JPH05297898A - Data quantity converting method - Google Patents

Abstract

PURPOSE:To convert a variable number of input data into a fixed number of data without generating any ringing at a block end point in a nonlinear area. CONSTITUTION:A nonlinear compression part 12 performs the nonlinear compression of a variable number of data by inputted blocks and a dummy data addition part 14 adds dummy data for interpolation from the final data value in the block to the starting data value to increase the number of data; and the band limiting type oversampling part 15 consisting of an FFT(fast Fourier transformation) processing part 16, an intermediate 0-filling process part 17, and an IFFT(inverse FFT) process part 18 performs oversampling, a linear interpolation part 19 performs linear interpolation, and a thinning-out process part 20 thins out the data to convert the data into a fixed number of sample data.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data number conversion method, and more particularly to converting a variable number of data such as spectrum amplitude data calculated in a voice synthesis analyzer (vocoder) or the like into a fixed number of data. The present invention relates to such a data number conversion method.

[0002]

2. Description of the Related Art Various coding methods are known in which signal compression is performed by utilizing statistical characteristics of audio signals (including voice signals and acoustic signals) in the time domain and frequency domain and human auditory characteristics. ing. As this encoding method,
Broadly speaking, time domain coding, frequency domain coding,
Examples include analysis and synthesis coding.

As an example of high-efficiency coding of a voice signal or the like, M
BE (Multiband Excitation) coding, SBE (Singleband Excitation) coding, Harmonic coding, SBC
(Sub-band Coding), LPC (Linear
Predictive Coding: Linear predictive coding) or DC
T (Discrete Cosine Transform), MDCT (Modified DC)
T), FFT (Fast Fourier Transform), etc., conventionally, when quantizing various information data such as spectrum amplitude and its parameters (LSP parameter, α parameter, k parameter, etc.), 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 to further improve the quantization efficiency, the quantization noise (distortion) becomes large in the scalar quantization. , Was difficult to put into practical use. Therefore, time axis data, frequency axis data, filter coefficient data, etc. obtained at the time of these encodings are not individually quantized, but a plurality of data are combined into a set (vector) and expressed by one code. Attention has been paid to vector quantization for quantization.

However, the above MBE, SBE, LP
Since the number of spectral amplitude data such as C changes depending on the pitch, if vector quantization is to be performed as it is, variable-dimensional vector quantization is required, which not only complicates the configuration but also provides good characteristics. Hard to get.

Further, even when a difference between blocks (frames) of data is obtained before quantization, the difference cannot be obtained unless the numbers of data in preceding and following blocks (frames) match. .. As described above, it may be necessary to convert a variable number of data into a fixed number in the process of data processing, but it is desired to convert the number of data with good characteristics.

The present invention has been made in view of the above circumstances, and can convert a variable number of data into a fixed number of data, and can perform data number conversion with good characteristics in which ringing or the like does not occur at an end point. The purpose is to provide such a data number conversion method.

[0008]

A data number conversion method according to the present invention comprises a step of nonlinearly compressing data in which the number of waveform data in a block or parameter data representing the waveform is variable, and variable for each block. By using a data number converter for converting the variable number of non-linear compressed data into the constant number in order to compare the number of non-linear compressed data for each block with a fixed number of reference data in a non-linear region. , To solve the above problems.

Here, the number of data is expanded by adding dummy data for interpolating values from the last data in the block to the first data in the block to the variable number of non-linear compressed data for each block. After that, it is preferable to perform band-limited oversampling. Dummy data that interpolates the value from the last data in the block to the first data in the block is such that it does not cause a sudden change in the value at the end point of the block.
Or it is data that does not jump or discontinuity in the value, change the first data value in the block after holding the last data value in the block for a certain period,
An example of such a form of value change is such that the first data value in this block is held for a certain period. In the band-limited oversampling, orthogonal transformation such as FFT (Fast Fourier Transform) is performed, and 0s are packed (or low-pass filtered) only in a section according to a multiple of oversampling, and then IFFT (inverse FFT). Inverse orthogonal transformation such as

As the non-linearly compressed data, audio signals such as voice signals and acoustic signals converted into data on the frequency axis can be used. Specifically,
For example, spectral envelope amplitude data in the case of MBE (Multiband Excitation) encoding, S
BE (Singleband Excitation) coding, Harmonic coding, SBC (Sub-
band Coding: Band division coding, LPC (Linear Predi
ctive Coding: Linear predictive coding), DCT (discrete cosine transform), MDCT (modified DCT),
Spectral amplitude data in FFT (Fast Fourier Transform) or the like and its parameter (LSP parameter, α parameter, k parameter, etc.) data and the like can be used. In addition, it is possible to vector quantize the data converted into a fixed number, and before this vector quantization,
It is also possible to obtain a difference between blocks of a fixed number of data for each block and perform vector quantization on the difference data between blocks.

[0011]

It is possible to convert into a fixed number of non-linear compressed data and compare it with the reference data in the non-linear region, and it is possible to perform vector quantization by taking the difference between blocks. In addition, the continuity of the data values in the block before the data number conversion is enhanced, and the high-quality data number conversion that does not cause ringing at the block end point can be performed.

[0012]

Embodiments of the data number conversion method according to the present invention will be described below with reference to the drawings. FIG. 1 shows a schematic configuration of a data number conversion method according to an embodiment of the present invention.

