JPH0552511B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to a voice transient point detection method suitable for use in voice recognition.

BACKGROUND TECHNOLOGY AND PROBLEMS In speech recognition, methods based on word recognition for specific speakers have already been put into practical use. This involves having a specific speaker pronounce all the words to be recognized, and then detecting and storing (registering) the acoustic parameters using a bandpass filter bank, etc.
I'll keep it. Then, when a specific speaker utters a utterance, its acoustic parameters are detected and compared with the acoustic parameters of each registered word, and when these match, the word is recognized.

In such a device, if the time axis of the speaker's utterance is different from the time of registration, the time axis of the speaker's utterance is different from the time of registration,
The time series of acoustic parameters extracted every ~20 msec) is expanded or contracted to align the time axis. This makes it possible to cope with variations in speaking speed.

However, with this device, the entire acoustic parameters of every word to be recognized must be registered and stored in advance.
Requires huge storage capacity and calculations. For this reason, there was a limit to the number of words that could be recognized.

On the other hand, it has been proposed to perform recognition in units of phonemes (in Japanese, A, I, U, E, O, K, S, T, etc. when written in Roman letters) or syllables (KA, KI, KU, etc.). ing.

However, the system registers discretely pronounced sounds for each syllable and performs recognition by aligning the discretely pronounced sounds on the time axis in the same way as word recognition. It could only be used for specific purposes.

Furthermore, when recognizing unspecified speakers,
There is a large variance in acoustic parameters due to individual differences, and recognition cannot be achieved only by matching the time axis as described above. Therefore, for example, methods have been proposed such as registering multiple acoustic parameters for one word and recognizing approximate acoustic parameters, or converting the entire word into fixed-dimensional parameters and discriminating using a discrimination function. Either method requires a huge amount of storage capacity, a large amount of calculation, and the number of words to be recognized becomes extremely small.

In response to this, the inventor of the present application has previously proposed a new speech recognition method that allows speech recognition to be easily and reliably performed even for unspecified speakers. Let's first explain one example below.

By the way, when observing the phenomenon of phoneme production, phonemes such as vowels and fricatives (S, H, etc.) can be elongated and uttered. For example, when considering the utterance of "yes", the phoneme changes to "silence→H→A→I→silence" as shown in FIG. 1A. In response, the same "yes" can be uttered as shown in FIG. 1B. Here, the lengths of the quasi-stationary portions of H, A, and I change with each utterance, which causes fluctuations in the time axis. However, in this case, it has been found that there is relatively little variation in the time axis in the transitional part between each phoneme (indicated by diagonal lines).

Therefore, in FIG. 2, the audio signal supplied to the microphone 1 is passed through the microphone amplifier 2, the low-pass filter 3 of 5.5 kHz or less, and then passed through the A-D converter circuit 4.
is supplied to Also, from the clock generator 5
A sampling clock of 12.5kHz (80μsec interval) is supplied to the AD conversion circuit 4, and at this timing, each audio signal has a predetermined number of bits (=1 word).
is converted into a digital signal. This converted audio signal is supplied to a register 6 of 5×64 words. In addition, a frame clock with an interval of 5.12 msec from the clock generator 5 is supplied to the quinary counter 7, and this count value is supplied to the register 6 to shift the audio signal by 64 words. A signal is taken from register 6.

4 x 64 = 256 taken out from this register 6
The word signal is supplied to a fast Fourier transform (FFT) circuit 8. Here, in this FET circuit 8, for example, if the waveform function represented by n _f sampling data included in the time length of T is U _ofT (f) ...(1), this is Fourier transformed. Then, the following signal is obtained: U _ofT (f)=∫ ^T/2 _-T/2 U _ofT (f)e ^-2 〓 ^jft dt ≡U _1ofT (f)＋jU _2ofT (f) ……(2).

Furthermore, the signal from this FET circuit 8 is supplied to a power spectrum detection circuit 9, and a power spectrum signal of |U ² |=U ² _1ofT (f)+U ² _2ofT (f) (3) is extracted. Here, since the Fourier-transformed signal is symmetrical on the frequency axis, half of the n _f data extracted by Fourier transformation is redundant data. Therefore, half of the data is removed and 1/2n _f pieces of data are extracted. That is, the 256-word signal supplied to the FET circuit 8 described above is converted to extract a 128-word power spectrum signal.

