JPH051958B2 - - Google Patents

Description

[Detailed Description of the Invention] (Technical Field) The present invention provides a privacy device for polarizing voice communication;
In particular, CSM (Composite Sinusoidal Mode−
The present invention relates to a confidential communication device based on the principle of ling (composite sine wave model), and relates to an analog confidential communication device with high communication confidentiality.

(Prior art) In general, as a secret communication device, an audio signal is converted into a digital code by A/D conversion, the original audio information is transmitted in a confidential manner by performing specific code conversion, etc., and the receiving side converts the original audio information into a reverse code. Digital confidential communication devices that reproduce the original audio signal by D/A conversion after conversion are widely used. However, such a digital system has the drawback that high requirements are placed on the transmission quality of the transmission path, such as transmission capacity and error rate.

On the other hand, there are analog secret communication devices that use methods such as inverting the spectrum of the audio signal, or dividing it and switching the relative positions before transmitting it, but these generally do not meet the requirements for the transmission quality of the transmission path. However, since the spectral envelope of the original audio signal remains in some form, it generally has the disadvantage of low privacy.

(Object of the Invention) An object of the present invention is to provide a new type of analog confidential speech device with high speech confidentiality based on the principles of speech analysis and speech synthesis using CSM, which will be described later.

(Structure of the Invention) The apparatus of the present invention includes: a CSM analysis means; a means for converting the distribution range of CSM frequency data calculated by the means at a constant ratio;
Generates an additive composite waveform of multiple sine waves with the frequency and amplitude specified by CSM data
a transmitting side having a CSM spectrum generating means; and a spectrum for performing spectrum analysis of a signal from the CSM spectrum generating means on the transmitting side and extracting each parameter specifying the frequency and amplitude of each of the plurality of sine waves that have been added and synthesized. parameter extracting means; means for inversely transforming the extracted parameters on the transmitting side; and a plurality of variable frequency oscillators with phase reset functions set to respective frequencies specified by the inversely transformed parameters. and a variable gain amplifier, a variable length window function generator, and a random number generator for correspondingly setting the output of each variable frequency oscillator with a phase reset function to an amplitude specified by the extracted parameter, When synthesizing voiced sounds, the phase of each variable frequency oscillator with a phase reset function is reset in accordance with the pitch period, and when synthesizing unvoiced sounds, the phase is reset according to the pitch period calculated from the random number output from the random number generator. Correspondingly, the phase of each variable frequency oscillator with a phase reset function is reset so that the start and end points of the window function generated by the variable length window function generator almost coincide with the phase reset point. and a receiving side.

(Principle) As mentioned above, the secret speech device of the present invention is based on the principle of CSM speech analysis and speech synthesis.

First, we will explain the principles of CSM's speech analysis and speech synthesis.

CSM expresses an audio signal as the sum of a specific number of sine waves whose amplitude and frequency are freely selectable parameters. A predetermined number of 4 to 6 sine waves is used at most.

Therefore, when performing CSM speech synthesis, it is first necessary to express the speech signal as a sum of a predetermined number of sine waves by CSM speech analysis.
CSM voice analysis will be explained in detail later.
Only the main points will be explained here.

In CSM analysis, as in LPC analysis,
An autocorrelation coefficient is used as an intermediate parameter for the purpose of ignoring phase information, averaging the influence of sound sources, and avoiding instability due to noise components.

In other words, in CSM analysis, N low-order taps of sample autocorrelation coefficients directly calculated from the speech waveform to be expressed for each analysis frame are combined with N low-order taps of autocorrelation coefficients of a synthesized wave. The purpose is to determine the frequency of each sine wave to be synthesized and its strength (power amplitude) so that it matches the individual sine waves.

Now, if the number of sine waves to be combined is n, each frequency of each sine wave is ω _i (i=1, 2,...n), and the intensity of each sine wave is m _i , then the CSM composite wave yt is yt＝ _o 〓 ⁱ⁼¹ √sin(ωit＋φi), but the autocorrelation coefficient γ _l of this tap l is ωi,
It is easily expressed using mi, and γ _l = _o 〓 ⁱ⁼¹ micoslωi.

On the other hand, the sample of the audio waveform to be expressed is
If X _t is the sample autocorrelation coefficient υ _l of tap l in a certain frame, it is given as υ _l =1/M _M-1 〓 ^t=l X _t X _tl .

