JPH07104793A - Encoding device and decoding device for voice - Google Patents

Abstract

PURPOSE:To improve the reproducibility of strong consonants such as 's', 't', 'p', and 'k' and to reproduce a voice with good quality by performing a windowing process for a synthesized voiceless sound on a voice synthesis side by using information showing the state of a transmitted pulsing noise. CONSTITUTION:An input signal is divided into blocks in the direction of the time base and a V(voiced sound)/UV(voiceless sound) decision part 17 makes V/UV decisions, block by block. A maximum value detection part 24 detects maximum values in the blocks, a local RMS calculation part 26 calculates respective mean levels of small blocks nearby the maximum values and small blocks before and after it, and a pulse classification part 27 makes classifications of whether or not there is a pulsing noise on the basis of the respective mean levels and also finds information indicating the state of the pulsing noise. When a UV(voiceless sound) decision is made, information on the state of the pulsing noise of the signal waveform is made to substitute for pitch information from a high precision pitch search part 16 through a changeover switch 28, and transmitted through an encoding part 21.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a high-efficiency coding method in which an input voice signal is divided into blocks and the coding process is performed in units of the divided blocks.

[0002]

2. Description of the Related Art Various coding methods are known in which signal compression is performed by utilizing the statistical properties 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 encoding 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)
In T), FFT (Fast Fourier Transform), etc., when quantizing various information data such as spectrum amplitude and its parameters (LSP parameter, α parameter, k parameter, etc.), conventionally, scalar quantization is performed. There are many.

In the voice analysis / synthesis system such as the PARCOR method, since the excitation source is switched at each block (frame) on the time axis, voiced sound and unvoiced sound cannot be mixed in the same frame. , As a result, high quality voice was not obtained.

On the other hand, in the above MBE coding, for each voice in one block (frame), each harmonic (harmonic) of the frequency spectrum or each band (band) in which a few harmonics are grouped together. Since voiced / unvoiced sound discrimination (V / UV discrimination) is performed for each band divided for each band or for each band divided by a fixed bandwidth (for example, 300 to 400 Hz), sound quality Is recognized. V / U for each band
V discrimination mainly looks at how strongly the spectrum in the band has a harmonic structure.

[0006]

By the way, the start and end points of a strong consonant marked by a noise pulse are
It has a drawback that it cannot be reproduced well. That is, the change part (transient part) of the unvoiced sound is heard as a sound that is too smoothed and band-limited.

Further, not only the unvoiced part but also the unvoiced to voiced part or the voiced to unvoiced part (transient part) lacks time resolution. For example, 8
1 block 256 samples per kHz sampling
The noise pattern of the block 32 msec always has the same shape (only the average level is different), and in particular, consonant sounds such as s, t, p and k become unclear. Further, the reverberation is often strong, or a plurality of speakers may sound as if they are speaking, which adversely affects the synthesized sound quality.

The UV (unvoiced) signal carries the bits used to encode the pitch, which does not carry any information in the unvoiced frame, without being distinguished from the V (voiced) signal.

The present invention has been made in view of the above circumstances, and it is possible to prevent the sound quality from deteriorating at the start point and the end point of a strong consonant accompanied by a noise pulse, and to obtain a synthesized sound that is easy to hear. It is an object of the present invention to provide a coding device and a decoding device for a simple audio signal.

[0010]

In order to solve the above-mentioned problems, an audio signal encoding apparatus according to the present invention divides an input audio signal into blocks on a time axis and encodes each block. In the apparatus for encoding a voice signal for performing a voiced sound or an unvoiced sound for each block, a means for detecting the maximum value in the block of the input signal,
Means for calculating the average level of the small blocks near the detected maximum value and each average level of the preceding and following small blocks, and classifying whether or not it is pulse noise based on the average level of each of these small blocks, and pulse And means for obtaining information indicating the state of sex noise.

Here, it is preferable that, when the unvoiced sound is discriminated, the information indicating the state of the pulse noise is sent instead of the pitch information at the time of the voiced sound discrimination. Further, it is preferable that the voiced / unvoiced sound discriminating means uses at least one of the signal level, the zero-cross count of the signal, and the maximum value of the autocorrelation of the LPC residual for pitch detection.

An example of a speech signal coding apparatus to which the present invention is applied is one in which speech signals are coded by a speech analysis / synthesis method using so-called multiband excitation.

Next, when the maximum value exists in the vicinity of the end position in the block, the maximum value search start position of the next block is set next to the end point of the small block for detecting the average level of the previous block. The position is preferably.

As information indicating the state of the pulse noise, a ratio r1 / of the average rms value r1 of the small block having the maximum value and the average rms value r2 of the next small block of the small block. It is preferable to use r2.

Next, the speech signal decoding apparatus according to the present invention is such that, from the speech signal encoding apparatus, the discrimination information of voiced sound or unvoiced sound is obtained for each block obtained by dividing the input speech signal into blocks on the time axis. The pitch information, the amplitude information of the spectrum, and a decoding device of the audio signal supplied with the information indicating the state of the pulse noise in the block that is determined to be unvoiced, the determination information, the pitch information, A voiced sound synthesizer that synthesizes voiced sound based on amplitude information, the discrimination information, pitch information, an unvoiced sound synthesizer that synthesizes unvoiced sound based on amplitude information, and the above based on information indicating the state of the pulse noise. The above problem is solved by having an unvoiced sound windowing processing unit that performs windowing processing on the unvoiced sound signal from the unvoiced sound synthesis unit.