This power spectrum signal is supplied to an emphasis circuit 10 and weighted to perform auditory correction. Here, as the weighting, for example, correction is performed to enhance high frequency components.

This weighted signal is supplied to a band division circuit 11, and is divided into, for example, 32 bands according to a frequency mel scale matched to auditory characteristics. Here, if the dividing point of the power spectrum is different, the signal is divided into each band in proportion and a signal corresponding to the amount of signal in each band is extracted. As a result, the 128-word power spectrum signal described above is compressed into 32 words while preserving the acoustic characteristics.

This signal is supplied to a logarithm circuit 12 and converted into a logarithm value of each signal. This eliminates redundancy due to weighting or the like in the above-mentioned emphasis circuit 10. Here, this logarithmic power spectrum log｜U ² _ofT (f)｜ ...(4) is the spectrum parameter x _(i) (i=0, 1...31)
It is called.

This spectral parameter x _(i) is supplied to a discrete Fourier transform (DFT) circuit 13 . here,
In this DFT circuit 13, for example, if the number of divided bands is M, this M-dimensional spectral parameter x _(i) (i=0, 1...M-1) is 2M-1
Considering the real symmetric parameters of the points, 2M−2 points
Perform DFT. _{Therefore ,} ^_ _{_} ^_ _{_} ^_ _{_} 2) m=0, 1, ..., 2M-3. Furthermore, since the function that performs this DFT is considered to be an even function, W ^mi _2M-2 = cos (2π・i・m/2M−2) = cosπ・i・m/M−1, and from these, X _(n) = _2M-3 〓 ⁱ⁼⁰ x _(i) cosπ・i・m/M−1 ...(6). This DFT extracts acoustic parameters that express the envelope characteristics of the spectrum.

Regarding the spectrum parameter x _(i) DFTed in this way, 0 to P-1 (for example, P = 8)
If we extract the values of the P dimensions up to the next and set them as local parameters L _(p) = (P = 0, 1, ..., P-1), then L _(p) = _2M-3 〓 ⁱ⁼⁰ x _{(i )} cosπ・i・p/M−1 ……(7), and considering that the spectral parameters are symmetric, set x _(i) = x _(2M-i-1) ……(8) Then, the local parameter L _(p) is L _(p) = x _(p) + _M-2 〓 〓 ⁱ⁼⁰ x _(i) {cosπ・i・p/M−1+cosπ・(2M−2
-i)・p/M−1}+x(M−1)cosπ・p/M−
1...(9) However, p=0, 1,..., P-1. In this way, a 32 word signal is compressed into P (for example 8) words.

This local parameter L _(p) is the memory device 14
supplied to This memory device 14 has a memory section of P words per row arranged in a matrix of 16 rows, for example, and stores local parameters L _(p) sequentially for each dimension.
A frame clock with an interval of 5.12 msec is supplied from the frame clock, and the parameters of each row are sequentially shifted in the horizontal direction. As a result, the memory device 14
P-dimensional local parameter L _(p) at 5.12msec intervals
are stored for 16 frames (81.92 msec), and updated to new parameters at every frame clock.

Further, for example, a signal from the emphasis circuit 10 is supplied to a speech transition point detection circuit 20 to detect transition points between phonemes.

This transient point detection signal T _(t) is supplied to the memory device 14, and reading from the memory device 14 is performed when the local parameter L _(p) corresponding to the timing of this detection signal is shifted to the 8th row. .
Here, when reading out the memory device 14, signals for 16 frames are read out in the horizontal direction for each dimension P. The read signal is then supplied to the DFT circuit 15.

In this circuit 15, DFT is performed in the same manner as described above, and the envelope characteristics of the time-series changes in the acoustic parameters are extracted. 0 to 0 from this DFT signal
The values of the Q dimension up to the Q-1 (for example, Q=3) order are extracted. This DFT is performed for each dimension P, and the entire transition point parameter K _(p,q) of P×Q (=24) words is
(p=0,1,...,P-1) (q=0,1,...
..., Q-1) are formed. Here, since K _(0,0) expresses the power of the audio waveform, q may be set to 1 to Q when p=0 for power normalization.