However, M is the number of samples in one analysis frame.

Now, in the CSM analysis, the value of each m _i and ω _i is determined so that the above-mentioned γ _l is equal to the given υ _l for N low-order values.

That is, γ _l =υ _l : However, the values of m _i and ω _i are determined so that l=0, 1, 2, . . . N holds true.

This specific method will be explained in detail later, but here, m _i and ω _i of the n sine waves mentioned above are shall be obtained.

FIG. 1 shows an example of a speech feature vector pattern based on the CSM parameters m _i and ω _i obtained in this way.

In addition, the 9th order (N = 9) CSM (number of sine waves n =
5) The second example shows the correspondence between the line spectrum and the 9th-order LPC spectrum envelope (frequency transmission characteristics of the LPC synthesis filter) obtained from the same audio sample.
As shown in the figure.

Note that there is a relationship of N=2n-1 between the above-mentioned order N and the number n of sine waves, as described later.

From these figures, it can be seen that the CSM contains information that extracts the features of the original speech that should be expressed.

However, using n sets of m _{i and} ω _i values obtained as a result of CSM analysis, the intensity specified by m _i and ω _i (the actual amplitude is √ _i as mentioned above)
If you create n sine waves with angular frequencies and simply add and synthesize them, the human ear will hear:
It has the characteristic that it can only be heard as a synthesized sound of sine waves and cannot be heard as the original sound at all.

This means that even if you simply add sine waves, the spectrum of the generated signal is just a discretized n line spectrum, whereas the spectrum of the audio signal has a continuous spectral envelope, and In addition, voiced sounds have a pitch structure, while unvoiced sounds have a fine spectral structure that is represented by a stochastic process. This is because the spectral structure of a simply added CSM and a speech signal is completely different. Conceivable.

Therefore, in order to synthesize audible speech using CSM, it is necessary to spread the line spectrum into a continuous spectrum using some special method. In other words, what is CSM speech synthesis?
The purpose of this method is to diffuse a voice characteristic spectrum pattern expressed by a line spectrum as shown in FIGS. 1 and 2 using a special method, and generate a voice spectrum pattern from it. If such CSM speech synthesis is not performed, the simple summation synthesized waveform of multiple sine waves with the frequency and amplitude specified by CSM will contain the information necessary to reproduce the original speech in the most basic way. Even though it is present in the form, it cannot be heard as a sound at all.

The secret speech device of the present invention utilizes the above-mentioned characteristics of the CSM simple addition synthesis waveform.

In other words, the input audio signal is
A CSM analysis is performed, and an analog signal of a simple summation composite waveform of a plurality of sine waves having the angular frequency and amplitude specified by the CSM is generated and sent to the transmission line. As mentioned above, although this synthesized waveform contains the information necessary to reproduce speech in the most basic form, it has a highly confidential nature in that it cannot be heard as speech at all. Also,
CSM if specifically required, as detailed below.
The confidentiality can be further improved by applying a predetermined specific transformation to the parameters of .

Now, on the receiving side, special CSM speech synthesis is performed to reproduce the original speech. CSM using spread spectrum on the receiving side of the secure communication device of the present invention
The speech synthesis method is as follows.

That is, since voiced sounds have a clear pitch structure, the phase of each of the n sine waves of the CSM specified as described above is reset every pitch period. This makes it possible to easily generate a spectral envelope and a fine pitch spectral structure.

Furthermore, by applying special time window function processing to the above-mentioned phase reset waveform as detailed in the explanation of the embodiment, discontinuity of the synthesized waveform at the time of phase reset is removed and continuity of the audio waveform is ensured. are doing.

Through the above implementation, the CSM line spectrum shown in Figure 2 is diffused as shown in Figure 3A, and changed to a spectrum having a spectral envelope and pitch fine structure, which is sufficiently practical for auditory purposes. Experimental results have shown that durable sound quality can be obtained.

For reference, the CSM spectrum obtained by simple addition without performing the above processing is shown in Figure 3B.
Shown below. As mentioned above, this is the spectrum of the signal transmitted via the transmission path, and a waveform with such a spectrum reproduces the original sound to the extent that it sounds just like a synthesized sine wave. It is not possible.