In FIG. 1, an input terminal 11 is supplied with amplitude data of a spectral envelope calculated by an MBE vocoder which will be described later. For this amplitude data, for example, a voice signal having a spectrum as shown in A of FIG. 2 is analyzed to obtain a pitch frequency (angular frequency) ω, and the periodicity of the spectrum corresponding to the pitch frequency ω is taken into consideration. When the amplitude at each harmonic (harmonic) position is obtained and the amplitude data representing the spectrum envelope (envelope) as shown in B of FIG. 2 is obtained, the amplitude within a certain effective band (for example, 200 to 3400 Hz) is obtained. The number of amplitude data changes depending on the pitch frequency ω. Therefore, a fixed frequency (angular frequency) ω _c is assumed, and the amplitude data of the spectrum envelope at each harmonic position of this constant frequency ω _c is obtained, thereby making the number of data constant.

In the example of FIG. 1, a variable number (m _MX +1) of input data from the input terminal 11 is compressed (logarithmically compressed) in the non-linear compression unit 12 into, for example, the dB area, and then the data number conversion main body unit. At 13, the data is converted into a fixed number (M) of data. The data number conversion main body unit 13 has a dummy data addition unit 14 and a band-limited oversampling unit 15, and the band-limited oversampling unit 15 performs orthogonal transformation, for example, FFT (Fast Fourier Transform) processing unit 16,
Intermediate zero padding processing unit 17 and inverse orthogonal transform, for example, IFF
It is composed of a T (inverse FFT) processing unit 18. The data subjected to the band-limited oversampling is linearly interpolated by the linear interpolation unit 19, thinned by the thinning processing unit 20, and
It becomes a fixed number of data and is taken out from the output terminal 21.

Here, the amplitude data string of m _MX +1 pieces calculated in the MBE vocoder described later is a (m).
m is the order of the above harmonics (harmonics) or the band number, and m _MX is the maximum value, but the number of amplitude data in all bands is m _MX +1 including the amplitude data in the band of m = 0. Becomes This amplitude data a (m) is converted into, for example, a dB region by the non-linear compression unit 14. That is, assuming that the obtained data is a _dB (m), a _dB (m) = 20 log ₁₀ a (m) (1). Since the number m _MX +1 of the logarithmically converted amplitude data a _dB (m) changes depending on the pitch as described above, it is converted into a fixed number (M) of amplitude data b _dB (m). This is a kind of sampling rate (sample rate) conversion. The compression processing in the non-linear compression unit 12 includes, for example, so-called μ-l in addition to logarithmic compression in the dB area.
Pseudo-logarithmic compression processing such as aw and α-law may be performed.
Thus, by compressing the amplitude, efficient coding is realized.

The sampling frequency fs for a voice signal on the time axis input to the MBE vocoder is usually 8 kHz.
The total bandwidth is 3.4 kHz (the effective bandwidth is 200-
3400Hz), the pitch lag from the higher female voice to the lower male voice (the number of samples corresponding to the pitch period)
Is about 20 to 147. Therefore, the pitch (angle) frequency ω is 8000/147 ≒ 54 (Hz) to 8000/20 = 400
It will fluctuate up to about (Hz). Therefore, about 8 to 63 pitch pulses (harmonics) stand up to the above 3.4 kHz on the frequency axis. That is, as the waveform of the dB region on the frequency axis, the data of 8 to 63 samples is converted into a fixed number of samples, for example, 44 samples, by the number of samples conversion. This corresponds to obtaining a sample of the position of the harmonics for each constant pitch frequency (angular frequency) ω _C , as shown in C of FIG.

Next, in the dummy data adding section 14, the m _MX +1 compressed data a _dB (m) is extended to a number (eg, a power of 2) N _F (for example, N _F = 256) that facilitates FFT. That is, as dummy data a ′ _dB (m) in the section from m _MX +1 to N _F , for example, m _MX + 1 ≦ m <N _F / 2: a ′ _dB (m) = a _dB (m _MX ) N _F / 2 ≦ m <3N _F / 4: a ′ _dB (m) = a _dB (m _MX ) × k ₁ + a _dB (0) × k ₂ where k ₁ = (3N _F / 4−n) / (N _F / 4) k ₂ = (n−N _F / 2) / (N _F / 4) 3N _F / 4 ≦ m <N _F : a ′ _dB (m) = a _dB (0) (2) Use to extend. This is 0 to m as shown in FIG.
_The original amplitude data a _dB (m) is placed in the _MX section and m _MX +1
In the section of ≤m <N _F / 2, the last data in the block, a _dB (m _MX ), is held, and in the section of N _F / 2 ≤m <3N _F / 4, linear interpolation is performed to obtain 3N _F. In the section of / 4 ≦ m <N _F, the line becomes a broken line that holds the first data a _dB (0) in the block.

That is, data is created and packed so that the left end and the right end of the original waveform for rate conversion shown in FIG. 3 are gradually connected. In the FFT, since the waveform before conversion is regarded as a repetitive waveform as shown by the broken line in FIG. 3, the point of m = N _F is connected to m = 0.