That is, in FIG. 3, when a transient point like B is detected for an input audio signal (HAI) like A, the entire power spectrum of this signal is like C. For example, if the power spectrum at the transition point of "H→A" is as shown in D, this signal is emphasized to become as shown in E, and compressed using the mel scale as shown in F. This signal is subjected to DFT to become something like G, and the previous and following 16 frames are matrixed like H, and this signal is sequentially moved in the time axis t direction.
DFT is performed to form transient point parameters K _(p,q) .

This transition point parameter K _{(p, q)} is supplied to the Mahalanobis distance calculation circuit 16, and the cluster coefficients from the memory device 17 are supplied to the circuit 16 to calculate the Mahalanobis distance with each cluster coefficient. Here, the cluster coefficients are obtained by extracting transient point parameters from the pronunciations of multiple speakers in the same manner as described above, classifying them according to phoneme content, and performing statistical analysis.

The calculated Mahalanobis distance is then supplied to the determination circuit 18, which determines which phoneme to which phoneme the detected transition point is a transition point, and outputs it to the output terminal 19.

For example, “Yes”, “No”, “0 (zero)”
Regarding the 12 words of ~9 (Kiyuu), the voices of a large number of speakers (more than 100 people) are supplied in advance to the above-mentioned device, the transition point is detected, and the transition point parameter is extracted. These transient point parameters are classified into a table as shown in Figure 4, and this classification (cluster)
Perform statistical analysis for each. * in the figure indicates silence.

For these transient point parameters, any sample R ^(a) _r,s (r=1,2...24) (a is a cluster index, for example, a=1 is *→H, a=2 is H→A) n is the speaker number), the covariance matrix A ^(a) _r,s ≡E(R ^(a) _r,o − _r ^(a) ) (R ^(a) _s,o − _s ^(a
) )...(15) However, _r ^(a) = E (R ^(a) _r,o ) E counts the ensemble average, and this inverse matrix B ^(a) _r,s = (A ^(a) _{t, o} ) ^-1 _r,s ...(16) is found.

Here, the distance between any transient point parameter K _r and cluster a is Mahalanobis distance D (K _r,a )≡d 〓 ^r 〓 ^s (K _r − _r ^(a) )・B ^(a) _r,s (K _r − _s ^(a) )
...(17) is obtained.

Therefore, the above-mentioned B ^(a) _r,s and _r ^(a) are stored in the memory device 17.
By determining and storing the above, the Mahalanobis distance calculation circuit 16 calculates the Mahalanobis distance between the input voice and the transition point parameter.

As a result, the minimum distance to each cluster and the ranking of the transition points are extracted from the circuit 16 for each transition point of the input audio. These are supplied to the determination circuit 18,
Recognition determination is made when the input voice becomes silent. For example, for each word, the word distance is determined by the average value of the square root of the minimum distance between each transition point parameter and the cluster. In addition, taking into account the dropout of some of the transition points, word distances are calculated for multiple types assuming that each word is dropped. However, if the ranking relationship of the transition points is different from the table, it will be rejected.
Then, the word with the minimum word distance is recognized and determined.

Therefore, this device detects changes in phoneme at transition points in speech, so there is no change in the time axis.
It is possible to perform good recognition for non-specific speakers.

In addition, by extracting the parameters described above at the transition point, one transition point can be
It can be recognized in 24 dimensions, making recognition extremely easy and accurate.

Furthermore, as a result of learning using the above-mentioned device with 120 speakers and conducting experiments on the above-mentioned 12 words with speakers other than these 120, an average recognition rate of 98.2% was obtained.

Furthermore, in the above example, “H → A” of “Yes” and “8
“H→A” of “(Hachi)” can be classified into the same cluster. Therefore, if αP is the number of phonemes of the language to be recognized and approximately ₂ clusters are calculated in advance and the cluster coefficients are stored in the memory device 17, this method can be applied to the recognition of various words, and can be used to recognize many words. can be easily recognized.

The present invention relates to an audio transition point detection method suitable for use in the detection circuit 20 in such an apparatus.