The above is for voiced sounds, but in the case of unvoiced sounds, it is performed as follows. In other words, in the case of voiced sounds mentioned above, the phase reset and special time window function processing performed every pitch period are used, and in the case of unvoiced sounds, pulses that are randomly generated as a stochastic process are used instead of the pitch periods. The process is performed every time this pulse occurs.

By using the above method, it is possible to perform CSM synthesis that is auditorily sufficient for practical use.

Note that the above CSM synthesis is a synthesis method that does not use a filter, and therefore does not require consideration of stability on the synthesis side. Therefore, when using a communication means such as this device, which transmits the information of m _i and ω _i to the synthesis side and reproduces the voice on the synthesis side, when the line quality is relatively poor, it is better than a vocoder. Another possible feature is that it provides good sound quality.

However, as mentioned above, simply adding sine waves will not be perceived as speech. The confidential communication device of the present invention is based on such a principle, and further improves communication confidentiality by means of converting the distribution range of the sine wave.

(Example) Next, the present invention will be explained in detail using examples.

FIG. 4 is a block diagram showing one embodiment of the present invention.

This embodiment consists of a transmitting side 1 and a receiving side 2.

The transmitting side 1 further includes an A/D converter 101, a Hamming window processor 102, and an autocorrelation coefficient measuring device 10.
3. CSM analyzer 104, pitch/V/UV analyzer 105, parameter converter 106, n variable frequency oscillators 107-1 to 107-n, n variable gain amplifiers 108-1 to 108-n, addition It includes a combiner 109, a variable gain amplifier 110, a variable frequency converter 111, a V/UV switch 112, and a summing combiner 113.

Further, the receiving side 2 further includes a spectrum analyzer 2.
01, power separator 202, parameter inverter 2
03, n variable frequency oscillators with phase reset function 204-1 to 204-n, n variable gain amplifiers 205-1 to 205-n, addition synthesizer 206,
Multiplier 207, multiplier 208, V/UV switch 2
09, variable length window function generator 210, period calculator 2
11 and a random number generator 212.

Now, the operation of this embodiment is as follows. The audio waveform to be transmitted is fed via an input line 1000 to an A/D converter 101, where it is converted into digital data whose amplitude and time axis are quantized, and the outputs of which are each passed through a Hamming window processor. 102, and Pituchi V/UV analyzer 10
5.

The digital data supplied to the Hamming window processor 102 is subjected to weight multiplication using a known Hamming window function for each predetermined frame, and is supplied to the autocorrelation coefficient measuring device 103 for each frame of data. Ru.

The autocorrelation coefficient measuring device 103 calculates the lowest N autocorrelation coefficients υ _l (where l=
1, 2,...N).

In other words, data for one frame is X _t (where t
=0, 1, ..., M-1), each of N υ _l is obtained by performing the calculation process υ _l =1/M _M-1 〓 ^t=l X _t X _tl .

The set of υ _l for each frame obtained in this way is supplied to the next CSM analyzer 104, and υ ₀ in this (that is, υ ₀ = 1/M _M-1 〓 ^t=0 X _2/t ) parameter converter 106 as the power information in this frame.
supply to.

Now, the autocorrelation coefficient υ _l for each frame mentioned above
The CSM analyzer 104, which has been supplied with the set m _i , ω _i ( However, i=1, 2,...n)
and supplies this to the parameter converter 106.

Also, the pitch V/UV which receives the digital data of the original audio signal from the A/D converter 101
The analyzer 105 extracts information regarding the pitch period and voiced (V)/unvoiced (UV) sounds and supplies it to the parameter converter 106.

Now, the parameter converter 106 is a converter that converts the parameters of this information and outputs it, but to make the explanation easier to understand, initially this converter does not perform any conversion and converts the input signal. It shall be output as is.

Thus, the n pieces of angular frequency information ω _i obtained by the CSM analyzer 104 are supplied to the variable frequency oscillators 107-1 to 107-n via the converter 106, and specify the output angular frequencies of these oscillators, respectively. set to ω _i . In addition, the CSM analyzer 104
The n pieces of intensity information m _i obtained in
8-n as gain control information, and sets the output of each of the oscillators 107-1 to 107-n to a designated intensity _mi .