This means that if filtering is performed such that multiplication is performed on the frequency axis after the FFT, convolution is performed on the original axis shown in FIG. Therefore, as shown in FIG. 4, the part other than the original waveform (section where m _MX <m <N _F )
If 0 is simply padded, ringing as shown by the broken line R in FIG. 4 occurs at the discontinuity point, and good rate conversion cannot be performed. In order to prevent such a problem, as shown in FIG. 3, dummy data is packed so as not to cause a sudden change in the value at the block end point. In addition to the specific example of the dummy data, the entire data from the last data of the block to the first data of the block may be linearly interpolated as shown by a broken line I in FIG. You may do it.

Next, the FFT (Fast Fourier Transform) processing unit 16 of the band-limited oversampling unit 15 applies the N _F points to the number sequence (data sequence) expanded to the N _F points (N _F samples). FFT is performed to obtain a sequence (spectrum) of 0 to N _F as shown in A of FIG. A portion corresponding to 0~π of this sequence, between a portion corresponding to Pai～2pai, the intermediate zero-filled section _{17, (O S -1) N} F
Pack 0s. O _S at this time, the ratio of the oversampling (ratio). For example, in the case of O _S = 8 is stuffed 7N _F-number of 0 between the 0~π corresponding parts and π~2π substantial portion of the sequence as shown in B of FIG. 5, 8N _F
A series of points (for example, 2048 points when N _F = 256) is used.

Here, the zero padding operation can be regarded as an LPF process. That is, the digital filter operating at O _S N _F with respect to the number sequence of O _S N _F as the sampling rate is π / 8 as shown by the thick line in A of FIG.
The low-pass processing is performed with the cut-off to obtain a sample string as shown in B of FIG. In this filter operation, there is a possibility that ringing as shown by the broken line R in FIG. 4 may occur. In this embodiment, in order to prevent the ringing from occurring, the left end and the right end of the original waveform are smoothly connected (so that there is no abrupt change in the differential coefficient).

Next, the IFFT (Inverse FFT) unit 18, O _S N _F point The reverse FFT (e.g. 2048 points), O _S times the oversampled amplitude data as shown in FIG. 7 was ( (Including dummy data) is obtained. Valid part of the data sequence, i.e. 0~O _S × (m _MX +
1) is extracted, the original waveform (original amplitude data a _dB (m))
Becomes a density of O _S times and an oversampled product is obtained. This is still a variable number according to the pitch m _MX +
It is a data string that depends on 1.

Next, linear interpolation is performed in order to convert this into a fixed number of data. For example, FIG. 8A shows the above-mentioned m _MX.
= 19 (the total number of bands before conversion, the number of amplitude data is 2
Shows the case of 0), this by 8 times oversampling as O _S = 8, as shown in B of FIG. _{8, O S × (m MX} +1 during 0~π) = 160 Although there are a number of sample data, the linear interpolation unit 19 linearly interpolates the sample data to a fixed number N _M , for example N _M = 2048.

FIG. 9A shows a fixed number N _M (for example, N _M = 204) obtained by linearly interpolating by the linear interpolating unit 19.
8) data is shown, and in order to convert this 2048-sample data into a predetermined sample number M (for example, M = 44), the thinning-out processing unit 20 thins out 2048-point data to obtain 44-point data. ing. Where 0th
Since it is not necessary to transmit the DC value (DC data value, 0th data value) of the 2047th sample data, the value of nint (2048/44) · i) is adopted as the thinning value, It suffices to obtain individual data. However, 1 ≦ i ≦ 44, and nint is neares
t integer, that is, a function indicating the nearest integer value.

In this way, a sequence b _dB (n) converted into data of a fixed number M of samples, where 1 ≦ n ≦ M, is obtained. The sequence of the fixed number of data may be subjected to vector quantization, if necessary, between blocks or frames, vector quantization may be performed, and the index may be transmitted.

On the receiving side (combining side or decoder side), vector-quantized and dequantized sequence b _VQdB (n) of M points of waveform data is obtained from the index.
The data string is processed in the same manner, that is, by performing reverse operations of band-limited oversampling, linear interpolation, and decimation, respectively, so that m _MX +1 of the required number of points is obtained.
Convert to a sequence of points. Note that m _MX (or m _MX +1)
Can be obtained from the separately transmitted pitch information. For example, when the pitch period normalized to the sampling period is p, the pitch frequency (angular frequency) ω
Is calculated by 2π / p, and from π / ω = p / 2, m _MX +
It can be calculated as 1 = inint (p / 2). This m _MX
Decoding processing is performed based on the amplitude information of +1 point.

Next, a specific example of a kind of MBE (Multiband Excitation) vocoder of a voice signal synthesis analysis coding apparatus (so-called vocoder) to which the above-described data number conversion method can be applied is shown in the drawings. Will be described with reference to. The MBE vocoder described below is based on DW Griffin and JS Lim, "Multiband Excitatio
n Vocoder, "IEEE Trans.Acoustics, Speech, and Signal
Processing, vol.36, No.8, pp.1223-1235, Aug. 1988
The conventional PARCOR (PA
In the Rtial auto-CORrelation vocoder and the like, the voiced section and the unvoiced section were switched for each block or frame when modeling the voice, whereas in the MBE vocoder, the same time (the same block or It is modeled on the assumption that there are voiced sections and unvoiced sections in the frequency axis region of (in the frame).

