the result.

Next, FIG. 9 shows a schematic configuration of an encoding device (encoder) for explaining a high-efficiency encoding method as a third embodiment of the present invention. In FIG. 9, the audio signal supplied to the input terminal is divided into blocks by the frequency axis conversion processing unit 12 and converted into the frequency axis, and the obtained data is processed by the nonlinear compression unit 13 as necessary. Processing unit 14 for obtaining a difference between blocks
Sent to. Note that the block of N samples has an overlapping portion with an adjacent block, and L blocks in the time axis direction.
In the case of a format in which the processing proceeds in units of samples (L <N), the processing unit 14 calculates the difference between frames. Between this block (or between frames)
The M data obtained by taking the difference are sent to the M-dimensional vector quantization unit 15, and the quantized index data is extracted from the output terminal 16. This vector quantization unit 15
May or may not have a multilayer structure.

The processing section 14 for calculating the difference between the blocks (or between frames) may delay the input data by one block (or one frame) to obtain the difference from the original input data. In the example of No. 9, the subtractor 31 is inserted and connected to the input side of the vector quantization unit 15, and the code vector (consisting of M element data) from the M-dimensional vector quantization unit 15 is stored in one block (or one frame). The delayed data is subtracted from the input data (M-dimensional vector). In this case, since it is difference data of the data to be vector-quantized, the code vector from the vector quantization unit 15 is sent to the adder 32, and the adder 32
Block (or frame) delay circuit 3
The value obtained by multiplying the coefficient α by the multiplier 34 through 3 is sent to the adder 32. The output from the multiplier 34 is sent to the subtractor 31. When the M-dimensional vector quantization unit 15 having the two-stage hierarchical structure shown in FIG. 1 is used, data from the output terminal 29 is sent to the adder 32 as an M-dimensional code vector for vector quantization. Will be.

As described above, by taking the difference between blocks (frames), the region where the input amplitude data on the frequency axis exists in the M-dimensional space can be narrowed. This is because, during the block (frame) interval, the amount of change in the spectrum amplitude is usually small and has a strong correlation. Therefore, quantization noise can be reduced, and data compression efficiency can be further increased.

Next, the data on the frequency axis obtained by the frequency axis conversion processing unit 12 is converted into a spectrum in which the amplitude of the spectrum is converted into a dB area by the non-linear compression unit 13. After the difference is obtained, a specific example in which the M-dimensional vector quantization is performed by the vector quantization unit 15 having a multilayer structure as shown in FIG. 1 will be described in detail below. The frequency axis conversion processing unit 1
In No. 2, processing of various coding methods can be adopted. For example, MBE (Multiband Excitation) processing described later can be used. At the time of block formation in the frequency axis conversion processing unit 12, it is assumed that a block of N samples is advanced on a time axis in frame units of L samples. At this time, the analysis is performed on one block N samples, but the analysis result is obtained (updated) in one frame L sample period.

First, the value (amplitude value) of the data such as the spectrum amplitude of the MBE analysis result obtained from the frequency axis conversion processing unit 12 is defined as a (m), and 0 (zero) is set for each frame.
It is assumed that m _MX +1 samples of ≦ m ≦ m _MX are obtained.

The amplitude value a (m) of this m _MX +1 sample
Is converted to the dB region, and the obtained data is set to a _dB (m), where a _dB (m) = 20 log ₁₀ a (m) (1). Here, in the MBE analysis or the like, the number m _MX +1 of the samples changes for each frame depending on the pitch period. Therefore, for the inter-frame difference and the vector quantization, the dB existing in one frame (block) is used.
It is desirable to keep the number of amplitude values a _dB (m) constant. Therefore, by sampling rate conversion (data number conversion)
m _MX +1 The dB amplitude value a _dB (m) is converted into a fixed number M of data b _dB (n). This sample number n is 1
It is assumed that each frame (block) has a value of 1 ≦ n ≦ M, and the data b _dB (0) of n = 0 corresponding to the dB amplitude value a _dB (0) of m = 0 is a DC (direct current) component. Since it has a corresponding amplitude, it is not transmitted (always 0).

As described above, by taking the inter-frame difference after the conversion into the dB area, the area where the data b _dB (n) is present can be further narrowed. That is, the amplitude of the spectrum rarely changes greatly during a frame interval (for example, about 20 msec), and has a very strong correlation. That is, vector quantization is performed on a value c _dB (n) obtained by taking a difference such as c _dB (n) = b _dB (n) −b ′ _dB (n) (2). Here, b ′ _dB (n) is a predicted value of b _dB (n), and as shown in FIG. 9, an inverse quantization output from the vector quantization unit 15 (local decoder output, code "output b _dB (n) is obtained by one-frame delay in the delay circuit 33" _dB (n) p equivalent) b vector and alpha multiplied by the multiplier _{34, b 'dB (n)} = α · b "dB ( n) p (3) Here, p of b ″ _dB (n) p represents one frame before.

As described above, when the difference between the amplitudes of the frames is calculated, the quantization noise can be further reduced, but it is susceptible to a code error. This is because an error in a certain frame propagates to the next frame, the next frame, and so on. Therefore, in general, α is set to about 0.7 to 0.8, and a so-called leaky difference is calculated. If you still want to be strong against code errors, decrease α,
Further, the process may be shifted to the next processing step with α = 0 (that is, without taking the difference between frames). This needs to be determined in consideration of the balance of the performance of the entire system.