By the way, as a conventional method of detecting a transient point, for example, there is a method of using the sum of the amount of change in the acoustic parameter L _(p) . In other words, when P-order parameters are extracted for each frame, and when the parameters of G frames are L _(p) (G) (p=0, 1...P-1), T (G) = _{p- 1} 〓 ^p=0 L _(p) (G) - L _(p) (G - 1) | ...(9') Detection is performed using the sum of the absolute values of the differences.

Here, for example, when P=one dimension, a peak of the parameter T _(G) is obtained at a change point of the parameter L _(p) (G), as shown in FIGS. 5A and 5B.

Note that in the above description, L _(p) (G) is a continuous quantity, but in reality, this parameter L _(p) (G) is a discrete quantity. However, such voice recognition devices analyze predetermined M frames at a time to obtain one-dimensional parameters, and the parameter values cannot keep up with sudden changes. It was almost impossible to detect the transition point to the plosive consonant that occurs.

OBJECTS OF THE INVENTION In view of the above-mentioned problems, it is an object of the present invention to provide a method for detecting a vocal transition point that can effectively detect a transition point from silence to a plosive consonant.

Summary of the Invention The present invention extracts acoustic parameters by equally weighting input audio signals according to human auditory characteristics.
In an audio transient point detection method that normalizes the level of this acoustic parameter, monitors this normalized acoustic parameter over multiple frames, and detects the peak of the acoustic parameter, The transition point parameter for gender consonants and the transition point parameter for speech from silence to sounds other than plosive consonants are separately obtained, and then a transition point signal is obtained from their levels.

Embodiment Hereinafter, an embodiment of the speech recognition apparatus of the present invention will be described with reference to FIG. In FIG. 6, parts corresponding to those in FIG. 2 are given the same reference numerals, and detailed explanation thereof will be omitted.

In FIG. 6, the emphasis circuit 1 of FIG.
The weighted signal from 0 is sent to the band division circuit 21
The signal V _(o) (n=0, 1...
N-1) is taken out. This signal is supplied to the biased logarithm circuit 22 to form v' _(o) =log(V _(o) +B)...(10). Further, the signal V _(o) is supplied to the accumulator circuit 23 to form V _a = ₂₀ 〓 ⁿ⁼¹ V _(o) /20, and this signal V _a is supplied to the logarithm circuit 22 to form v' _a = log (V _a +B) ...(11) is formed. These signals are then supplied to the arithmetic circuit 24 to form v _(o) = v' _a - v' _(o) . . . (12).

Here, by using the signal V _(o) as described above, this signal has the same degree of change for each order (n=0, 1...N-1) with respect to the change from phoneme to phoneme, It is possible to avoid variations in the amount of change depending on the type of phoneme. Further, by forming the normalized parameter v _(o) by taking a logarithm and performing an operation, fluctuations in the parameter v _(o) due to changes in the level of input audio are eliminated. Furthermore, by adding bias B and performing calculations, if B → ∞, the parameter
As is clear from the fact that v _(o) → 0, the sensitivity to minute components (noise, etc.) of the input voice can be lowered.

This parameter v _(o) is supplied to the memory device 25 and stored for 2w+1 (for example, 9 frames, assuming w=4). This stored signal is supplied to the arithmetic circuit 26, and Y _o,t =min I∈GFN {v _(o) (I)} ...(13) However, GF _N = {I; -w+t≦I≦ w+t} is formed, and this signal and parameter v _(o) are supplied to the arithmetic circuit 27. T _1(t) = _N-1 〓 ⁿ⁼⁰ _W 〓 ^I=-w (v _(o) (I+t)−Y _o,t ) ...(14) is formed. This T _1(t) is the first transient point detection parameter, and this first transient point detection parameter T _1(t) is supplied to the first peak detection circuit 28 to detect a predetermined value of the input audio signal. The transition point signal of the transition point of the phoneme is supplied to the adder circuit.