FIG. 10 is a block diagram showing the overall schematic construction of an embodiment in which the present invention is applied to the MBE vocoder. In FIG. 10, an audio signal is supplied to the input terminal 101, and the input audio signal is
It is sent to a filter 102 such as an HPF (high-pass filter) to remove a so-called DC (direct current) offset component and at least a low frequency component (200 Hz or less) for band limitation (for example, 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.
The pitch extraction unit 103 divides the input voice signal data into blocks in units of a predetermined number N (for example, N = 256) (or cuts out with a rectangular window), and pitches the voice signals in this block. ..
Such a cut block (256 samples) is used for L samples (for example, L = 1 as shown in FIG. 11A).
It is moved in the time axis direction at the frame interval of 60),
The overlap between blocks is NL samples (96 samples, for example). Also, the windowing processing unit 10
4, the predetermined window function for 1 block N samples,
For example, a Hamming window is provided, and this windowing block is sequentially moved in the time axis direction at intervals of 1 frame L samples.

When this windowing process is expressed by a mathematical expression, _xw (k, q) = x (q) w (kL-q) (3) In this equation (1), k is a block number,
q represents the time index (sample number) of the data, and the q-th data x (q) of the unprocessed input signal is windowed by the window function (w (kL-q)) of the kth block. It is shown that the data x _w (k, q) can be obtained by doing so. The window function w _r (r) in the case of a rectangular window as shown in FIG. 11A in the pitch extraction unit 103 is w _r (r) = 1 0 ≦ r <N (4) = 0 r <0, N ≦ r Further, the window function in the case of Hamming window as shown in B of FIG. 11 in the windowing processing unit 104 w _h (r) _{is, w h (r) = 0.54} - 0.46 cos (2πr / (N-1)) 0 ≦ r <N (5) = 0 r <0, N ≦ r. Such a window function w _r (r) or (3) when using the w _h (r) formula of the window function w (r) (= w (KL-
q)), the zero-zero interval is 0 ≦ kL−q <N, and is modified as follows: kL−N <q ≦ kL Therefore, for example, in the case of the above rectangular window, the window function w _r (kL−q) = 1
As shown in FIG. 12, kL−N <q ≦ kL
It will be when. In addition, the above equations (3) to (5) have a length N
The window of (= 256) samples is shown to advance by L (= 160) samples. Below, above (4)
Each N point (0 ≦ r cut out by each window function of equations (5))
The non-zero sample sequence of <N) is represented by x _wr (k, r) and x _wh
Let us denote it as (k, r).

In the windowing processing unit 104, as shown in FIG. 13, the sample sequence x _wh (k, r) of 256 samples in one block subjected to the Hamming window of the equation (5) is 17
The 0 data for 92 samples are added (so-called zero padding) to form 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 extraction unit 103, the above x _wr (k, r)
Pitch extraction is performed based on the sample sequence (1 block N samples). The pitch extraction method is known to include periodicity of time waveform, periodic frequency structure of spectrum, and autocorrelation function.
The center correlation waveform autocorrelation method is used. Regarding the center clip level in the block at this time, one clip level may be set for one block, but the peak level of the signal of each part (each sub-block) obtained by subdividing the block is detected and When there is a large difference in the peak level of each sub-block, the clip level is changed stepwise or continuously within the block. The peak 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 for the data of N samples of one block), and the maximum peak among the plurality of peaks is equal to or larger than a predetermined threshold. In the case of, the maximum peak position is set as the pitch period, and in other cases, the pitch is within a pitch range that satisfies a predetermined relationship with the pitch other than the current frame, for example, the pitch of the previous frame and the pitch of the previous frame. As a result, a peak in the range of ± 20% is obtained, and the pitch of the current frame is determined based on this peak position. In this pitch extraction unit 103, a relatively rough pitch search is performed by an open loop, and the extracted pitch data has a high precision (fine) pitch search unit 10.
Then, the high precision pitch search (pitch fine search) is performed by the closed loop.

High precision (fine) pitch search section 106
Includes rough pitch data of integer (integer) values extracted by the pitch extraction unit 103 and the orthogonal transformation unit 10.
5, the data on the frequency axis subjected to FFT, for example, is supplied. In this high precision pitch search unit 106,
Centering on the above coarse pitch data value, it is ± 0.2 in increments of ±
Shake several samples at a time to drive to the optimum fine pitch data value with a decimal point (floating). As a fine search method at this time, 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 etc. is expressed as S (j) = H (j) | E (j) | 0 <j <J (6) It is assumed that the model. Here, J corresponds to πω _s = f _s / 2, and the sampling frequency f _s =
When 2πω _s is, for example, 8 kHz, it corresponds to 4 kHz.
In the above equation (6), when the spectrum data S (j) on the frequency axis has a waveform as shown in A of FIG. 14, H (j)
Is the original spectrum data S as shown in B of FIG.
The spectral envelope of (j) is shown as E
(j) shows the spectrum of the excitation signal (excitation) which is periodic at the same level as shown in C of FIG. That is, the FFT spectrum S (j) is the spectrum envelope H (j) and the power spectrum | E of the excitation signal.
(j) | is modeled as the product.