Next, the inter-frame difference data c _dB (n)
, I.e., the array c _dB (n) is transformed into an M element having M elements.
An example in which vector quantization is performed as a dimensional vector will be described. Even if the difference is not taken, if α = 0 is considered, c
can be included in _dB (n). Such M data to be quantized in the M-dimensional vector is replaced with x (n). In the case of the present embodiment, x (n) ≡c _dB (n), and 1 ≦ n ≦ M. When the number of bits the index of such an M-dimensional vector quantization output is b, straight vector quantization directly searching a codebook having a 2 ^b number of code vectors of the M-dimensional also logically although, the codebook search during vector quantization, in proportion to M · 2 ^b calculates the amount of increase is the same for ROM size of the table. Thus, the use of vector quantization with a structured codebook becomes a more realistic method. In this embodiment, the M-dimensional vector is divided into a plurality of low-dimensional vectors, an average value of each vector is calculated, and a vector (upper layer) including the average value and a vector group (lower layer) from which the average value is removed are set. Layers) and performs vector quantization.

First, M such as the difference data c _dB (n)
Data x (n) is divided into S vectors.

[0034]

(Equation 1)

[0035] in this equation _{(4), X 1, X 2, ...} , X
_S each d _1, d 2 _{is ...,} represent d _{S-dimensional} vector,
d ₁ + d ₂ +... + d _S = M. Also, t indicates transposition of a vector. The above-described specific example of FIG.
_1, X _2, ... _, X _S are all 4 dimensions (d ₁ =
d ₂ =... = d _s = 4).

The S vectors X _1, X _2, ... _, X
_Assuming that the average value of each element of _S is y _1, y _2, ... _, Y _S , y _i (1 ≦ i ≦ S) becomes

[0037]

(Equation 2)

Can be expressed as follows. S whose average value is an element
The dimensional average vector Y is defined as Y = (y ₁ , y ₂ ,..., Y _s ) ^t (6). This corresponds to FIG. First, this S
Vector quantization of the dimension vector Y is performed. Vector y
For the vector quantization of (1), methods such as straight vector quantization and shape-gain vector quantization can be considered, but in this embodiment, shape-gain vector quantization is used. This shape-gain vector quantization is MJ
Sabin, RM Gray, "Product Code Vector Quantizer
for Waveformand Voice Coding ", IEEE Trans. on ASS
P, vol. ASSP-32, No. 3, June 1984.

The result of the vector quantization of the S-dimensional vector Y (S-dimensional vector) is defined as Y _VQ . The S-dimensional vector Y _VQ after the _{quantization, Y VQ = (y VQ1,} y VQ2, ..., y VQS) expressed as (7). This Y _VQ is equal to the original array x (n) (≡c _dB (n),
(1 ≦ n ≦ M). Therefore, it is necessary to protect against transmission line errors relatively strongly.

Next, based on the S-dimensional vector Y _VQ , an input array x (n) (≡c _dB (n)) of the original M-dimensional vector is estimated (dimension expansion) by some method. The error between the estimated value and the original input sequence is used as an input signal for the next stage of vector quantization. Examples of the estimation method include non-linear interpolation (for example, A. Gersho, "Optimal Nonlinear InterpolativeVec
tor Quantization ", IEEE Trans. on comm., vol.38, N
o.9, Sep. 1990, etc.), spline interpolation, polynomial interpolation, linear interpolation (first-order interpolation), zero-order hold, and the like. When excellent interpolation is performed here, the area where the input vector exists in the next stage of vector quantization is narrowed, and as a result, quantization with lower distortion can be performed. In this embodiment, the simplest zero-order hold example (see FIG. 5) is employed.

[0041] Here, the average value removing vectors corresponding to the S vector (residual vector obtained by removing the quantized average value) R _1, R _{2, ...,} when expressed as R _S, these vectors R _{1 ,} R _2, … _, R _S

[0042]

(Equation 3)

Is obtained. The vector I in this equation (8)
_i (1 ≦ i ≦ S) is a unit column vector in which all the elements are 1 in the _di dimension. This specific example corresponds to FIG.

[0044] Each of these residual vector R _1, R _2, ...,
R _S is vector quantized using each separate codebook. Here, the vector quantization uses straight vector quantization, but other structured vector quantization is of course also possible. That is, each residual vector R
_1, R _2, … _{, RS} are represented by elements

[0045]

(Equation 4)

On the other hand, the data after vector quantization is
_{R VQ1, R VQ2, ...,} R VQS, generally referred to in the R _VQi.

[0047]

(Equation 5)

This can be regarded as a result of adding a quantization error vector ε _i to the residual vector R _i . That is, R _VQi = R _i + ε _i (11)

[0049]

(Equation 6)

Each of the quantized residual vectors R _VQ1,
A specific example of each element of R _{VQ2 ,} ... _, R _VQS corresponds to FIG.

The index output to be transmitted on the encoder side is one index representing the Y _VQ and the S index
Number of residual vector R _VQ1, R _{VQ2, ...,} the S number of the index representing the R _VQS. In the shape-gain vector quantization, one output index is represented by one shape index and one gain index.

Next, the following processing is performed to obtain a decoded value of vector quantization. By table lookup from the transmitted index, Y _VQ , R
_{After obtaining VQi} (1 ≦ i ≦ S), the following processing is executed. That is, y _VQi is obtained from the above equation (7), and X _VQi is obtained as follows. _{X VQi = R VQi + y VQi} I i (1 ≦ i ≦ S) = R i + ε i + y VQi I i = X i -y VQi I i + ε i + y VQi I i = X i + ε i ··· (13)

Therefore, the quantization noise appearing in the decoder output is only ε _i generated during the quantization of R _i . The quality of the first-stage Y quantization does not directly appear in the final noise. However, depending on the _quality , the performance of vector quantization of R _{VQi in} the second stage is affected, and as a result, it contributes to the magnitude of the quantization noise of the decoder output.