[0016]

[Function] Unvoiced sound (U
Since the information indicating the pulse noise state of the block determined as V) is transmitted, the pulse noise state can be known on the synthesizing side, and the occurrence of unclearness or reverberation such as consonants. It can be prevented. In addition, since it is not necessary to send pitch information in a block determined to be unvoiced (UV), it is possible to insert information indicating the state of the unvoiced pulse noise into the slot for sending this pitch information and send it. The quality of the reproduced sound (synthesized sound) can be improved without increasing the data transmission amount.

[0017]

Embodiments of the high efficiency coding method according to the present invention will be described below. The high-efficiency coding method is, for example, MBE (Multiband Excitation) coding described later, in which a signal for each block is converted on a frequency axis, divided into a plurality of bands, and V ( Voiced sound) or UV
It is possible to use an encoding method for determining whether or not (unvoiced sound).

That is, as a general high-efficiency coding method to which the present invention is applied, the FFT is performed by dividing the voice signal into blocks for every fixed number of samples (for example, 256 samples).
While converting to spectrum data on the frequency axis by orthogonal transformation such as, to extract the pitch of the voice in the block,
The spectrum on the frequency axis is band-divided at intervals according to this pitch, and V (voiced sound) / U is obtained for each of the divided bands.
V (unvoiced sound) is discriminated. This V / UV discrimination information is encoded and transmitted together with spectrum amplitude data. However, in the embodiment of the present invention, V / U
Various information such as V discrimination information and spectrum amplitude data is transmitted in units of frames (for example, 160 samples) described later.

Here, assuming a speech synthesis analysis system such as an MBE vocoder, the sampling frequency fs for an input speech signal on the time axis is usually 8 kHz,
The total bandwidth is 3.4 kHz (However, the effective bandwidth is 200 to 34
00Hz), and the pitch lag (the number of samples corresponding to the pitch period) from the higher female voice to the lower male voice is 2
It is about 0 to 147. That is, the pitch frequency is 8000
It will vary from / 147≈54 (Hz) to about 8000/20 = 400 (Hz). Therefore, about 8 to 63 pitch pulses (harmonics) stand up to 3.4 kHz on the frequency axis.

In this way, considering that the number of divided bands (the number of bands) for each block (frame) changes between about 8 and 63 when the bands are divided at intervals according to the pitch, It is preferable to reduce or reduce the number of divided bands to a fixed number (for example, about 12).

Further, in the embodiment of the present invention, when all the bands of the block (frame) are determined to be unvoiced (UV), the state of pulse noise (noise pulse) is used instead of the pitch information as described later. The information shown is transmitted, and the reproducing side (decoder) performs windowing processing on the synthesized voiceless signal according to the state information of the pulse noise.

Next, a specific example of a kind of MBE (Multiband Excitation) vocoder of a speech signal synthesis analysis coding apparatus (so-called vocoder) to which the high efficiency coding method as described above can be applied, A description will be given with reference to the drawings.

The MBE vocoder described below is a DW G
riffin and JS Lim, ^ MultibandExcitation Vocode
r, "IEEE Trans.Acoustics, Speech, and Signal Process
ing, vol.36, No.8, pp.1223-1235, Aug. 1988, the conventional PARCOR (PARtial au
In a to-CORrelation (partial autocorrelation) vocoder or the like, a voiced sound section and an unvoiced sound section were switched for each block or frame when modeling a voice, whereas in the MBE vocoder, at the same time (same block or frame). The model is based on the assumption that there are voiced and unvoiced intervals in the frequency domain of (in).

FIG. 1 is a block diagram showing an overall schematic structure of an embodiment in which the present invention is applied to the MBE vocoder. In FIG. 1, an audio signal is supplied to an input terminal 11, and the input audio signal is sent to a filter 12 including an HPF (high-pass filter) and a so-called D signal is sent.
Removal of C (DC) offset and band limitation (for example, 20
At least low frequency components (200 Hz or less) are removed for 0 to 3400 Hz). This filter 12
The signal obtained through the above is sent to the pitch extraction unit 13, the windowing processing unit 14, and the maximum value detection unit 24 described later, respectively.

In the pitch extraction unit 13, the input voice signal data is divided into blocks in units of a predetermined number N (for example, N = 256) (or cut out by a rectangular window), and pitch extraction is performed on voice signals in this block. Is done. Such a cut block (256 samples) is moved in the time axis direction at a frame interval of L samples (for example, L = 160) as shown in FIG. 2A, and the overlap between blocks is N−. There are L samples (for example, 96 samples). In addition, the windowing processing unit 14 applies a predetermined window function, for example, a Hamming window, to one block of N samples, and sequentially moves the windowed block in the time axis direction at intervals of one frame of L samples.