Power spectrum of the excitation signal | E (j)
Is a spectral waveform corresponding to the waveform of one band (band) for each band on the frequency axis in consideration of the periodicity (pitch structure) of the waveform on the frequency axis determined according to the pitch. It is formed by arranging it repeatedly. The waveform for this one band is obtained by FFT by regarding the waveform obtained by adding 0-data for 1792 samples to the 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 above pitch, a value (a kind of amplitude) | A that represents the above H (j) (minimizes the error for each band) | A _m
Ask for |. 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

[0036]

[Equation 1] Can be expressed as | A _m | that minimizes this error ε _m
Is

[0037]

[Equation 2] Therefore, when | A _m | in the equation (8), the error ε _m is minimized. Such an amplitude | A _m | is obtained for each band, and the obtained amplitude | A _m | is used to obtain the error ε _m for each band defined by the above equation (7). Next, the sum Σε of all the bands of such error ε _m for each band
_{Find m} . Further, such an error sum value Σε _m of all bands is obtained for some slightly different pitches, and a pitch that minimizes the error sum value Σε _m is obtained.

That is, several kinds of vertical pitches are prepared with the rough pitch determined by the pitch extraction unit 103 as the center, for example, in 0.25 steps. The error sum value Σε _m is obtained for each of these plural kinds of slightly different pitches. In this case, when the pitch is determined, the bandwidth is determined, and from the above formula (8), the power spectrum | S (j) |
(j) | can be used to find the error ε _m in the above equation (7), and the total sum Σε _m of all bands can be found. This error sum value Σε _m is obtained for each pitch, and the pitch corresponding to the minimum error sum value is determined as the optimum pitch. As described above, the high-precision pitch search unit 106 obtains the optimum fine (eg, 0.25 step) pitch, and the amplitude | A _m | corresponding to this optimum pitch.
Is determined.

In the above description of the pitch fine search, in order to simplify the explanation, all bands are voiced (Vo
Assuming the case of iced), as described above, MBE
In the vocoder, unvoiced sound (Un
Since a model in which a voiced) area exists is used, it is necessary to distinguish voiced sound / unvoiced sound for each band.

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

[0041]

[Equation 3] When this NSR value is larger than a predetermined threshold value (for example, 0.3) (error is large), | S (j) | is approximated by | A _m || E (j) | in that band. Is not good (the above excitation signal | E (j) | is unsuitable as a basis)
Therefore, the band is determined to be UV (Unvoiced, unvoiced sound). In other cases, it can be determined that the approximation is performed to some extent, and the band is determined to be V (Voiced, voiced sound).

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

[0043]

[Equation 4] Required at.

The data from the amplitude re-evaluation unit 108 is
Data number conversion (a kind of sampling rate conversion) unit 10
Sent to 9. This data number conversion unit 109 is for making the number constant, considering that the number of divided bands on the frequency axis differs according to the pitch and the number of data (especially the number of amplitude data) differs. is there. That is, for example, if the effective band is up to 3400 kHz, this effective band is divided into 8 bands to 63 bands according to the pitch, and the amplitude | obtained for each of these bands |
A _m | will change _(UV including UV band amplitude | | A _m) the number of data m _MX +1 also 8-63. Therefore, the data number conversion unit 109 converts the variable number m _MX +1 of amplitude data into a fixed number M (for example, 44) of data.

In this embodiment, as described above with reference to FIGS. 1 to 9, the amplitude data of one block of the effective band on the frequency axis is processed from the last data in the block to the first data in the block. Data is expanded to N _F by adding dummy data for interpolating the values up to the data, and then band-limited O _S times (for example, 8 times) oversampling is applied to increase the O _S times. The amplitude data of the number is obtained, and this number is multiplied by O _S ((m _MX +1) × O _S )
Of the amplitude data is linearly interpolated and expanded to a larger number of N _M pieces (for example, 2048 pieces), and the N _M pieces of data are thinned out to be converted into the fixed number of M pieces (for example, 44 pieces) of data.

The data from the data number converting unit 109 (the above-mentioned fixed number M of amplitude data) is sent to the vector quantizing unit 110, and a predetermined number of data is collected into a vector, and vector quantization is performed. Is given. The quantized output data from the vector quantizer 110 is output to the output terminal 1
It is taken out via 11. The high-precision (fine) pitch data from the high-precision pitch search unit 106 is coded by the pitch coding unit 115, and the output terminal 11
It is taken out via 2. Further, the voiced sound / unvoiced sound (V / UV) discrimination data from the voiced sound / unvoiced sound discrimination unit 107 is taken out through the output terminal 113. The data from these output terminals 111 to 113 are transmitted as signals in a predetermined transmission format.

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

Next, a synthesizing side (decoding side) for synthesizing an audio signal based on the above-mentioned respective data obtained by transmission.
The general configuration of will be described with reference to FIG.
In FIG. 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. It The quantized amplitude data from the input terminal 121 is sent to the inverse vector quantization unit 124 and inversely quantized, and the data number inverse conversion unit 12
It is sent to 5 for inverse conversion. This data number inverse conversion unit 12
In 5, the inverse transformation, which is in contrast to the above description of FIGS. 1 to 9, is performed, 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 decoding unit 128 and sent to the data number inverse conversion unit 125, the voiced sound synthesis unit 126, and the unvoiced sound synthesis unit 127. The input terminal 123
The V / UV discrimination data from is sent to the voiced sound synthesis unit 126 and the unvoiced sound synthesis unit 127.