As described above, by making the vectorbook codebook a hierarchical structure, (i) the number of times of multiplication and addition for the codebook search is reduced, (ii) the ROM capacity for the codebook is reduced, iii) Hierarchical error correction code can be effectively used.

Specific examples of the effects (i) and (ii) will be described. Now, M = 44, S = 7, d ₁ = d ₂ = d ₃
= D ₄ = 5 and d ₅ = d ₆ = d ₇ = 8. The number of bits used for quantization of the data x (n) (≡c _dB (n)) and 1 ≦ n ≦ M is 48 bits.

First, simply, an M = 44-dimensional vector is expressed as 4
When the vector quantization of the 8-bit output is performed, the table size of the codebook becomes 2 ⁴⁸ ≒ 2.81 × 10 ¹⁴ , which is multiplied by the word width (= 44) to obtain 2 ⁴⁸ ≒ 44 ≒ 1.2.
A table of 38 × 10 ¹⁶ words is required. The calculation amount of the table search is also a value on the order of 2 ⁴⁸ × 44.

On the other hand, consider the following bit allocation. Y → 13 bits (8 bits: shape, 5 bits: gain), dimension S = 7 X ₁ → 6 bits, dimension d ₁ = 5 X ₂ → 5 bits, dimension d ₂ = 5 X ₃ → 5 bits, dimension d ₃ = 5 X ₄ → 5 bits, Dimension d ₄ = 5 X ₅ → 5 bits, Dimension d ₅ = 8 X ₆ → 5 bits, Dimension d ₆ = 8 X ₇ → 4 bits, Dimension d ₇ = 8 Total 48 bits , (M =) 44 dimensions As table capacity at this time, Y: shape: 7 × 2 ⁸ = 1792, gain: ²⁵ = 32 X ₁ : 5 × 2 ⁶ = 320 X ₂ : 5 × 2 ⁵ = 160 ^{_{X 3: 5 × 2 5 =}} 160 X 4: 5 × 2 5 = 160 X 5: 8 × 2 5 = 256 X 6: 8 × 2 5 = 256 X 7: 8 × 2 4 = 128 words, a total of 3264 words Is required. The amount of calculation in the table search is basically the same order as the total size of the table, and is therefore on the order of approximately 3264.
This is a value that poses no problem in practical use.

Regarding the above (iii), the 13 bits of the quantized output index of the vector Y in the first stage are:
FEC (Forward Error Correcti
on), for X _{1 to} X ₇ , for example, the top 3, 3, 2, 2,
2, 2, and 1 bits can be protected, and the lower bits can be used without error correction. In this case, there is a correlation between the Hamming distance of the binary data representing the index of the vector quantizer and the Euclidean distance of the code vector referred to by the index (that is, a code vector having a short Euclidean distance). Allocating an index having a short Hamming distance, etc.), so that FEC can be applied more effectively.

By the way, in the case of speech synthesis analysis coding such as MBE, which will be described later, in consideration of the fact that the voiced unvoiced degree, pitch, etc. of voice have already been extracted as feature quantities, these feature quantities, especially voiced It is possible to switch the codebook for vector quantization depending on the result of the / non-voice determination. That is, at the time of voiced sound, at the time of unvoiced sound, the shape of each existing spectrum is greatly different, and having a codebook trained separately for each state,
This is very favorable in terms of improving characteristics. In the case of hierarchical structured vector quantization, the upper layer vector quantization may be a fixed codebook, and only the lower layer vector quantization codebook may be switched between voiced and unvoiced. . In addition, the bit allocation on the frequency axis may be switched to, for example, emphasizing low sounds for voiced sounds and emphasizing high sounds for unvoiced sounds. The switching control, the presence or absence of the pitch, including voiced / unvoiced on the frequency axis
Yes ratio spectrum level and inclination, etc. can be utilized.

Next, a specific example of a kind of MBE (Multiband Excitation) vocoder which is a kind of speech signal synthesis / analysis coding apparatus (so-called vocoder) to which the high efficiency coding method as described above can be applied will be described. This will be described with reference to the drawings. The MBE vocoder described below is compatible with DW Griffin and JS Lim, "MultibandExcitatio
n Vocoder, "IEEE Trans.Acoustics, Speech, and Signal
Processing, vol. 36, No. 8, pp. 1223-1235, Aug. 1988
And the conventional PARCOR (PA
Rtial auto-CORrelation: In vocoders and the like, voiced sections and unvoiced sections are switched for each block or frame when modeling speech, whereas MBE vocoders use the same time (for the same block or The modeling is performed on the assumption that a voiced section (Voiced) section and an unvoiced section (Unvoiced) section exist in the frequency axis region (in a frame).

FIG. 10 is a block diagram showing the overall schematic configuration of an embodiment in which the present invention is applied to the MBE vocoder. In FIG. 10, an audio signal is supplied to an input terminal 101, and this input audio signal is
The signal is sent to a filter 102 such as an HPF (high-pass filter) to remove a so-called DC (direct current) offset and to remove at least a low-frequency component (200 Hz or less) for band limitation (for example, limited to 200 to 3400 Hz). . The signal obtained through the filter 102 is sent to the pitch extraction unit 103 and the windowing processing unit 104, respectively.
In the pitch extracting section 103, the input audio signal data is divided into blocks (or cut out by a rectangular window) in units of a predetermined number N (for example, N = 256), and the pitch of the audio signal in this block is extracted. .
For example, as shown in FIG. 11A, an L sample (for example, L = 1)
60) is moved in the time axis direction at the frame interval,
The overlap between the blocks is NL samples (for example, 96 samples). In addition, the windowing processing unit 10
4, a predetermined window function for one block N samples,
For example, a hamming window is applied, and the windowed block is sequentially moved in the time axis direction at intervals of one frame L samples.