Here, when the windowing processing in the windowing processing section 14 is expressed by a mathematical expression, x _w (k, q) = x (q) w (kL-q) (1) 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 input signal before processing 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 the rectangular window as shown in A of FIG. 2 in the pitch extraction unit 13 is w _r (r) = 1 0 ≦ r <N (2) = 0 r < 0, N ≦ r Further, the window function w _h in the case of Hamming window as shown in B of FIG. 2 in the windowing processing section 14 (r) _{is, w h (r) = 0.54} - 0.46 cos (2πr / (N-1)) 0 ≦ r <N (3) = 0 r <0, N ≦ r. Such a window function w _r (r) or w (1) when using the _h (r) formula of the window function w (r) (= w (KL-
q)) has a non-zero interval, 0 ≦ kL−q <N, and transforms it into kL−N <q ≦ kL. Therefore, for example, in the case of the above rectangular window, the window function w _r (kL−q) = 1
This is when kL−N <q ≦ kL, as shown in FIG. In addition, the above formulas (1) to (3) have a length N
The window of one block of (= 256) samples is L (= 16
0) It shows that the frame is advanced one frame at a time. Hereinafter, the non-zero sample sequence of each N point (0 ≦ r <N) cut out by each window function of the above equations (2) and (3) is
_These are respectively expressed as x _wr (k, r) and x _wh (k, r).

In the windowing processing section 14, as shown in FIG. 4, 179 is applied to the sample sequence x _wh (k, r) of one block of 256 samples to which the Hamming window of the above equation (3) is applied.
Two samples of 0 data are added (so-called zero padding) to form 2048 samples, and the orthogonal transform unit 15 performs orthogonal transform such as FFT (Fast Fourier Transform) on the time-axis data sequence of 2048 samples. Processing is performed. Or, it is FF with 256 points without zero padding
There is also a method of applying T to reduce the processing amount.

The pitch extraction unit 13 performs pitch extraction based on the x _wr (k, r) sample sequence (1 block N samples). The pitch extraction method is known to use periodicity of time waveform, periodic frequency structure of spectrum, autocorrelation function, etc. In this embodiment, the autocorrelation method of center clip waveform is adopted. is doing. For the center clip level in the block at this time,
One clip level may be set for each block, but when the peak level of the signal of each part (each sub-block) obtained by subdividing the block is detected, and the difference in peak level between these sub-blocks is large, In addition, the clip level is changed stepwise or continuously within the block. The pitch period is determined based on the peak position of the autocorrelation data of this 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 of the plurality of peaks is equal to or larger than a predetermined threshold value. In the case of, the maximum peak position is set as the pitch cycle, and in other cases, the pitch is within the 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 ± 2
The peak within the range of 0% is obtained, and the pitch of the current frame is determined based on this peak position. In this pitch extraction unit 13, a relatively rough pitch search is performed by an open loop, and the extracted pitch data is sent to a high precision (fine) pitch search unit 16 and a closed loop high precision pitch search (pitch) is performed. Fine search) is performed.

A method of obtaining the pitch from the autocorrelation of the residual waveform obtained by LPC analysis of the input waveform instead of the center clip waveform may be used.

The high-precision (fine) pitch search unit 16 includes coarse (rough) pitch data of integer (integer) values extracted by the pitch extraction unit 13 and, for example, FFT-processed on the frequency axis by the orthogonal transformation unit 15. And the data of. In this high-precision pitch search unit 16, with the coarse pitch data value as the center, ± several samples are shaken in increments of 0.2 to 0.5 to give an optimum decimal point (floating).
To the value of the fine pitch data of. As a fine search method at this time, so-called synthesis analysis
Using the (Analysis by Synthesis) method, 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
Let S (j) as spectrum data on the frequency axis orthogonally transformed by T etc. be expressed as S (j) = H (j) | E (j) | 0 <j <J (4) It is assumed that the model. Here, J corresponds to ω _s / 4π = f _s / 2, and the sampling frequency f
_{When s} = ω _s / 2π is, for example, 8 kHz, it corresponds to 4 kHz. In the equation (4), when the spectrum data S (j) on the frequency axis has a waveform as shown in FIG.
(j) indicates the spectrum envelope (envelope) of the original spectrum data S (j) as shown in B of FIG.
(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 |.

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 one band is obtained by FFT by regarding a waveform obtained by adding (filling with 0) 1792 samples of 0 data to a Hamming window function of 256 samples as shown in FIG. 4 as a time axis signal. 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 limit point and the upper limit point 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

[0034]

[Equation 1]

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

[0036]

[Equation 2]

Thus, when | A _m | in the equation (6),
Minimize the error ε _m .

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 (5). Next, the sum Σε of all the bands of the error ε _m for each band as described above.
_{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 types of vertical pitches are prepared with the rough pitch determined by the pitch extraction unit 13 as the center, for example, in steps of 0.25. The error sum value Σε _m is obtained for each of these plural kinds of slightly different pitches. In this case, if the pitch is determined, the bandwidth is determined, and from the above formula (6), the power spectrum | S (j) |
(j) | can be used to find the error ε _m in the above equation (5), and the total sum Σε _m of all the bands can be found. This error sum total 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 obtains the optimum fine (for example, 0.25 step) pitch, and the amplitude | A _m | corresponding to this optimum pitch is determined. The calculation of the amplitude value at this time is performed in the voiced sound amplitude evaluation unit 18V.

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), MBE as described above
In the vocoder, unvoiced sound (Un
Since a model in which a voiced) region exists is used, it is necessary to distinguish voiced sound / unvoiced sound for each band.