In the voiced sound synthesis unit 126, for example, cosine (cosin)
e) A voiced sound waveform on the time axis is synthesized by wave synthesis, and in the unvoiced sound synthesis unit 127, for example, white noise is filtered by a bandpass filter to synthesize the unvoiced sound waveform on the time axis, and these voiced sound synthesized waveforms are combined. The unvoiced sound synthesized waveform is added and synthesized by the addition unit 129 and is taken out from the output terminal 130. In this case, the amplitude data, the pitch data, and the V / UV discrimination data are updated and given for each frame (L sample, for example, 160 samples) at the time of the analysis, but the continuity between the frames is improved (smoothed). Therefore, each value of the amplitude data and the pitch data is set as each data value at, for example, the center position in one frame, and each data value up to the center position of the next frame (one frame at the time of composition) is interpolated. Ask by. That is,
In one frame (for example, from the center of the above analysis frame to the center of the next analysis frame) at the time of synthesis, each data value at the tip sample point and the end (the tip of the next synthesis frame)
Each data value at the sample point is given, and each data value between these sample points is obtained by interpolation.

The synthesis processing in the voiced sound synthesis unit 126 will be described in detail below. The one combined frame (L sample, for example, 160 samples) on the time axis in the m-th band (band of the m-th harmonic) determined to be V (voiced sound)
When the voiced sound for a 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 composite frame. ≦ n <L can be expressed as (11). The final voiced sound V (n) is synthesized by adding (ΣV _m (n)) the voiced sounds of all the bands which are determined to be V (voiced sound) of all the bands.

A _m (n) in the equation (11) is the amplitude of the m-th harmonic wave which is interpolated from the beginning to the end of the composite frame. The simplest way is to linearly interpolate the value of the m-th harmonic of the amplitude data updated in frame units.
That is, the m-th frame at the tip (n = 0) of the composite frame
The amplitude value of the harmonic is A _0m , the end of the composite frame (n =
L: the amplitude value of the m-th harmonic at the end of the next composite frame) is A _Lm , A _m (n) = (Ln) A _{0 m} / L + nA _Lm / L ... (12) It is sufficient to calculate A _m (n).

Next, the phase θ _m (n) in the above equation (11) is calculated by θ _m (0) = mω _O1 n + n ² m (ω _L1 −ω ₀₁ ) / 2L + φ _{0 m} + Δω n (13) You can ask. In this equation (13), φ _0m represents the phase (frame initial phase) of the m-th harmonic at the tip (n = 0) of the composite frame, and ω ₀₁ is the tip of the composite frame (n = 0). The fundamental angular frequency, ω _L1, represents the fundamental angular frequency at the end of the combined frame (n = L: the leading end of the next combined frame). Δω in the equation (13) is
Set a minimum Δω such that the phase φ _{Lm at} n = L is equal to θ _m (L).

Hereinafter, how to obtain 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. To do. When the m-th band is V (voiced sound) for both n = 0 and n = L, the amplitude A _m (n) is the amplitude value A _0m transmitted by the above-mentioned equation (12), A _Lm is linearly interpolated and the amplitude is A
_It suffices to calculate _m (n). The phase θ _m (n) is θ when n = 0
Δω is set so that θ _m (L) becomes φ _Lm when _m (0) = φ _{0 m} and n = L.

Next, when n = 0, V (voiced sound) and n =
Amplitude A _m (n) when UV (unvoiced sound) when L
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 when} n = L is the amplitude value of unvoiced sound and is used in unvoiced sound synthesis described later. The phase θ _m (n) is θ _m (0) = φ _{0 m} , and Δω =
Set to 0.

Further, when n = 0, UV (unvoiced sound)
When V = voiced sound when n = L, amplitude A _m
(n) is linearly interpolated so that the amplitude A _m (0) at n = 0 is 0 and the transmitted amplitude value A _Lm is n = L. Phase θ
_{For m} (n), using the phase value φ _Lm at the end of the frame as the phase θ _m (0) at n = 0, θ _m (0) = φ _Lm −m (ω _O1 + ω _L1 ) L / 2 ... (14), 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.
A method of setting will be described. In the above equation (13), n
= L, then 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 . Then, Δω becomes Δω = (mod2π ((φ _Lm −φ _0m ) −mL (ω _O1 + ω _L1 ) / 2) / L (15). In this equation (15), mod 2π (x) Is the principal value of x
It is a function that returns a value between π and + π. For example, x = 1.3
mod2 π (x) = -0.7π when π, mod2 when x = 2.3π
When π (x) = 0.3π and x = -1.3π, mod2π (x) = 0.7
π, and so on.

Here, A of FIG. 16 shows an example of the spectrum of the voice signal, and the bands with band numbers (harmonics number) m of 8, 9, and 10 are UV (unvoiced sound), and other bands. Is V (voiced sound). This V
The time axis signal of the (voiced sound) band is the voiced sound synthesis unit 1 described above.
26, and the time axis signal of the UV (unvoiced sound) band is synthesized by the unvoiced sound synthesis unit 127.