When such a windowing process is represented by a mathematical formula, _xw (k, q) = x (q) w (kL-q) (14) In the equation (14), k is a block number, q
Represents the time index (sample number) of the data,
For the q-th data x (q) of the input signal before processing, the k-th data
This shows that data x _w (k, q) can be obtained by performing a windowing process using a block window function (w (kL-q)). The window function _wr (r) in the case of a rectangular window as shown in FIG. 11A in the pitch extracting unit 103 is represented by _wr (r) = 1 0 ≦ r <N (15) = 0r <0, N ≦ r Further, the window function w _h (r) in the case of the Hamming window as shown in FIG. 11B in the windowing processing unit 104 is represented by w _h (r) = 0.54−0.46 cos (2πr / (N-1)) 0 ≦ r <N (16) = 0 r <0, N ≦ r. When such a window function w _r (r) or w _h (r) is used, the window function w (r) (= w (kL−
The non-zero section of q)) is expressed as follows: 0 ≦ kL−q <N By modifying this, kL−N <q ≦ kL Therefore, for example, in the case of the rectangular window, the window function w _r (kL−q) = 1
As shown in FIG. 12, kL−N <q ≦ kL
It is time of. Further, the above equations (14) to (16) indicate that the length N
(= 256) indicates that the sample window advances by L (= 160) samples. The following (15)
Each N point (0 ≦ r) extracted by each window function of the equation (16)
<N) are represented by x _wr (k, r) and x _wh , respectively.
(k, r).

As shown in FIG. 13, the windowing processing unit 104 takes 17 samples from the sample sequence x _wh (k, r) of 256 samples per block to which the Hamming window of the above equation (16) is applied.
Zero data for 92 samples is added (so-called zero padding) to make 2048 samples, and the orthogonal transform unit 105 performs orthogonal transform such as FFT (fast Fourier transform) on the time axis data sequence of 2048 samples. Processing is performed.

In the pitch extracting section 103, the above x _wr (k, r)
Is extracted based on the sample sequence (1 block N samples). As the pitch extraction method, a method using a periodicity of a time waveform, a periodic frequency structure of a spectrum, an autocorrelation function, and the like are known.
The autocorrelation method of the center clip waveform is adopted. As for the center clip level in the block at this time, one clip level may be set for each block, but the peak level of the signal of each part (each sub-block) obtained by subdividing the block is detected, and these are detected. When the difference between the peak levels of each sub-block is large, the clip level in the block is changed stepwise or continuously. The pitch period is determined based on the peak position of the autocorrelation data of the center clip waveform. At this time, a plurality of peaks are obtained from the autocorrelation data belonging to the current frame (the autocorrelation is obtained from data of one block N samples), and the maximum peak among the plurality of peaks is equal to or larger than a predetermined threshold. In this case, the maximum peak position is used as the pitch period, and in other cases, a pitch within a pitch range that satisfies a predetermined relationship with the pitch obtained in a frame other than the current frame, for example, the previous and next frames, for example, the center of the pitch of the previous frame As a result, a peak within a range of ± 20% is obtained, and the pitch of the current frame is determined based on the peak position. In the pitch extracting section 103, a relatively rough pitch search by an open loop is performed, and the extracted pitch data is input to a high-precision (fine) pitch searching section 10
6 to perform a high-precision pitch search (fine search of pitch) by a closed loop.

High precision (fine) pitch search section 106
Contains the coarse (rough) pitch data of the integer value extracted by the pitch extracting unit 103 and the orthogonal transform unit 10.
5, for example, the data on the frequency axis that has been subjected to the FFT. In the high precision pitch search unit 106,
With the coarse pitch data value as the center, ± 0.2 steps
Shake several samples at a time to drive to the optimal fine pitch data with a decimal point (floating). At this time, as a method of fine search, a so-called analysis by synthesis method is used, and the pitch is selected so that the synthesized power spectrum is closest to the power spectrum of the original sound.

The fine search of the pitch will be described. First, in the MBE vocoder, the FF
S (j) as spectrum data on the frequency axis orthogonally transformed by T or the like is expressed as S (j) = H (j) | E (j) | 0 <j <J (17) Model is assumed. Where J is
_{ω s / 4π = f s /} 2 in response, the sampling frequency f
_s = when ω _s / 2π is, for example, 8kHz corresponding to 4kHz. In the above equation (17), when the spectrum data S (j) on the frequency axis has a waveform as shown in FIG.
H (j) indicates the spectral envelope of the original spectral data S (j) as shown in FIG.
E (j) indicates the spectrum of the excitation signal (excitation) at the same level and periodic as shown in FIG. 14C. That is, the FFT spectrum S (j) is represented by the spectrum envelope H (j) and the power spectrum of the excitation signal |
E (j) |

Power spectrum | E (j) of the above excitation signal
| Takes into account the periodicity (pitch structure) of the waveform on the frequency axis determined according to the pitch, and converts the spectrum waveform corresponding to the waveform of one band (band) for each band on the frequency axis. It is formed by arranging it repeatedly. The waveform for one band is obtained by performing a FFT by regarding a waveform obtained by adding 0 data of 1792 samples (0 padding) to a Hamming window function of 256 samples as shown in FIG. It can be formed by cutting out an impulse waveform having a certain bandwidth on the frequency axis according to the pitch.