The optimum pitch from the high precision pitch search section 16 and the amplitude | A from the amplitude evaluation section (voiced sound) 18V
The data of _m | is sent to the voiced sound / unvoiced sound discrimination unit 17,
The voiced sound / unvoiced sound is discriminated for each band. NSR (noise to signal ratio) is used for this determination. That is, the NSR that is the NSR of the m-th band
_m is

[0042]

[Equation 3]

This NSR _m is a predetermined threshold Th.
_{If it} is larger than ₁ (for example, 0.2 or 0.3) (that is, if the error is large), | A _m |
It can be judged that the approximation of | S (j) | by | E (j) | is not good (the above excitation signal | E (j) | is unsuitable as a basis), and the band is regarded as UV (unvoiced). Determine. In other cases, it can be determined that the approximation is performed to a certain extent, and the band is determined to be V (Voiced, voiced sound).

By the way, as described above, the number of bands divided by the fundamental pitch frequency (the number of harmonics) is
The number of V / UV flags for each band similarly varies because the voice varies in the range of about 8 to 63 depending on the pitch (the size of the pitch). Therefore, in this embodiment, the V / UV discrimination results are collected (or degenerated) for each fixed number of bands divided by a fixed frequency band.
I am trying. Specifically, a predetermined band (for example, 0 to 4000 Hz) including a voice band is divided into N _B (for example, 12) bands, and according to the NSR value in each band,
For example, the weighted average value is discriminated by a predetermined threshold value Th ₂ , and the V / UV of the band is judged.

Next, in the unvoiced sound amplitude evaluation unit 18U, frequency-axis data from the orthogonal transformation unit 15, fine pitch data from the pitch search unit 16, and voiced sound amplitude evaluation unit 1 are included.
Data of amplitude | A _m | from 8 V and V / UV (voiced sound / unvoiced sound) discrimination data from the voiced sound / unvoiced sound discrimination unit 17 are supplied. This amplitude evaluation section (unvoiced sound) 1
In 8U, the amplitude is obtained again (amplitude re-evaluation is performed) with respect to the band that is determined to be unvoiced sound (UV) by the voiced sound / unvoiced sound determination unit 17. The amplitude | A _m | _UV for this UV band is

[0046]

[Equation 4]

It is calculated by

The data from the amplitude evaluation unit (unvoiced sound) 18U is sent to the data number conversion (a kind of sampling rate conversion) unit 19. The data number conversion unit 19 has a different number of divided bands on the frequency axis according to the pitch,
This is for keeping the number constant, considering that the number of data (especially the number of amplitude data) is different. That is, for example, when the effective band is set to 3400 kHz, the effective band is divided into 8 bands to 63 bands according to the pitch, and the amplitude | A _m | (UV The number of data m _MX +1 including the band amplitude | A _m | _UV also changes from 8 to 63.
Therefore, in the data number conversion unit 19, this variable number m _MX +
One amplitude data is converted into a fixed number M (for example, 44) of data.

Here, in the present embodiment, for example, for the amplitude data for one block of the effective band on the frequency axis, the values from the last data in the block to the first data in the block are interpolated. After adding the dummy data to increase the number of data to N _F , the band-limited O _S
By multiplying (e.g., 8 times) oversampling, the amplitude data of O _S times is obtained, and the amplitude data of this O _S times ((m _MX +1) × O _S ) are linearly interpolated. Expand to many N _M (eg 2048),
The N _M pieces of data are thinned out and converted into the fixed number M of data (for example, 44 pieces).

The data from the data number conversion unit 19 (the above-mentioned fixed number M of amplitude data) is the vector quantization unit 20.
To a vector, and a predetermined number of pieces of data are combined into a vector, and vector quantization is performed. The quantized output data from the vector quantizer 20 (main part thereof) is the high-precision (fine) pitch data from the high-precision pitch search unit 16 and the voiced / unvoiced sound from the voiced / unvoiced sound discrimination unit 17. It is sent to the encoding unit 21 and encoded together with the (V / UV) discrimination data.

However, the pitch data from the highly accurate pitch search section 16 is sent to the encoding section 21 via the selected terminal a of the changeover switch 28. This is because all bands in the block are UV (unvoiced), as described later.
When the pitch information becomes unnecessary, the information indicating the state of pulse noise (noise pulse) is switched from the pitch information and sent.

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 period. As the V / UV discrimination data from the voiced sound / unvoiced sound discrimination unit 17, as described above, data reduced to about 12 bands (degenerate) may be used as necessary, and 1 in all bands. You may make it use the data showing the division position of a voiced sound (V) area and unvoiced sound (UV) area below a location.

In the encoding unit 21, for example, CR
C addition and rate 1/2 convolutional code addition processing is performed. That is, the pitch data (or the feature extraction data of the time waveform of an unvoiced sound signal described later), the voiced sound /
The unvoiced sound (V / UV) discrimination data and important data in the quantized output data are subjected to CRC error correction coding and then convolutional coding. The encoded output data from the encoding unit 21 is sent to the frame interleaving unit 22, is subjected to interleave processing together with some (for example, low importance) data from the vector quantization unit 20, and is extracted from the output terminal 23. It is transmitted (or recorded / reproduced) to the combining side (decoding side).