The unvoiced sound synthesizing process in the unvoiced sound synthesizing section 127 will be described below. The white noise signal waveform on the time axis from the white noise generation unit 131 is windowed with an appropriate window function (for example, Hamming window) with a predetermined length (for example, 256 samples), and the STFT processing unit 132 performs STFT (short circuit). By performing the (Term Fourier Transform) process, a power spectrum of white noise on the frequency axis as shown in B of FIG. 16 is obtained. The power spectrum from the STFT processing unit 132 is sent to the band amplitude processing unit 133, and as shown in C of FIG. 16, the above amplitude | for the band (for example, m = 8, 9, 10) set as the UV (unvoiced sound). A _m | _UV is multiplied to set the amplitude of other V (voiced sound) bands to 0. The band amplitude processing unit 133 is supplied with the above amplitude data, pitch data, and 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 into 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 the overlap addition unit 135.
Then, overlapping and addition are repeated with appropriate weighting (so that the original continuous noise waveform can be restored) on the time axis to synthesize a continuous time axis waveform. The output signal from the overlap adder 135 is sent to the adder 129.

As described above, the signals of the voiced sound portion and the unvoiced sound portion which are synthesized in the respective synthesis units 126 and 127 and are returned to the time axis are added by the addition unit 129 at an appropriate fixed mixing ratio, The reproduced audio signal is taken out from the output terminal 130.

Regarding the configuration on the speech analysis side (encoding side) in FIG. 10 and the configuration on the speech synthesis side (decoding side) in FIG. 15, each unit is described as hardware, but a so-called DSP (digital) is used. It is also possible to realize it by a software program using a signal processor or the like.

The present invention is not limited to the above embodiment, and for example, not only a voice signal but also a sound signal can be used as an input signal.

[0062]

As is apparent from the above description, according to the data number conversion method of the present invention, a variable number of data are non-linearly compressed within a block and converted into a constant number of data. It is possible to take the difference between frames), vector quantization, etc., which is effective in improving the coding efficiency. Also, when performing band-limited oversampling processing for data number conversion (sample number conversion),
Since the number of data is expanded by adding dummy data that interpolates between the last data value and the first data value of the data in the block before processing, ringing occurs at the end points due to subsequent filter processing etc. Such a problem can be avoided, and good coding, particularly highly efficient vector quantization can be realized.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration for explaining a data number conversion method according to the present invention.

FIG. 2 is a waveform diagram for explaining an example of a change in the number of data.

FIG. 3 is a diagram showing an example of a waveform in which the number of data before FFT is enlarged.

FIG. 4 is a diagram showing a comparative example of waveforms in which the number of data before FFT is enlarged.

FIG. 5 is a diagram for explaining a waveform after FFT and an oversampling operation.

FIG. 6 is a diagram for explaining a filtering operation on a waveform after FFT.

FIG. 7 is a diagram showing a waveform after IFFT.

FIG. 8 is a diagram showing an example of sample number conversion by oversampling.

FIG. 9 is a diagram for explaining linear interpolation and thinning processing.

FIG. 10 is a functional block diagram showing a schematic configuration of an analysis side (encoding side) of a speech signal synthesis analysis coding apparatus as a specific example of an apparatus to which the data number conversion method according to the present invention is applied.

FIG. 11 is a diagram for explaining a windowing process.

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

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

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 on the synthesis side (decoding side) of a speech signal synthesis analysis coding apparatus as a specific example of an apparatus to which the data number conversion method according to the present invention is applied.

FIG. 16 is a diagram for explaining unvoiced sound synthesis when synthesizing voice signals.

[Explanation of symbols]

12: Non-linear compression section 13: Data number conversion main body section 14: Dummy data addition section 15: Band-limited oversampling section 16: FFT ( Fast Fourier transform) processing unit 17 ... intermediate zero padding processing unit 18 ... IFFT (inverse FFT) processing unit 19 ... linear interpolation unit 20 ... thinning processing unit 103. --- Pitch extraction unit 104 --- Windowing processing unit 105 --- Orthogonal transform (FFT) unit 106 --- High precision (fine) pitch search unit 107 --- Existence Voice / unvoiced (V / UV) discrimination unit 108 ... Amplitude re-evaluation unit 109 ... Data number conversion (data rate conversion) unit 110 ... Vector quantization unit 126 ...・ Voice sound synthesis section 127 ・ ・ ・ None Sound synthesis unit

[Procedure amendment]

[Submission date] June 8, 1993

[Procedure Amendment 1]

[Document name to be amended] Statement

[Correction target item name] 0026

[Correction method] Change

[Correction content]

On the receiving side (composite side or decoder side), the vector-quantized and dequantized sequence b _VQdB (n) of M-point waveform data is obtained from the index. The data string is processed in the same way, i.e., band-limited oversampling, linear interpolation, and decimation.
By doing so, the necessary number of points is converted into a series of m _MX +1 points. Note that m _MX (or m _MX +1)
Can be obtained from the separately transmitted pitch information. For example, when the pitch period normalized to the sampling period is p, the pitch frequency (angular frequency) ω
Is calculated by 2π / p, and from π / ω = p / 2, m _MX
It can be calculated as + 1 = inint (p / 2). This m
Decoding processing is performed based on the amplitude information of _MX + 1 points.