Thus, the output of the adder synthesizer 109 has the CSM
A waveform is obtained by simply adding and synthesizing a plurality of sine waves having the amplitude and frequency specified by .

The above composite waveform is further applied to the variable gain amplifier 110.
After the total power is controlled to be proportional to the power υ ₀ obtained by the measuring device 103, the combiner 113
supplied to

On the other hand, the pitch period information obtained by the analyzer 105 is supplied to a variable frequency oscillator 111, and the frequency of this oscillator 111 is controlled so as to become the fundamental frequency of a given pitch. Furthermore, the analyzer 105
Voiced/unvoiced (V/UV) information from V/UV
The output of the oscillator 111 is supplied to the synthesizer 113 in the case of a voiced sound (V), and is controlled to be turned off in the case of an unvoiced sound (UV).

The transmission line 12 thus connected to the combiner 113
00, the summation and synthesis waveform of the pitch information and each sine wave of the power-controlled CSM is transmitted in the form of an analog signal, but as mentioned above, even if this signal is converted directly into audio, it cannot be heard as audio. It has a confidential nature in that it cannot be taken.

Now, on the receiving side 2, the signal transmitted in this way is subjected to spectrum analysis by a spectrum analyzer 201, and each angular frequency information ω _i specified by the line spectrum of the CSM and each intensity weighted by the total power are Information m _i ' is obtained, and the latter is separated in the power separator 202 into total power information υ ₀ and intensity information m _i of each sine wave of the CSM.

Furthermore, pitch period information and V/UV information are extracted in the spectrum analyzer 201, and each of the above information is supplied to a parameter inverse converter 203.

Now, the parameter inverse converter 203 is a converter that performs the inverse transformation of the parameter converter 106 on the transmitting side described above, but in order to make the explanation easier to understand as before, first, the parameter converter 1 on the transmitting side
Corresponding to the case where the parameter inverse converter 203 outputs the input signal as it is without performing any conversion, it is assumed that the parameter inverse transformer 203 also outputs the input signal as it is.

Thus, ω _i (ω ₁
~ _ωo ) is added to the frequency control input of the n variable frequency oscillators with phase reset function 204-1 to 204-n, and changes the output angular frequency of these oscillators to the specified angular frequency _ω1 to _ωo. Set to .

Also, the intensity (power amplitude) of each of the n waves of CSM
m ₁ to _{m o} specifying are supplied to the gain control terminals of the n variable gain amplifiers 205-1 to 205-n, so that the oscillation power of each frequency becomes the specified value. Control.

The n outputs obtained in this way are
After addition and synthesis are performed in step 06, the signals are supplied to the next multiplier 207.

Now, the pitch period data and V/UV information (voiced sound/unvoiced sound information) extracted by the spectrum analyzer 201 are sent to the parameter inverse converter 20, respectively.
3 to the V/UV switch 209.

On the other hand, the random number generated by the random number generator 212 is
It is supplied to the period calculator 211, where it is converted so that the random number distribution width and its lower limit become specific values, and is supplied to the V/UV switch 209 as a data string that determines the phase reset time interval during unvoiced sound. be done.

Thus, when the above-mentioned V/UV information of the spectrum analyzer 201 specifies unvoiced sound (V), the switch 209 selects the above-mentioned pitch period data side and sends it to the variable length window function generator 210. supply Furthermore, if the aforementioned V/UV information specifies unvoiced sound (UV), the switch 209 selects the data string side representing a random time interval generated in the stochastic process of the output of the aforementioned period calculator 211. , is supplied to the window function generator 210 instead of the digital data string representing the pitch period described above.

Now, the window function generator 210 is for generating a window function that ensures the continuity of the audio waveform except for the discontinuity that occurs in the output waveform due to the phase reset, and furthermore, the window function A related phase reset pulse is also generated.

As described above, a data string specifying the interval between the next phase reset pulse is input to the window function generator 210 via the switch 209. impulses having a time interval specified by , are supplied via line 2100 to the phase reset terminals of variable frequency oscillators with phase reset function 204-1 to 204-n, thereby Performs a phase reset of the oscillator.

Now, the window function generator 210 generates a variable length window function ω(x) as shown below in synchronization with the generation of the above-mentioned phase reset pulse.