Next, for each band divided according to the pitch, a value (a kind of amplitude) | A representative of the above H (j) (to minimize the error for each band) _m
| Here, for example, when the lower and upper points of the m-th band (band of the m-th harmonic) are a _m and b _m , respectively, the error ε _m of the m-th band is

[0069]

(Equation 7)

Can be expressed by | A _m | that minimizes this error ε _m is

[0071]

(Equation 8)

When | A _m | in equation (19),
Minimize the error ε _m . The amplitude | A _m | is obtained for each band, and the error ε _m for each band defined by the above equation (18) is obtained using the obtained amplitude | A _m |. Next, a total value Σε _m of all the bands of the error ε _m for each band is obtained. Further, the error sum Shigumaipushiron _m of all such bands, calculated for different pitches to some small, obtaining the pitch as an error sum Shigumaipushiron _m is minimized.

That is, several types are prepared above and below, for example, in increments of 0.25 with the rough pitch obtained by the pitch extraction unit 103 as the center. Each respective pitches of different pitches to these plurality of types of fine finding the error sum Σε _m. In this case, when the pitch is determined, the bandwidth is determined. From the above equation (19), the power spectrum | S (j) | of the data on the frequency axis and the excitation signal spectrum | E
(j) | seek error epsilon _m of the equation (18) using a, can be obtained sum Shigumaipushiron _m of all the bands. Obtains the error sum Shigumaipushiron _m for each pitch, is not to determine the pitch corresponding to error sum total value which is the smallest as the optimal pitch. As described above, the optimum fine (for example, in increments of 0.25) pitch is obtained by the high-precision pitch search unit 106, and the amplitude | A _m |
Is determined.

In the above description of the fine search of the pitch, in order to simplify the description, all the bands are voiced (Vo).
iced), but the MBE
In a vocoder, an unvoiced sound (Un
Since the model in which a voiced) region exists is used, it is necessary to discriminate voiced / unvoiced sounds for each band.

The data of the optimum pitch and amplitude | A _m | from the high-precision pitch search unit 106 is sent to the voiced / unvoiced sound discriminating unit 107, and the voiced / unvoiced sound is discriminated for each band. For this determination, NSR (noise-to-signal ratio) is used. That is, the NSR of the m-th band is

[0076]

(Equation 9)

When this NSR value is larger than a predetermined threshold value (for example, 0.3) (error is large), | S (j) by | A _m || E (j) | | Is not good (the excitation signal | E (j) | is inappropriate as a basis), and the band is identified by UV (Unvoice
d, unvoiced sound). In other cases, it can be determined that the approximation has been performed to some extent, and the band is
(Voiced, voiced sound).

Next, the amplitude re-evaluation unit 108 receives the on-frequency data from the orthogonal transformation unit 105 and the amplitude | A _m evaluated as a fine pitch from the high-precision pitch search unit 106.
| And the voiced / unvoiced sound discriminating unit 107
V / UV (voiced sound / unvoiced sound) discrimination data is supplied. The amplitude reevaluating unit 108 calculates the amplitude again for the band determined to be unvoiced (UV) by the voiced / unvoiced sound determining unit 107. The amplitude | A _m | _UV for this UV band is

[0079]

(Equation 10)

Is obtained.

The data from the amplitude reevaluating unit 108 is
Data number conversion (a kind of sampling rate conversion) unit 10
9 The data number conversion unit 109 is provided to make the number constant in consideration of the fact that the number of division bands on the frequency axis differs according to the pitch and the number of data (particularly the number of amplitude data) differs. is there. That is, for example, if the effective band is up to 3400 Hz, this effective band is divided into 8 to 63 bands according to the pitch, and the amplitude | A obtained for each of these bands is obtained.
The number m _MX +1 of _m | (including the amplitude of the UV band | A _m | _UV ) data also changes from 8 to 63. Therefore, the data number conversion unit 109 converts the variable number m _MX +1 of amplitude data into a fixed number N _C (for example, 44) data.

Here, in this embodiment, dummy data which interpolates values from the last data in a block to the first data in a block with respect to the amplitude data of one effective band on the frequency axis. Is added to expand the number of data to N _F , and then the band-limited K _OS times (for example, 8
Obtain an amplitude data of K _OS times the number by performing oversampling multiplied), the K _OS times the number ((m _MX +
1) × K _OS amplitude data is linearly interpolated and expanded to more N _M (for example, 2048), and this N _M data is decimated to obtain the constant number N _C (for example, 44). Convert to data.

The data (the fixed number N _C of amplitude data) from the data number conversion unit 109 is sent to the vector quantization unit 110 and is grouped into a predetermined number of data to form a vector. Will be applied. As the vector quantization unit 110, the above-described hierarchically structured vector quantization unit 15 as shown in FIG. 1 is used.
Therefore, the quantized output data from the vector quantizing unit 110 extracted via the output terminal 111 includes upper-order index data extracted from the S-dimensional vector quantizer 22 shown in FIG. Each of the vector quantizers 25 _{1 to} 25 _S to each output terminal 27 _{1 to} 27
_This is a combination of the lower index data extracted through _S.

The high-precision pitch search unit 106
, High-precision (fine) pitch data is encoded by the pitch encoding unit 115 and extracted via the output terminal 112. Further, the voiced / unvoiced sound (V / UV) discrimination data from the voiced / unvoiced sound discriminating unit 107 is extracted via an output terminal 113. The data from each of these output terminals 111 to 113 is transmitted as a signal of a predetermined transmission format.