Here, the entire block (frame) is UV.
When it is determined to be (unvoiced sound) and the following condition is satisfied, input signal class classification is performed to obtain information indicating the state of pulse noise, which will be described later. (U1) All bands are UV (unvoiced sound) (u2) Signal level is smaller than a predetermined threshold Th _s (u3) Zero cross number is larger than a predetermined threshold Th _z (u4) Autocorrelation coefficient at the time of pitch extraction is predetermined Threshold Th _a
Less than

The condition (u1) above requires that the analysis frame is completely UV (unvoiced sound) by V / UV discrimination. In the above conditions (u2) and (u3), the signal level is low, and the zero-cross count number is the predetermined threshold Th.
requires greater than _z . The above condition (u4)
Requires that the cross-correlation coefficient of the signal be less than _a predetermined threshold Th _a . This correlation coefficient is an autocorrelation value at the lag that gives the maximum autocorrelation of the prediction residual signal obtained as a result of performing LPC analysis of the order of 10 for the input signal for pitch extraction.
This is the maximum value of the residual autocorrelation. The other conditions added to the above condition (u1) are used to reduce the possibility of voiced sounds being erroneously classified as noise pulses and synthesized.

Here, 2500 can be given as a specific example of the signal level threshold value Th _s of the above condition (u2), which is about −20 dB (correctly when assuming full-scale sine wave input and 0 dB). Corresponds to −19.70 dB). Further, the threshold value Th of the number of zero crossings of the above condition (u3)
_As a specific example of _z , 90 can be mentioned, and as _a specific example of the threshold value Th _a of the autocorrelation coefficient of the above condition (u4), 0.47 can be mentioned. The voiced sound / unvoiced sound determination unit 17 determines such conditions.

Next, the class classification for obtaining the information indicating the state of pulse noise will be described. The position of the maximum value sample within one frame is determined. The starting point of the maximum value search changes so as to avoid double detection of pulses in the frame area.

Here, with respect to the position k of the maximum value sample detected in the frame as a center, the following three averages rms (root mean) are calculated for each predetermined region T in the region including the position k and the regions before and after it. square) Calculate the values r0, r1, r2. r0 = RMS {x [k-3T / 2, ..., k-T / 2]} r1 = RMS {x [k-T / 2, ..., k + T / 2]} r2 = RMS {x [k + T / 2,2 ..., k + 3T / 2]} In these equations, x represents an input signal, and when k = i, x
It represents [i] = max {x}. The area T may be about 30 samples, for example.

The average rms value r thus obtained
0, r1, and r2 are used as conditions for pulse classification (classification) as follows. _{(C1) r0 <Th 0 (} c2) 3r0 < the r1 and r1> 2.5r2 (c3) 4r0 < r1 and r1> 1.7r2 (c4) 35r0 < r1 these conditions, (c1) and ((c2) Or (c3)
Or (c4)).

The above condition (c1) serves to reduce erroneous classification of voiced sound as a pulse. Condition (c2) is normal pulse noise (noise pulse)
Is for. The inequality asymmetry is based on the observation that the decay time is usually longer than the attack time. Condition (c3) is for long decay time noise pulses and condition (c4) is for very sharp changes (transients).

Next, the pulse code pc as the pulse noise state information to be actually transmitted is calculated as follows. pc = 0: non-pulse signal or UV not classified
signal. = P: pulse signal, p is a pulse relative peak value. This p is a value between a predetermined value p1 and p2 (p1 ≦ p
≦ p2), and the values of 3 and 8 are selected as p1 and p2, respectively.

Specifically, the average rms values r1 and r2
The ratio r1 / r2 of the above is used, and this ratio r1 / r2 and the above predetermined values p1 and p
Based on the relationship with 2, when r1 / r2 <p1, pc = 0 p1 ≦ r1 / r2 ≦ p2, pc = int [r1 / r2] p2 <r1 / r2, pc = p2 Then, the pulse code pc is obtained. Where int
[X] represents the value of the integer part of the real number x.

Here, in FIG. 1, the maximum value detection unit 24
Detects the maximum value and its position within the frame. The delay means (memory) 25 is for holding the position of the maximum value of the previous frame, and this is because the area T (for example, 30 samples) is centered on the position of the maximum value detected in one frame. Since it is set, if the maximum value exists at the end position of the previous frame, 1 of the next frame
This is because the maximum value search is started from the fifth sample. In the next local RMS calculator 26, the average r
The ms values r0, r1, r2 are calculated. The obtained average rms values r0, r1 and r2 are calculated by the pulse classification unit 27.
The pulse classification unit 27 sends the above conditions (c
Pulse classification is performed based on 1) to (c4), and the above ratio r1 / r
The pulse code pc as the transmission information is obtained according to 2,
It is sent to the selected terminal b of the changeover switch 28.

The pitch data from the high precision pitch search section 16 is sent to the selected terminal a of the changeover switch 28, and the output from the changeover switch 28 is sent to the encoding section 21. The changeover switch 28 is switch-controlled by a discrimination output signal from the voiced sound / unvoiced sound discriminating section 17, so that during normal voiced sound transmission, that is, even if one of all bands in the block is V. When it is determined to be (voiced sound), the selected terminal a is switched, and when all bands in the block are UV (unvoiced sound), the selected terminal b is switched and connected.