That is, if T is the value of the interval between phase reset pulses at that point specified by the input data, and x is the elapsed time since the previous phase reset pulse was generated, then ω(x) = 0.5. +0.5cos(πx/T) However, a window function expressed as 0<xT is generated. This window function ω(x) is shown in FIG. 5A. The value of T mentioned above is
In the case of voiced sounds, it represents the pitch period, and in the case of unvoiced sounds, it represents variables that occur in stochastic processes, so
In either case, it changes over time. Therefore, this window function ω(x) has a variable length and is synchronized with the generation of the phase reset pulse described above in the relative time relationship shown in FIG. 5B (the start and end points of the window function are (This almost coincides with the time point at which the phase reset pulse is generated.)

The window function thus generated is shown on line 2200
is supplied to multiplier 207 via. As a result, the multiplier 207 generates n sine waveforms whose phase is reset for each phase reset pulse synthesized by the adder synthesizer 206, and the above-mentioned window function generated in synchronization with each phase reset pulse. ω
(x) is obtained. The waveform obtained in this way is such that each sine wave is continuously converged to 0 by multiplication by the window function ω(x) just before the phase is reset, and each sine wave rises from 0 at the time of the phase reset, so the waveform is The continuity of the phase reset waveform is ensured, and thus the discontinuity that occurs in the phase reset waveform can be removed by multiplication by the window function ω(x).

The output of the multiplier 207 from which the discontinuity has been removed is supplied to the next multiplier 208, where the power separator 2
It is weighted by the power information υ ₀ of each frame sent from the transmitter 1, separated by 02, and output on line 2000 as synthesized speech.

As explained above, on the receiving side 2 of this embodiment, the CSM speech synthesis necessary for the above-mentioned speech synthesis is executed, and as a result, the original input to the transmitting side 1, which was hidden on the transmission path 1200, is Sound reproduction will be performed with good sound quality.

In the above description, the parameter converter 106 on the transmitting side 1 and the parameter inverse converter 203 on the receiving side 2 output the input data as is to the output side and do not perform any conversion. Of course, even in this case, confidentiality on the transmission path is maintained as described above.

That is, it is also possible to configure a confidential communication device by completely omitting the parameter converter 106 on the transmitting side 1 and the parameter inverse converter 203 on the receiving side 2.

However, in order to obtain a higher degree of confidentiality,
If the parameter converter 106 performs parameter conversion, and the parameter inverse converter 203 performs inverse conversion, each data input to the parameter converter 106 can be obtained at the output side of the parameter inverse converter 203. good.

As an example of parameter conversion, the distribution range of CSM frequency data is converted at a constant ratio. A simple transformation such as ω′ _i =a·ω _i +θ (i=1, 2, . . . N) can be used, where a and θ are constants. Note that when 0<a<1, there is also an audio band compression transmission effect. Conversion on the receiving side can be performed as ω _i =(ω′ _i −θ)/a (i=1, 2, . . . N).

Furthermore, in the embodiment described above, the pitch frequency information is transmitted from the transmitting side 1, but instead, the transmission of the pitch frequency information from the transmitting side 1 may be omitted as follows. can.

In other words, by utilizing the property of voice that the pitch frequency increases when the voice energy is large and the pitch frequency decreases when the voice energy is small, we empirically created a correspondence table between voice energy and pitch frequency. Create. Then, the receiving side 2 is provided with a pseudo-pitch period generating means that generates a pitch period from the total audio power information transmitted from the transmitting side 1 using this correspondence table, and the pseudo-pitch period generated thereby is used on the receiving side 2. In this way, transmission of pitch frequency information from the transmitting side can be omitted.

Next, variable frequency oscillator 2 with phase reset function
An example of the circuit of No. 04 is shown in FIG. Frequency control terminal 2
041, the constant current power supply 204
2 and 2043, and thereby the oscillation frequency is made variable. The oscillation voltage waveform at the υ point is a triangular waveform that linearly rises and falls between the reference voltage +Vr and -Vr.When an impulse is applied to the phase reset terminal 2045, the υ point is momentarily grounded and forced to The potential is pulled back to 0, and oscillation is restarted from there to perform a phase reset. The triangular wave oscillation output at the υ point is input to a sine wave converter 2046, converted to a sine wave, outputted from a terminal 2047, and used as the output of the oscillator 204. The sine wave converter 2046 can be easily realized, for example, by reading out a sine function value stored in a ROM using an input waveform.