Each of these data is obtained by processing the data in the block of N samples (for example, 256 samples). , The data to be transmitted is obtained in the frame unit. That is, the pitch data, V / UV discrimination data, and amplitude data are updated in the frame cycle.

Next, an embodiment of a decoding method according to the present invention will be described with reference to the drawings. FIG. 15 shows a schematic configuration of a synthesizing side (decoder side) according to an embodiment of the present invention for synthesizing an audio signal based on each of the data obtained by transmission. 15, the input terminal 121 is supplied with the vector-quantized amplitude data, the input terminal 122 is supplied with the encoded pitch data, and the input terminal 123 is supplied with the V / UV discrimination data. You. The quantized amplitude data from the input terminal 121 is sent to the inverse vector quantizer 124 and inversely quantized. This inverse vector quantization unit 124 is also used in FIG.
Is a hierarchically structured inverse vector quantization unit corresponding to the hierarchically structured vector quantization unit 15, and synthesizes and outputs the inverse vector quantized data based on the index data of each layer.

The output data from the inverse vector quantization unit 124 is sent to the data number inverse conversion unit 125 where it is inversely transformed, and the obtained amplitude data is sent to the voiced sound synthesis unit 126 and the unvoiced sound synthesis unit 127. The encoded pitch data from the input terminal 122 is decoded by the pitch decoder 128 and sent to the data number inverse converter 125, the voiced sound synthesizer 126, and the unvoiced sound synthesizer 127. The V / UV discrimination data from input terminal 123 is sent to voiced sound synthesis section 126 and unvoiced sound synthesis section 127.

The voiced sound synthesizing section 126 has, for example, a cosine (cosin
e) A voiced sound waveform on the time axis is synthesized by wave synthesis, and the unvoiced sound synthesis unit 127 synthesizes an unvoiced sound waveform on the time axis by filtering, for example, white noise with a band-pass filter. The unvoiced sound synthesized waveform is added and synthesized by the adder 129 and extracted from the output terminal 130. In this case, the amplitude data, the pitch data, and the V / UV discrimination data are updated and provided every frame (L samples, for example, 160 samples) at the time of the analysis, but the continuity between the frames is improved (smoothness). For example, each value of the amplitude data and the pitch data is set as a data value at, for example, a center position in one frame, and each data value up to the center position of the next frame (one frame at the time of synthesis) is interpolated. Ask by That is,
In one frame at the time of synthesis (for example, from the center of the analysis frame to the center of the next analysis frame), each data value at the leading sample point and the end point (the leading edge of the next combined frame)
Each data value at a sample point is given, and each data value between these sample points is obtained by interpolation.

Hereinafter, the synthesizing process in voiced sound synthesizing section 126 will be described in detail. The one synthesized frame (L samples, for example, 160 samples) on the time axis in the m-th band (m-th harmonic band) determined as V (voiced sound)
Assuming that the voiced sound of the minute is V _m (n), V _m (n) = A _m (n) cos (θ _m (n)) 0 using the time index (sample number) n in this synthesized frame. ≦ n <L (22) The final voiced sound V (n) is synthesized by adding (ΣV _m (n)) the voiced sounds of all the bands determined as V (voiced sound) in all the bands.

A _m (n) in the equation (22) is the amplitude of the m-th harmonic interpolated from the top to the end of the composite frame. In the simplest case, the value of the m-th harmonic of the amplitude data updated for each frame may be linearly interpolated.
That is, the m-th position at the end (n = 0) of the composite frame
The amplitude value of the harmonic is A _0m , and the end of the synthesized frame (n =
L: When the amplitude value of the m-th harmonic at the tip) of the next synthesized frame is A _Lm, the formula _{A m (n) = (Ln} ) A 0m / L + nA Lm / L ··· (23) A _m (n) may be calculated.

[0091] Next, the (9) phase theta _m (n) in the expression by _{θ m (n) = mω O1} n + n 2 m (ω L1 -ω 01) / 2L + φ 0m + Δωn ··· (24) You can ask. In the equation (24), φ _0m indicates the phase of the m-th harmonic (frame initial phase) at the front end (n = 0) of the composite frame, and ω ₀₁ indicates the phase at the front end (n = 0) of the composite frame. The fundamental angular frequency ω _L1 indicates the fundamental angular frequency at the end of the combined frame (n = L: the leading end of the next combined frame). Δω in the above equation (24) is
The minimum Δω is set so that the phase φ _{Lm at} n = L becomes equal to θ _m (L).

Hereinafter, a method of obtaining the amplitude A _m (n) and the phase θ _m (n) according to the V / UV discrimination result when n = 0 and n = L in an arbitrary m-th band will be described. I do. When the m-th band is V (voiced sound) for both n = 0 and n = L, the amplitude A _m (n) is calculated by the above equation (23) using the transmitted amplitude values A _0m , A Amplitude A by linear interpolation of A _Lm
m (n) may be calculated. The phase θ _m (n) is n = 0 and θ
Δω is set so that θ _m (L) becomes φ _Lm when _m (0) = φ _0m and n = L.

Next, when n = 0, V (voiced sound) and n =
If L (unvoiced sound) at L, the amplitude A _m (n)
Is linearly interpolated so that 0 A _m (L) from the transmission amplitude value A _{0 m} of A _m (0). The transmission amplitude value A _Lm at n = L is the amplitude value of the unvoiced sound, and is used in unvoiced sound synthesis described later. Phase θ _m (n) is set to _{θ m (0) = φ 0m} , and [Delta] [omega =
Set to 0.