Therefore, the pulse code pc, which is the information indicating the state of the pulse noise, that is, the information indicating the state where the peak of the waveform of the unvoiced portion exists (or the information indicating the degree of the peak) is originally the pitch information. It will be transmitted by inserting it into the slot that was transmitting. That is, when all the bands in the block are determined to be UV (unvoiced sound), the pitch information is not necessary, and the U / U discrimination flag from the voiced sound / unvoiced sound determination unit 17 is checked to find that all U
Only when it is V, the pulse code pc is transmitted instead of the pitch information.

Next, FIG. 6 shows a schematic structure of a synthesizing side (decoding side) for synthesizing an audio signal based on the above-mentioned respective data obtained by being transmitted (or recorded / reproduced).
Will be described with reference to.

In FIG. 6, a data signal at the input terminal 31 is substantially equal to the data signal taken out from the output terminal 23 on the encoder side shown in FIG. 1 (ignoring signal deterioration due to transmission and recording / reproduction). Is supplied. The data from the input terminal 31 is sent to the frame deinterleave unit 32, subjected to deinterleave processing which is the reverse processing of the interleave processing of FIG. 1 above, and subjected to CRC and convolutional coding on the main portion (encoder side). In the portion, generally, the data portion having high importance) is subjected to decoding processing by the decoding unit 33, and the remaining portion (data of low importance not subjected to the encoding processing) is directly subjected to bad frame mask processing in the bad frame mask processing unit 34. To the section 34. In the decoding unit 33, for example, so-called Viterbi decoding processing or error detection processing using a CRC check code is performed. Bad frame mask processing unit 34
Performs processing such as obtaining a parameter of a frame with many errors by interpolation, and also performs the above pitch data, voiced sound /
Unvoiced sound (V / UV) data and vector-quantized amplitude data are separated and extracted.

The vector-quantized amplitude data from the bad frame mask processing unit 34 is sent to the inverse vector quantization unit 35 and inversely quantized, and the data number inverse conversion unit 3
It is sent to 6 and is inversely transformed. This data number inverse conversion unit 36
Then, the inverse conversion, which is in contrast to the above-described data number conversion unit 19 of FIG.
And the unvoiced sound synthesizer 38. The pitch data from the bad frame mask processing unit 34 is sent to the voiced sound synthesis unit 37 and the unvoiced sound synthesis unit 38. The V / UV discrimination data from the bad frame mask processing unit 34 is also sent to the voiced sound synthesis unit 37 and the unvoiced sound synthesis unit 38.

Here, the pitch data (data in the pitch information transmission slot) is information indicating the state of the above-described noise pulse (pulse noise) when all bands in the block are UV (unvoiced sound). Since a certain pulse code pc is sent instead of the pitch information, it is sent to the changeover switch 51, and the output from the selected terminal a of this changeover switch 51 is the voiced sound synthesizing unit 3
7, the output from the selected terminal b is sent to the UV windowing processing unit 52 at the subsequent stage of the unvoiced sound synthesizing unit 38. The configuration and operation of the UV windowing processing unit 52 will be described later.

In the voiced sound synthesizer 37, for example, cosine
The voiced sound waveform on the time axis is synthesized by wave synthesis or sine wave synthesis, and the unvoiced sound synthesis unit 38 synthesizes the unvoiced sound waveform on the time axis by filtering white noise with a bandpass filter, for example. The voiced voice synthesized waveform and the unvoiced voice synthesized waveform are added and synthesized by the adding section 41 and taken out from the output terminal 42. In this case, the amplitude data, pitch data and V / UV discrimination data are
The value is updated and given for each frame (L sample, for example, 160 samples) at the time of the analysis, but in order to enhance (smooth) the continuity between frames, each value of the amplitude data and the pitch data is changed by one frame. For example, each data value at the center position in the inside is calculated, and each data value up to the center position of the next frame (one frame at the time of composition) is obtained by interpolation. That is, in one frame at the time of composition (for example, from the center of the above analysis frame to the center of the next analysis frame), each data value at the leading end sample point and each data value at the end (leading edge of the next composition frame) sample point Are given, and each data value between these sample points is obtained by interpolation.

Further, the above V / UV is used as V / UV discrimination data.
When the UV code is transmitted, all bands can be divided into a voiced sound (V) region and an unvoiced sound (UV) region at one division position according to this V / UV code. Accordingly, V / UV discrimination data for each band can be obtained. If the analysis side (encoder side) reduces (degenerates) the number of bands to a certain number (for example, about 12), it is solved (restored) and changed at intervals according to the original pitch. Of course, the number of bands is set.

The synthesis process in the voiced sound synthesizer 37 will be described in detail below. M-th discriminated as V (voiced sound)
1 on the time axis in the band (band of the mth harmonic)
When a voiced sound for a synthetic frame (L samples, for example 160 samples) is V _m (n), V _m (n) = A _m (n) using the time index (sample number) n in this synthetic frame ) cos (θ _m (n)) 0 ≦ n <L (9) The final voiced sound V (n) is synthesized by adding (ΣV _m (n)) the voiced sounds of all bands that are determined to be V (voiced sound) of all bands.

A _m (n) in the equation (9) 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 next composite frame) is A _Lm, and the following equation is used: A _m (n) = (Ln) A _0m / L + nA _Lm / L (10) It suffices to calculate A _m (n).