The parameter v _(o) is also supplied to the memory device 25', and 2(w-a)+1 (for example, 5, where 0<a=2) frames are stored. This stored signal is supplied to the arithmetic circuit 26', and Y _o,t =min I∈GFN {v _(o) (I)} ...(13') However, GF _N = {I;-(w -a)+t≦I≦(w-
a) +t} is formed, and this signal and parameter v _(o) are supplied to the arithmetic circuit 27', T _2(t) = _N-1 〓 ⁿ⁼⁰ _Wa 〓 ^I=-(wa) (v _{( o)} (I+t)-Y _o,t ) ...(14') is formed. Here, the size of a is set so that the parameter value changes sharply even at the transition point from silence to a plosive consonant, and the transition point can be detected from the parameter value. This T _2(t) is the second
The transient point detection parameter T _2(t) is supplied to the second peak detection circuit 28' to detect a transient point associated with a sudden change in the input audio signal, for example from silence to a plosive. Thus, when the switch circuit 29 is on, a transition point signal is supplied to the addition circuit 30, which transmits to the addition circuit 30 that a transition point has been detected. Here, the switch circuit 29 is
Set it to “ON” the next time. That is, the second transient point detection parameter from the arithmetic circuit 27'
When T _2(t) exceeds a predetermined threshold set in the level detection circuit 31, the mono multivibrator 32 sends a signal for a predetermined length (for example, the length of 5 frames).
A control pulse of a predetermined length is generated, and the switch circuit 29 is turned on while this control pulse of a predetermined length is supplied to the switch circuit 29. And the second transient point detection parameter
When T _2(t) exceeds a predetermined threshold and a transition point signal is supplied from the second peak detection circuit 28' to the switch circuit 29 while the switch circuit 29 is "ON", the transition point The signal is added to the adder circuit 30
It will be supplied as is. Therefore, the first
Regarding the transition point from silence to plosive consonant, for which a transition point signal is not normally generated well from the peak detection circuit 28 of the second peak detection circuit 28', the transition point signal from the second peak detection circuit 28' is sent to the output terminal 33 via the addition circuit 30. It will be taken out from.

Note that other parts have the same configuration as in FIG. 2.

According to this embodiment configured in this way, multiple sets of acoustic parameters are monitored over a predetermined set of multiple frames, and the first and second transient point detection parameters T _{1 (t)} and T _{2 (t)} and then obtain the transient point signal from those levels. Therefore, for the transition point from silence to plosive consonant, the transient point signal from the second peak detection circuit 28' is sent to the output terminal 33. This has the advantage of allowing better detection of the transition point from silence to plosive consonant.

Note that the present invention is not limited to the above-described embodiments, and it goes without saying that various other configurations can be made without departing from the gist of the present invention.

Effects of the Invention As described above, according to the speech transition point detection method of the present invention, since multiple sets of acoustic parameters are monitored over a plurality of predetermined sets of frames, a transition point from silence to a plosive consonant can be detected. There is a benefit in being able to do this well.

[Brief explanation of the drawing]

Figures 1 to 4 are diagrams for explaining the speech recognition device, Figure 5 is a diagram for explaining transient point detection, and Figure 6 is a diagram for explaining the transient point detection.
The figure is a system diagram of an example of the audio transient point detection method of the present invention. 1 is a microphone, 3 is a low pass filter,
4 is an AD conversion circuit, 5 is a clock generator, 6 is a register, 7 is a counter, 8 is a fast Fourier transform circuit, 9 is a power spectrum detection circuit, 10 is an emphasis circuit, 21 is a band division circuit, 22 is a logarithmic circuit, 23, 24, 26, 27 are arithmetic circuits, 25
is a memory device, 28 and 28' are peak detection circuits,
29 is a switch circuit, 30 is an adder circuit, 31 is a level detection circuit, 32 is a mono multivibrator,
33 is an output terminal.

Claims (1)

[Claims] 1. Acoustic parameters are extracted by weighting the input audio signal equally according to human auditory characteristics, the level of this acoustic parameter is normalized, and the normalized acoustic parameters are In a voice transition point detection method that monitors frames over frames and detects the peak of the above-mentioned acoustic parameters, a transition point parameter from silence to a plosive consonant and a transition point parameter from silence to a voice other than a plosive consonant are detected. This is an audio transient point detection method that obtains the levels separately and then obtains the transient point signal from those levels.