Further, UV (unvoiced sound) when n = 0,
If V = voiced sound when n = L, the amplitude _Am
(n) sets the amplitude A _m (0) at n = 0 to 0 and performs linear interpolation so that the transmitted amplitude value A _Lm at n = L. Phase θ
The _m (n), the phase theta _m as (0) at n = 0, by using the phase value phi _Lm at the frame _{end, θ m (0) = φ} Lm -m (ω O1 + ω L1) L / 2 (25) and Δω = 0.

When both n = 0 and n = L are V (voiced sound), Δω is set so that θ _m (L) becomes φ _Lm.
The method for setting is described. In the above equation (11), n
= By placing the _{L, θ m (L) =} mω O1 L + L 2 m (ω L1 -ω 01) / 2L + φ 0m + ΔωL = m (ω O1 + ω L1) L / 2 + φ 0m + ΔωL = φ Lm becomes, organize this Then, [Delta] [omega is, Δω = (mod2π ((φ Lm -φ 0m) - and mL become _{(ω O1 + ω L1) /} 2) / L ··· (26) this equation (26) in mod2π (x). Is the main value of x
This function returns a value between π and + π. For example, x = 1.3
mod2π (x) = -0.7π when π, mod2 when x = 2.3π
When π (x) = 0.3π, x = -1.3π, mod2π (x) = 0.7
π, and so on.

FIG. 16A shows an example of the spectrum of an audio signal. Bands (harmonic numbers) m of 8, 9, and 10 are set to UV (unvoiced sound) and other bands are set to UV (unvoiced sound). Is V (voiced sound). This V
The time axis signal of the (voiced sound) band is the voiced sound synthesis unit 1
26, and the time axis signal of the UV (unvoiced sound) band is synthesized by the unvoiced sound synthesis unit 127.

Hereinafter, the unvoiced sound synthesizing process in the unvoiced sound synthesizing section 127 will be described. The white noise signal waveform on the time axis from the white noise generating unit 131 is windowed with an appropriate window function (for example, a hamming window) with a predetermined length (for example, 256 samples), and the STFT (short circuit) is performed by the STFT processing unit 132. By performing a term Fourier transform) process, a power spectrum on the frequency axis of white noise as shown in FIG. 16B is obtained. The power spectrum from the STFT processing unit 132 is sent to the band amplitude processing unit 133, and as shown in FIG. 16C, the amplitude | for the UV (unvoiced) band (for example, m = 8, 9, 10) _Am | _UV is multiplied, and the amplitude of the other V (voiced sound) bands is set to zero. The band amplitude processing section 133 is supplied with the amplitude data, the pitch data, and the V / UV discrimination data. The output from the band amplitude processing unit 133 is sent to the ISTFT processing unit 134, and the phase is converted to a signal on the time axis by performing inverse STFT processing using the phase of the original white noise. The output from the ISTFT processing unit 134 is output to an overlap adding unit 135.
And repeats the overlap and addition while weighting appropriately (to restore the original continuous noise waveform) on the time axis to synthesize a continuous time axis waveform. The output signal from the overlap adding unit 135 is sent to the adding unit 129.

As described above, the respective signals of the voiced sound portion and the unvoiced sound portion that have been synthesized in the synthesis portions 126 and 127 and returned on the time axis are added by the addition portion 129 at an appropriate fixed mixing ratio. The reproduced audio signal is extracted from the output terminal 130.

As for the configuration on the voice analysis side (encoding side) in FIG. 10 and the configuration on the voice synthesis side (decoding side) in FIG. 16, each part is described in terms of hardware. It can also be realized by a software program using a signal processor) or the like.

The present invention is not limited to the above embodiment. For example, not only audio signals but also audio signals can be used as input signals.

[0101]

As is apparent from the above description, according to the encoding method of the present invention, an M-dimensional vector data is divided into a plurality of groups by using a structured codebook, and each group is divided into a plurality of groups. By calculating the representative value, the dimension is S
Dimensional (S <M), perform first vector quantization on the data of this S-dimensional vector, and obtain an S-dimensional code vector which is a local decode output at the time of the first vector quantization. The S-dimensional code vector is extended to an original M-dimensional vector to obtain data representing the relationship between the original M-dimensional vector and the data on the frequency axis, and a second vector quantization is performed on this data. , The amount of codebook search computation is reduced, the memory capacity for the codebook is reduced, and the error correction code can be effectively applied to the upper and lower layers of the hierarchical structure. Become.

According to the decoding method of the present invention,
The data on the frequency axis is obtained by using the audio signal as an M-dimensional vector, the dimension is reduced from the M-dimension to the S-dimension (S <M), the first vector quantization is performed on the S-dimension vector data, and the index is encoded. , The first vector quantized output data is inversely quantized to obtain a corresponding S-dimensional code vector, and this S-dimensional code vector is converted to the original M
A second vector quantization on the data representing the relationship between the expanded M-dimensional vector and the original data on the frequency axis of the M-dimensional vector,
When receiving and decoding a code string generated by encoding the index, the second vector quantization index is inversely quantized to obtain a corresponding code vector, and the first vector quantization The index is inversely quantized to obtain a corresponding S-dimensional code vector, the S-dimensional code vector is extended to an M-dimensional vector, and the code vector obtained based on the second vector quantization index and the By obtaining data on the frequency axis of the M-dimensional vector from the expanded M-dimensional vector and restoring the audio signal based on the data on the frequency axis of the M-dimensional vector, the data generated by the above-described encoding method is obtained. The decoded code string can be decoded.