Further, such a variable frequency oscillator with a phase reset function can be easily realized using a computer program.

Next, a circuit example of the variable gain amplifier 205 is shown in FIG. By applying a signal to be amplified to a terminal 2051 and a control signal to a terminal 2052, the amount of negative feedback is controlled, and an output having a controlled amplitude is obtained at an output terminal 2053.

In addition, it can also be realized by using an analog multiplier, or by using an analog waveform input as the reference voltage of the D/A converter and using control information expressed in digital quantities as the digital input. This can also be easily realized by

Variable gain amplifiers 108 and 110 can be implemented in a similar manner.

Next, an example of a circuit of the random number generator 212 is shown in FIG. The 15-stage shift register 2121 and one half adder 2112 have a period of 2 ¹⁵ -1.
Generate the next M series of pseudo-random numbers. The next random value is obtained by applying a shift pulse to clock terminal 2123 at the required time.

Next, the block diagram of the period calculator 211 is shown in FIG. 9A.
Shown below. This is suitable for using the random numbers uniformly distributed in the range of 0 to 2 ¹⁵ -1 output from the random number generator 212 described above as random numbers for specifying the time interval of the phase reset pulse during unvoiced speech. It is made up of a constant multiplier 2121 and a constant adder 2122. Thereby, as shown in FIG. 9B, the random number distribution width D and the lower limit L can be set to appropriate values.

Next, an embodiment of the window function generator 210 is shown in FIG. This includes a register 2101, a presettable down counter 2102, and a counter 210.
3. Contains read-only memory (ROM) 2104. Data T specifying the phase reset pulse interval supplied from the switch 209 is stored in the register 2101. Down counter 2102
is a counter that counts the high-speed clock CLK of a constant period. First, the content T of the register 2101 is preset, and this is counted down using the clock CLK. The content of counter 2102 is 0
When this happens, a pulse is generated from the output terminal, which presets the contents of the register 2101 again and starts counting down this value. Thus, the output 2102-1 of the down counter 2102 has T.
A pulse train having a period proportional to (for example, T/k) is generated. This pulse train is added as a clock to counter 2103. Count output 2103 of counter 2103 incremented by this clock
-1 is added to the ROM2104 as an address designation signal, and the window function ω(x) written there
data are read out in order and output on line 2200. When the contents of the counter 2103 reach k,
The last data in ROM 2104 is read and outputs a reset pulse on line 2100. This reset pulse is generated by the oscillators 204-1 to 204-n.
It is used as the aforementioned phase reset pulse supplied to the phase reset terminal of the register 2101, and is also used to set the next input data in the register 2101. In this way, phase reset pulses whose interpulse intervals have successively specified values and a variable length window function ω(x) synchronized with these pulses as shown in FIG. 5B are generated.

Finally, I will explain CSM analysis.

As mentioned above, CSM analysis uses, for each analysis frame, N low-order tap values of sample autocorrelation coefficients calculated directly from the speech waveform to be represented;
Determine the angular frequency ω _i of each sine wave to be synthesized and its intensity (power amplitude) m _i so that the N low-order tap values of the synthesized wave (sum of n sine waves) match. It is to be.

Now, if the autocorrelation coefficient of tap l of the composite wave is γ _l , then as mentioned above, γ _l = _o 〓 ⁱ⁼¹ m _i coslω _i .

On the other hand, the sample of the audio waveform to be represented X _t
Therefore, the sample autocorrelation coefficient υ _l of tap l in a certain frame is υ _l =1/M _M-1 〓 ^t=l X _t X _tl (1).

From this, γ _l =υ _l …(2) l=0, 1, 2, …N where N=2n−
If it is set to 1, the following matrix expression will be obtained.

However, the above equation cannot be solved by simple matrix operations because ω _i and m _i are unknown. Therefore, we set w _i = cos ^-1 x _i ...(4) and perform the substitution coslw _i = coslcos ^-1 x _i ≡Tl(x _i ) ...(5). This Tl(x) is a Tchebycheff polynomial. When this substitution is performed, equation (3) is converted as follows.