Next, the phase θ _m (n) in the above equation (9) is calculated by θ _m (n) = mω _O1 n + n ² m (ω _L1 −ω ₀₁ ) / 2L + φ _{0 m} + Δω n (11) You can ask. In the equation (11), φ _0m represents the phase of the m-th harmonic (frame initial phase) at the tip (n = 0) of the composite frame, and ω ₀₁ is at the tip of the composite frame (n = 0). The fundamental angular frequency, ω _L1, indicates the fundamental angular frequency at the end of the composite frame (n = L: leading end of the next composite frame). Δω in the above equation (11) 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 transmitted by the above equation (10). The amplitude A _m (n) may be calculated by linearly interpolating A _0m and A _Lm . Phase θ _m (n)
From n 0, θ _m (0) = φ _{0 m} to n = L, θ _m (L) is φ
Set Δω to be _Lm .

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 n = L, when V (voiced sound) is used, the 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 obtained at 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 ... (12) 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 formula (11), 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 (13). In this equation (13), 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.

Next, the unvoiced sound synthesizing process in the unvoiced sound synthesizing section 38 will be described below. The white noise signal waveform on the time axis from the white noise generation unit 43 is sent to the windowing processing unit 44, and windowed by a suitable window function (for example, Hamming window) with a predetermined length (for example, 256 samples),
The STFT (Short Term Fourier Transform) process is performed by the STFT processing unit 45 to obtain the power spectrum of the white noise on the frequency axis. The power spectrum from the STFT processing unit 45 is used as the band amplitude processing unit 4
6 is multiplied by the amplitude | A _m | _UV for the band designated as UV (unvoiced sound), and the amplitude of the other band designated as V (voiced sound) is set to zero. This band amplitude processing unit 46
The amplitude data, pitch data, and V / UV discrimination data are supplied to.

The output from the band amplitude processing unit 46 is IS
The signal is sent to the TFT processing unit 47, 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 47 is sent to the overlap adding unit 48, and the overlap and addition are repeated while appropriate weighting (so that the original continuous noise waveform can be restored) on the time axis,
Synthesize continuous time axis waveforms. The output signal from the overlap adding section 48 is sent to the adding section 41 as an output from the unvoiced sound synthesizing section 38 via a UV windowing processing section 52 described later.

In this way, the signals of the voiced sound part and the unvoiced sound part which are combined in the combining parts 37 and 38 and returned on the time axis are added by the adding part 41 at an appropriate fixed mixing ratio, The reproduced audio signal is taken out from the output terminal 42.

At least one of the bands in the frame is V
In the case of (voiced sound), the above-described processing is performed in each of the synthesizing units 37 and 38. However, when it is determined that all bands in the frame are UV (unvoiced), the changeover switch 51 is used. Is switched and connected to the selected terminal b side, and the information (pulse code pc) regarding the pulse noise of the unvoiced signal is sent to the UV windowing processing unit 52 instead of the pitch information.

That is, the UV windowing processing unit 52 has the pulse code pc from the pulse classification unit 27 of FIG.
Is supplied. In the UV windowing processing unit 52, as a window function W [i] for pulse generation, as shown in FIG. 7, W [i] = (i / 11) ² 0 ≦ i ≦ 11 = ((23 -i) / 11) ² 12 ≦ i ≦ 22 = 0 23 ≦ i In this window function W [i], i is a frame (for example, 1
The sampling frequency fs is 8 kHz.

Further, the pulse gain g _p by which the window function W [i] is multiplied is g _p = 0 when voiced sound (at least one band is voiced sound), and when unvoiced sound (all bands are unvoiced sound), Let g _p = pc. The output signal S _W [i] that has been subjected to the UV windowing process is S _W [i] = S _UV [i] × (1 + g _p ×), where S _UV [i] is the output signal from the unvoiced sound synthesizer 38. W [i]). FIG. 8 is a graph showing the multiplication coefficient (1 + g _p × W [i]) for the unvoiced sound signal S _UV [i] in this equation.

The unvoiced sound signal synthesized and windowed as described above is taken out from the UV windowing processing section 52, sent to the addition section 41, and added with the signal from the voiced sound synthesis section 37. The final audio signal is taken out from the output terminal 42.

Regarding the configuration on the speech analysis side (encoding side) in FIG. 1 and the configuration on the speech synthesis side (decoding side) in FIG. 6, although each unit is described as hardware, a so-called DSP (digital 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-mentioned embodiment, and for example, regarding the configuration of the voice analysis side (encoding side) of FIG. 1 and the configuration of the voice synthesis side (decoding side) of FIG. Although each unit is described as hardware, it can be realized by a software program using a so-called DSP (digital signal processor) or the like. Further, the high-efficiency coding method to which the present invention is applied is not limited to the voice analysis / synthesis method using the above-mentioned multiband excitation, and the sine wave synthesis is used for the voiced sound portion or the unvoiced sound portion is subjected to noise. It can be applied to various voice analysis / synthesis methods such as synthesis based on signals, and is not limited to transmission or recording / reproduction, but can be applied to various uses such as pitch conversion, speed conversion, or noise suppression. Of course.