[0103]

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a schematic configuration of an encoding device (encoder) to which a high-efficiency encoding method according to an embodiment of the present invention is applied.

FIG. 2 is a diagram for explaining an operation of vector quantization having a hierarchical structure.

FIG. 3 is a diagram for explaining an operation of vector quantization having a hierarchical structure.

FIG. 4 is a diagram for explaining an operation of vector quantization having a hierarchical structure.

FIG. 5 is a diagram for explaining an operation of vector quantization having a hierarchical structure.

FIG. 6 is a diagram for explaining an operation of vector quantization having a hierarchical structure.

FIG. 7 is a diagram for explaining an operation of vector quantization having a hierarchical structure.

FIG. 8 is a block diagram illustrating a schematic configuration of an encoding device (encoder) to which a high-efficiency encoding method according to a second embodiment of the present invention is applied.

FIG. 9 is a block diagram illustrating a schematic configuration of an encoding device (encoder) to which a high-efficiency encoding method according to a third embodiment of the present invention is applied.

FIG. 10 is a functional block diagram illustrating a schematic configuration of an analysis side (encoding side) of a speech signal synthesis analysis encoding apparatus as a specific example of an apparatus to which the high efficiency encoding method according to the embodiment of the present invention is applied. .

FIG. 11 is a diagram for explaining windowing processing.

FIG. 12 is a diagram for explaining a relationship between a windowing process and a window function.

FIG. 13 is a diagram showing time axis data as an object of orthogonal transform (FFT) processing.

FIG. 14 is a diagram showing spectrum data on a frequency axis, a spectrum envelope (envelope), and a power spectrum of an excitation signal.

FIG. 15 is a functional block diagram showing a schematic configuration of a synthesis side (decoding side) of a speech signal synthesis analysis coding apparatus as a specific example of a device to which the high efficiency coding method according to the embodiment of the present invention is applied; .

FIG. 16 is a diagram for explaining non-speech synthesis when synthesizing a speech signal.

[Explanation of symbols]

12: Frequency axis conversion processing unit 13: Non-linear compression unit 14: Difference processing unit between blocks (between frames) 15: Vector quantization unit 21: Dimension reduction unit 22 S-dimensional vector quantizer 23 Dimension extension unit 25 _{1 to} 25 _S Vector quantizer 103 Pitch extraction unit 104 ··· Window processing unit 105 ··· Orthogonal transform (FFT) unit 106 ··· High-precision (fine) pitch search unit 107 ··· Voiced / unvoiced (V / UV) discrimination Unit 108... Amplitude re-evaluation unit 109... Data number conversion (data rate conversion) unit 110... Vector quantization unit 126... Voiced sound synthesis unit 127. ..Unvoiced sound synthesizer

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-2-287500 (JP, A) JP-A-64-74600 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G10L 19/00

Claims (5)

(57) [Claims] A step of obtaining data on a frequency axis as an M-dimensional vector based on data obtained by dividing an input audio signal into blocks and converting the data into a frequency axis; Dividing the upper data into a plurality of groups and obtaining a representative value for each group to reduce the dimension to S dimension (S <M); and performing first vector quantization on the data of the S dimension vector. Applying, dequantizing the first vector quantized output data to obtain a corresponding S-dimensional code vector, extending the S-dimensional code vector to an original M-dimensional vector, Performing a second vector quantization on data representing the relationship between the extended M-dimensional vector and the original data on the frequency axis of the M-dimensional vector. Coding method comprising Rukoto. 2. The encoding method according to claim 1, wherein as the data on the frequency axis of the M-dimensional vector, data obtained by nonlinearly compressing the data converted to the frequency axis in block units is used. 3. The encoding method according to claim 1, wherein a difference between blocks of the data to be vector-quantized is obtained, and the inter-block difference data is subjected to vector quantization. 4. A step of dividing an input audio signal into blocks and converting the amplitude of a spectrum obtained by converting the signal into a frequency domain into a dB area to obtain data on the frequency axis as an M-dimensional vector; Dividing the data on the frequency axis of the M-dimensional vector into a plurality of groups and obtaining an average value for each group to reduce the dimension to S-dimension (S <M); Applying a first vector quantization to the first vector quantized output data to obtain a corresponding S-dimensional code vector by dequantizing the first vector quantized output data; Expanding to a vector, and performing a second vector quantization on the difference data between the expanded M-dimensional vector and the original data on the frequency axis of the M-dimensional vector. Encoding method characterized by having a to process. 5. An audio signal as an M-dimensional vector
Find the data on the frequency axis, and from the M dimension to the S dimension (S <M)
To the first dimension for the data of the S-dimensional vector.
The vector is quantized and the index is encoded.
The vector quantized output data of 1 is inversely quantized to correspond
Find the S-dimensional code vector and calculate the S-dimensional code vector.
Extend the vector to the original M-dimensional vector,
-Dimensional vector and the frequency axis of the original M-dimensional vector
A second vector quantity for the data representing the relationship with the upper data
By encoding the index
A decoding method that receives and decodes the generated code string
Then , the index of the second vector quantization is inversely quantized.
First inverse vector quantity for finding the corresponding code vector
Dequantizing the index of the first vector quantization
To find a corresponding S-dimensional code vector
Vector quantization process and expansion of the S-dimensional code vector to an M-dimensional vector
And the code vector obtained in the first inverse vector quantization step.
From the extended M-dimensional vector and the M-dimensional vector
Obtaining the data on the frequency axis of the vector, based on the data on the frequency axis of the M-dimensional vector,
And characterized by a step of restoring the audio signal
The decoding method to use.