However, in general, x ^l is T ₀ (x), T ₁ (x)...T _l
It can be expressed as a linear combination of (x).

That is, x ^l = _l 〓 ^j=0 S ^(l) _j Tj (x) ...(7) where S ^(l) _j is the inverse Tchebycheii coefficient.

Using this S ^(l) _j , the sample autocorrelation coefficient υ _j
Define the linear combination A _l as shown below.

A _l = _l 〓 ^j=0 S ^(l) _j υ _j …(8) However, l=0, 1, 2, …, 2n−1 Then, on the left and right sides of equation (6), respectively
By using the relationships of equations (7) and (8), the following relational expression is established.

_Now , here, the nth degree polynomial Pn( _x )≡ _o 〓 _K ⁼⁰ p ⁽ⁿ⁾ _k x ^k = _o Π ⁱ⁼¹ (x− x _i ), and using this Pn(x), create _o 〓 ⁱ⁼¹ m _i Pn(x _i ) x ^l _i , which is similar to the left side of equation (9), and examine this. It is clear that the above formula is 0, but it can be further changed as follows.

0 = _o 〓 ⁱ⁼¹ m _i Pn(x _i ) x ^l _i = _o 〓 ⁱ⁼¹ m _io 〓 ^K=0 p ⁽ⁿ⁾ _k x ^k+l _i = _o 〓 ^K=0 p ⁽ⁿ⁾ _ko 〓 ⁱ⁼¹ m _i x ^k+l _i = _o 〓 ^K=0 p ⁽ⁿ⁾ _k A _k+l From the above, the following formula is obtained with l=0, 1, 2,...n.

However, since p ⁽ⁿ⁾ _o = 1 holds true. The matrix formed by A _i on the left side is generally called the Hankel matrix. As described above, each A _i is given by equation (8) from the sample autocorrelation coefficient υ _j of the speech waveform to be expressed and is known.

Therefore, by solving equation (10), p ⁽ⁿ⁾ ₀ , p ⁽ⁿ⁾ ₁ , ...
The value of p ⁽ⁿ⁾ _o-1 can be found.

When each p ⁽ⁿ⁾ _i is found, the solution to the n-dimensional equation P _o(x) = x ⁿ + p ⁽ⁿ⁾ _o-1 x ^n-1 +... p ⁽ⁿ⁾ ₀ = 0 is {x ₁ , x ₂ ..., x _o } is found.

As a result, each CSM frequency ω _i is obtained from ω _i =cos ⁻¹ x _i in equation (4), and the CSM intensity m _i is obtained using the following equation derived from equation (9).

Note that the matrix on the left side of the above equation is generally Van der
This is called the Mon-de (Juander Monde) matrix.

To summarize the above, the analysis algorithm for CSM analysis is as follows.

(1) Calculate the sample autocorrelation coefficient.

υ _l =1/M _M-1 〓 ^t=l X _t X _tl (2) Define A _l using the inverse Tievisiev coefficient.

A _l = _l 〓 ^j=0 S ^(l) _j υ _j (3) Solve the Hankel matrix equation by A _l to find p ⁽ⁿ⁾ _i .

(4) p ⁽ⁿ⁾ _Find n x _i by solving an n-dimensional algebraic equation with i as a coefficient.

p _o(x) ≡x ⁿ +p ⁽ⁿ⁾ _o-1 x ^n-1 +p ⁽ⁿ⁾ _o-2 x ^n-2 +……+ p ⁽ⁿ⁾ ₁ x+p ₀ =0 (5) Cos inverse transformation Then, the CSM angular frequency {ω _i } is obtained.

ω _i =cos ^-1 x _i (6) Solve the Vander Monde matrix equation to find the CSM intensity {m _i }.

By performing each of the above steps, CSM
Each angular frequency {ω ₁ , ω ₂ ...ω _o } and the intensity of each wave {m ₁ , m ₂ , ... m _o } can be determined.

Note that, as an efficient method for solving the above-mentioned Hankel matrix equation, a method is known in which initial conditions are given and solutions are sequentially obtained.

Furthermore, since it has been proven that the above nth-order algebraic equation has only real roots, the roots can be found using the Newton-Raphson method or the like.

Furthermore, as an efficient method for solving the Van der Monde matrix equation, a method can be used in which triangular matrixing is performed and solutions are sequentially obtained.