[0089]

As is apparent from the above description, according to the audio signal encoding apparatus of the present invention, the input audio signal is divided into blocks on the time axis and encoded in each block. In the audio signal encoding device, it is determined whether voiced sound or unvoiced sound is included in each block, the maximum value in the block of the input signal is detected, and the average level of small blocks near the detected maximum value and the front and rear Calculated as each average level of the small blocks, to classify whether or not it is pulse noise based on the average level of each of these small blocks, and to obtain and transmit information indicating the state of pulse noise, The state of this pulse noise can be known on the synthesis side, and the occurrence of unclearness or reverberation such as consonants can be prevented in advance. In addition, since it is not necessary to send pitch information in a block determined to be unvoiced (UV), it is possible to insert information indicating the state of the unvoiced pulse noise into the slot for sending this pitch information and send it. , Playback sound (synthesized sound) without increasing the amount of data transmission
Can improve the quality of.

Further, according to the audio signal decoding apparatus of the present invention, it is determined from the audio signal encoding apparatus whether the input audio signal is divided into blocks on the time axis, that is, voiced sound or unvoiced sound. Information, pitch information, spectrum amplitude information, and a decoding device of a voice signal to which information indicating the state of pulse noise in a block that is determined to be unvoiced is supplied, wherein the determination information, pitch information , The voiced sound and the unvoiced sound are respectively synthesized based on the amplitude information, and since the unvoiced sound signal from the unvoiced sound synthesis unit is subjected to the windowing process based on the information indicating the state of the pulse noise, the unvoiced sound is generated on the synthesis side. It is possible to reproduce the state of pulsed noise in the generated block, and improve the reproducibility of strong consonants such as s, t, p, and k.

[Brief description of drawings]

FIG. 1 is a functional block diagram showing a schematic configuration on an analysis side of a speech signal analysis / synthesis encoding apparatus as an embodiment of a speech signal encoding apparatus according to the present invention.

FIG. 2 is a diagram for explaining a windowing process.

FIG. 3 is a diagram for explaining a relationship between windowing processing and a window function.

FIG. 4 is a diagram showing time axis data as an object of orthogonal transform (FFT) processing.

FIG. 5 is a diagram showing spectrum data on a frequency axis, a spectrum envelope (envelope), and a power spectrum of an excitation signal.

FIG. 6 is a functional block diagram showing a schematic configuration on the synthesizing side of a speech signal analysis / synthesis coding apparatus as an embodiment of a speech signal decoding apparatus according to the present invention.

FIG. 7 is a graph showing an example of a window function used in the UV windowing processing section.

FIG. 8 is a graph showing an example of a signal that is multiplied by an unvoiced sound signal in the UV windowing processing unit.

[Explanation of symbols]

13 Pitch extraction unit 14 Window processing unit 15 Orthogonal transform (FFT) unit 16 High precision (fine) pitch search unit 17 Voiced / unvoiced (V / UV) discriminating unit 18V: Voiced sound amplitude evaluation unit 18U: Unvoiced amplitude evaluation unit 19: Data number conversion (data rate conversion) unit 20, 27 ... Vector quantizer 24 ... Maximum value detector 26 ... Local RMS calculator 27 ... Pulse classifier 28, 51. Switch 37: Voiced sound synthesis section 38: Unvoiced sound synthesis section 41: Addition section 43: White noise generation section 44: Windowing processing section 46. .... Band amplitude processing unit 48 ... Overlap addition unit 52 ··· UV windowing processing unit

Claims (7)

[Claims] 1. An audio signal encoding apparatus for dividing an input audio signal into blocks on a time axis and encoding each block, and means for determining whether each block is voiced or unvoiced. Means for detecting the maximum value of the input signal in the block, means for calculating the average level of the small blocks near the detected maximum value and each average level of the preceding and following small blocks, and And a means for classifying whether or not it is pulse noise based on the average level and obtaining information indicating the state of pulse noise. 2. The code of the voice signal according to claim 1, wherein the information indicating the state of the pulse noise is sent in place of the pitch information at the time of the voiced sound determination when the voiceless sound is determined. Device. 3. The voiced sound / unvoiced sound discrimination means is at least one of the maximum value of the signal level, the number of zero-cross counts of the signal, and the LPC residual autocorrelation for pitch detection.
3. The method according to claim 1, wherein the determination is performed using one of the two.
An audio signal encoding device as described. 4. The speech signal coding apparatus according to claim 1, wherein the speech signal is coded by a speech analysis / synthesis method using multi-band excitation. 5. When the maximum value exists near the end position in the block, the maximum value search start position of the next block is set to the position next to the end point of the small block for detecting the average level of the previous block. The audio signal encoding device according to claim 1, 2, 3, or 4, wherein the position is a position. 6. The average rm of the small blocks having the maximum value as information indicating the state of the pulse noise.
s value r1 and the average rm of the next small blocks of the small block
6. The audio signal coding apparatus according to claim 1, wherein a ratio r1 / r2 to the s value r2 is used. 7. A voice signal encoding device, which blocks an input voice signal on a time axis, for each block, discriminates whether voiced sound or unvoiced sound, pitch information, spectrum amplitude information, and unvoiced sound. A device for decoding a voice signal supplied with information indicating a state of pulsed noise in a block determined to be a voiced sound synthesis for synthesizing a voiced sound based on the above determination information, pitch information, and amplitude information. Section, an unvoiced sound synthesizing section for synthesizing unvoiced sound based on the discrimination information, the pitch information, and the amplitude information, and a windowing process for the unvoiced sound signal from the unvoiced sound synthesizing section based on the information indicating the state of the pulse noise. A voice signal decoding device, comprising: an unvoiced windowing processing unit.