[Procedure Amendment 2]

[Document name to be amended] Statement

[Correction target item name] 0031

[Correction method] Change

[Correction content]

In the pitch extraction unit 103, the above X
Sample sequence of _wr (k, r) (1 block N samples)
Pitch extraction is performed based on This pitch extraction method is known to include periodicity of time waveform, periodic frequency structure of spectrum, and one using autocorrelation function,
In this embodiment, the center correlation waveform autocorrelation method is adopted. Regarding the center clip level in the block at this time, one clip level may be set for one block, but the peak level of the signal of each part (each sub-block) obtained by subdividing the block is detected and When there is a large difference in the peak level of each sub-block, the clip level is changed stepwise or continuously within the block. Based on the peak position of the autocorrelation data of this center clip waveform, the pitch
The period is fixed. At this time, a plurality of peaks are obtained from the autocorrelation data belonging to the current frame (the autocorrelation is obtained for the data of N samples of one block), and the maximum peak of the plurality of peaks is equal to or larger than a predetermined threshold value. When, the maximum peak position is the pitch period,
In other cases, peaks within a pitch range satisfying a predetermined relationship with the pitches obtained in frames other than the current frame, for example, the frames before and after, for example, within a range of ± 20% around the pitch of the previous frame are set. The pitch of the current frame is determined based on the obtained peak position. In this pitch extraction unit 103, a relatively rough pitch search is performed by an open loop,
The extracted pitch data is sent to the high precision (fine) pitch search unit 106, and a high precision pitch search (pitch fine search) is performed by the closed loop.

[Procedure 3]

[Document name to be amended] Statement

[Correction target item name] 0033

[Correction method] Change

[Correction content]

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 etc. is expressed as S (j) = H (j) | E (j) | 0 <j <J (6) It is assumed that the model. Where J is
ω _S / 4π = f _S. 4 kHz when the sampling frequency f _S = ω _S / 2π is 8 kHz, for example.
Corresponding to. In the above equation (6), when the spectrum data S (j) on the frequency axis has a waveform as shown in A of FIG. 14, H (j) is the original spectrum data S as shown in B of FIG. The spectrum envelope (envelope) of (j) is shown, and E (j) has shown the spectrum of the excitation signal (excitation) which is cyclic at equal level as shown in C of FIG. That is, the FFT spectrum S
(J) is modeled as the product of the spectral envelope H (j) and the power spectrum | E (j) | of the excitation signal.

[Procedure amendment 4]

[Document name to be amended] Statement

[Correction target item name] 0036

[Correction method] Change

[Correction content]

[0036]

[Equation 1] Can be expressed as | A _m | that minimizes this error ε _m
Is

[Procedure Amendment 5]

[Document name to be amended] Statement

[Name of item to be corrected] 0037

[Correction method] Change

[Correction content]

[0037]

[Equation 2] Therefore, when | A _m | in this equation (8), the error ε _m is minimized. Such an amplitude | A _m | is obtained for each band, and the obtained amplitude | A _m | is used to obtain an error ε _m for each band defined by the above equation (7). Next, the sum Σω of all the bands of the error ω _m for each band
_{Find m} . Further, such an error sum value Σε _m of all bands is obtained for some slightly different pitches, and a pitch that minimizes the error sum value Σε _m is obtained.

[Procedure Amendment 6]

[Document name to be amended] Statement

[Correction target item name] 0048

[Correction method] Change

[Correction content]

Next, a synthesizing side (decoding side) for synthesizing an audio signal based on the above-mentioned respective data obtained by transmission.
The general configuration of will be described with reference to FIG.
In FIG. 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. It The quantized amplitude data from the input terminal 121 is sent to the inverse vector quantization unit 124 and inversely quantized, and the data number inverse conversion unit 12
It is sent to 5 for inverse conversion. This data number inverse conversion unit 12
5, the same conversion as that described with reference to FIGS. 1 to 9 is performed, 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 decoding unit 128 and sent to the data number inverse conversion unit 125, the voiced sound synthesis unit 126, and the unvoiced sound synthesis unit 127. The V / UV discrimination data from the input terminal 123 is sent to the voiced sound synthesis unit 126 and the unvoiced sound synthesis unit 127.

[Procedure Amendment 7]

[Document name to be amended] Statement

[Correction target item name] 0052

[Correction method] Change

[Correction content]

Next, the phase θ _m (n) in the above equation (11)
Can be calculated by θ _m (n) = mω ₀₁ n + n ² m (ω _L1 −ω ₀₁ ) / 2L + φ _0m + Δωn (13). In this equation (13), φ
_{0 m} represents the phase of the m-th harmonic (frame initial phase) at the tip (n = 0) of the composite frame, ω ₀₁ is the fundamental angular frequency at the composite frame tip (n = 0), and ω _L1 is the composite. The basic angular frequencies at the end of the frame (n = L: leading end of the next composite frame) are shown. Δω in the equation (13) is set to a minimum Δω such that the phase φ _{Lm at} n = L becomes equal to θ _m (L).

Claims (2)

[Claims] 1. A step of nonlinearly compressing data in which the number of waveform data in a block or parameter data representing a waveform is variable, and a variable number of nonlinear compressed data in each block is stored in a fixed number in each block. A step of converting the variable number of non-linear compressed data into the constant number for comparison with reference data. 2. The number of data is expanded by adding dummy data for interpolating values from the last data in the block to the first data in the block to the variable number of non-linear compressed data for each block. The high-efficiency coding method according to claim 1, wherein band-limited oversampling is performed after that.