In the CSM analysis described above, this example uses a method of solving an equation in which the sample autocorrelation coefficient and the CSM autocorrelation coefficient are equal. A method based on line spectral frequency calculation and residue calculation can also be used.

Further, in this embodiment, a variable length window function having a specific function form is used, but this function form is merely an example, and it is clear that other function forms may be used.

Furthermore, the random number generator, period calculator, etc. are also shown as examples, and there is no need to be limited thereto.

In the above embodiments, m _i was used as the quantity specifying the intensity (power amplitude) of each sine wave of the CSM, but as a signal that actually controls the variable gain amplifier, √ _i which is proportional to the amplitude is used. It is clear that this can also be done using

(Effects of the Invention) As described above, by using the present invention, it is possible to provide a new type of confidential communication device that transmits analog signals based on the principle of speech analysis and synthesis using CSM. This privacy device is characterized by having relatively low requirements for the quality of the transmission path and being able to provide high privacy.

[Brief explanation of the drawing]

Figure 1 is a diagram showing an example of a speech feature vector pattern based on CSM parameters, Figure 2 is a diagram showing an example of the correspondence between a CSM line spectrum and an LPC spectrum envelope obtained from the same audio sample, and Figure 3A is a diagram showing an example of a speech feature vector pattern based on CSM parameters. FIG. 3B is a diagram showing the CSM spectrum obtained by simple addition. FIG. 4 is a block diagram showing an embodiment of the present invention. FIG. 5A is a diagram showing the functional form of the variable length window function, FIG. 5B is a diagram showing the relative time relationship between the variable length window function and the phase reset pulse, and FIG. 6 is a circuit of a variable frequency oscillator with a phase reset function. 7 is a diagram showing an example of a variable gain amplifier circuit, FIG. 8 is a diagram showing an example of a random number generator circuit, FIG. 9A is a block diagram of a period calculator, FIG. B is a diagram showing the random number distribution of the output of the period calculator, and FIG. 10 is a block diagram showing an example of a variable length window function generator. In the figure, 1...sending side, 2... receiving side, 1
01... A/D converter, 102... Hamming window processor, 103... Autocorrelation coefficient measuring device, 104...
...CSM analyzer, 105...Pitzchi V/UV analyzer, 106...Parameter converter, 107-1~
107-n...Variable frequency oscillator, 108-1~
108-n...variable gain amplifier, 109...addition combiner, 110...variable gain amplifier, 111...
Variable frequency oscillator, 112... V/UV switch, 113... Addition combiner, 201... Spectrum analyzer, 202... Power separator, 203... Parameter inverter, 204-1 to 204-n...
Variable frequency oscillator with phase reset function, 205-
1 to 205-n...variable gain amplifier, 206...
Addition synthesizer, 207... Multiplier, 208... Multiplier, 209... V/UV switch, 210... Variable length window function generator, 211... Period calculator, 212
...Random number generator.

Claims (1)

[Claims] 1. In a privacy device that privatizes voice communication, a CSM analysis means and a data calculated by the means are provided.
means for converting the distribution range at a constant ratio only for CSM frequency data; and a CSM spectrum generator for generating an additive composite waveform of a plurality of sine waves having a frequency and amplitude specified by the converted CSM data. spectral parameter extraction means for performing spectral analysis of the signal from the transmitting side and extracting each parameter specifying the frequency and amplitude of each of the plurality of sine waves that have been added and synthesized; a plurality of variable frequency oscillators with phase reset functions set to respective frequencies designated by the inversely transformed parameters; A variable gain amplifier that sets the output of each variable frequency oscillator with a phase reset function to an amplitude specified by the extracted parameter, a variable length window function generator, and a random number generator, and is used when synthesizing voiced sound. resets the phase of each of the variable frequency oscillators with a phase reset function in accordance with the pitch period, and when synthesizing unvoiced sounds, sets the distribution width and lower limit calculated from the random numbers output from the random number generator. The position of each variable frequency oscillator with a phase reset function is reset in accordance with the cycle, and the variable length window function generated by the variable length window function generator has a starting point of 1 and an ending point of 0 and has continuity. a receiving side, the start and end times of which substantially coincide with the time of the phase reset.