JPH052159B2 - - Google Patents

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a codec conversion device, and in particular to a codec conversion device.
The present invention relates to a codec conversion device that converts the code structure of an LPC (Linear Prediction Coding) vocoder type communication network and a multi-pulse coding type communication network in order to interface with each other.

[Conventional technology]

Conventionally, when connecting two codecs (CODECs) that use different encoding methods, a method has been used in which the audio waveforms are first converted into analog audio waveforms and then encoded using different encoding methods.

Even in voice communication between an LPC vocoder type communication network and a multipulse coding type communication network, the codes of both codecs are first converted into analog audio waveforms and then converted into the code of the other party. It is operated by this method.

The LPC vocoder-based communication network includes feature parameters of the input audio signal extracted by the LPC vocoder, ie, spectral envelope information as LPC such as α parameter and K parameter,
The sound source information as a residual signal obtained by removing this spectrum information from the input audio signal is converted into a code in a predetermined format and used on the transmitting side (analysis side) instead of the input audio signal.
to the receiving side (synthesizing side), and the receiving side uses these characteristic parameters to set LPC as the coefficient of a digital speech synthesis filter, and then reproduces the input audio signal by driving this with sound source information. This is the basic operating method. The above-mentioned sound source information is expressed by three elements: the pitch period of the input audio signal, whether it is voiced or unvoiced, and the strength of the sound source.In the case of an LPC vocoder, the pitch period is usually the number of repetitions corresponding to the number of extracted pitches. , and unvoiced sounds are expressed by replacing them with white noise.

In addition, in the multipulse encoding communication network, the sound source information to be sent together with the spectral envelope information in the LPC vocoder communication network is expressed by multiple impulse sequences, so-called multipulses, whose amplitude and temporal position are free. In this case, this multi-pulse is set so that the synthesized waveform synthesized using this multi-pulse as a driving sound source and the input audio signal most closely match.

These LPC vocoder-based communication networks and multi-pulse coding system communication networks both use normal coding systems, such as PCM (Pulse
It is possible to communicate at a much lower data bit rate than other methods such as Code Modulation, which makes it possible to greatly improve the efficiency of communication lines, and it also has the characteristic that it is essentially easy to conceal communication. It is being used frequently these days.

These communication networks do not require code conversion when communicating through codecs of the same type, such as when communicating between LPC vocoder communication networks, but when communicating through codecs of different types. If you do
Code conversion is required when communicating between an LPC vocoder communication network and a multipulse coding communication system.

Conventionally, when connecting different types of codecs, the transmitted and received content encoded in each format is first converted to the original audio waveform, and then this analog audio waveform is converted to a digital amount in the other party's encoding format. We are using this method.

[Problem that the invention seeks to solve]

However, the conventional codec conversion device described above has the following drawbacks.

In other words, according to the conventional method, when connecting different types of codecs, it is first converted to the audio waveform level, so deterioration of communication quality due to audio waveform conversion processing is unavoidable, and the processing The drawback is that an increase in the scale of hardware is inevitable.

The purpose of the present invention is to eliminate the above-mentioned drawbacks, to fundamentally eliminate the increase in deterioration of communication quality due to audio waveform conversion processing by connecting different types of codecs through direct conversion of feature parameters, and to The object of the present invention is to provide a codec conversion device that requires a small-scale hardware configuration.

[Means for solving problems]

The apparatus of the present invention is a codec conversion apparatus that converts the code structure of each communication network in order to interface an LPC vocoder-based communication network and a multi-pulse coding system communication network. pitch period determination means for determining the pitch period of a sound source from a multi-pulse train received and coded by a signal from an encoding system communication network; and an LPC synthesis filter of the LPC vocoder communication network to be excited from the multi-pulse train. excitation source amplitude determining means for determining excitation pulse amplitude or excitation noise amplitude;
Based on the first-order autocorrelation obtained by decoding the encoded spectral envelope information input from the multi-pulse encoding system communication network and the sound source power calculated based on the amplitude information of the encoded input voice. voiced/unvoiced discrimination means for determining the presence or absence of input voice or whether it is unvoiced, and directly converts the code structure of the sound source information in the multi-pulse coding type communication network into the code structure of the sound source information in the LPC vocoder type communication network. a first code converter for converting the signal, and a pitch period for converting the pitch period information extracted and encoded by the LPC vocoder after receiving the signal from the LPC vocoder communication network into the code format of the multi-pulse train and outputting the signal. a second code conversion section that includes code conversion means and directly converts the code structure of the sound source information in the LPC vocoder type communication network to the code structure of the sound source information in the multipulse coding type communication network. configured.

〔Example〕

Next, the present invention will be explained in detail with reference to the drawings.

FIG. 1 is a block diagram showing the basic configuration of a codec conversion apparatus according to the present invention.

The codec converting apparatus 1 shown in FIG. 1 includes a code converting section (1) 11 as a first code converting section and a code converting section (2) 12 as a second code converting section.

FIG. 1 further shows two communication networks having dissimilar codecs communicating through code conversion performed by a codec conversion device, namely a multipulse coding system network 2.
and an LPC vocoder type communication network 3 are also shown.

Multi-pulse coding system communication network 2
Terminal 21-1 is a voice communication terminal such as a telephone.
~21~N, and a multipulse encoding method as N codecs that encode audio signals input through these terminals and decode the encoded audio signals to reproduce the original audio signals. A multi-pulse coding system exchange 23 connects the codecs 22-1 to 22-N so that these N codecs and the LPC vocoder group of the LPC vocoder communication network can communicate in any combination.
It is composed of:

Multi-pulse encoding codec 22-1~
22-N uses LPC coefficients with the same contents as the LPC vocoder communication network 3 as the spectrum envelope information when encoding the characteristic parameters of the audio signals input via the terminals 21-1 to 22-N. However, while the LPC vocoder communication network 3 models and expresses sound source information using the pitch period, voiced/unvoiced, and sound source strength, the sound source information is expressed using a multi-pulse train. Expressing sound source information including waveform information,
A codec encodes these spectral envelope information and sound source information and sends them instead of an audio signal, and receives and decodes such encoded information, and then drives a speech synthesis filter to reproduce the input audio signal. It is.

On the other hand, LPC vocoder communication network 3 is
The input audio signals input through the terminals 31-1 to 31-N are supplied to the LPC vocoders 32-1 to 32-N, and LPC coefficients such as α parameters and K parameters are calculated using known LPC analysis techniques. The spectral envelope information is extracted at a predetermined order and encoded. In addition, the sound source information includes the pitch period, whether the sound is voiced or unvoiced, and the strength of the sound source, which are extracted using known techniques and then encoded and transmitted.
When this encoded information is received, it is decoded and the pitch period data is modeled using pulses with a repetition rate corresponding to the pitch period in the case of voiced sounds, and white noise in the case of unvoiced sounds to provide information on the strength of the sound source. At the same time, these are used as driving sound sources for a voice synthesis filter, and the aforementioned LPC coefficients are used as coefficients of the voice synthesis filter to operate the voice synthesis filter and reproduce the input voice signal.

The LPC vocoder exchange 33 supplies the encoded data of the input audio signal encoded in this way to any codec in the multipulse encoding communication network 2 via the codec conversion device 1, and It receives encoded data sent from the network via the codec conversion device 1, and performs a conversion operation to supply it to one of the vocoders.

FIG. 2 is an LPC vocoder code configuration diagram showing an example of a code configuration by an LPC vocoder used in the present invention, and FIG. 3 is a multipulse code configuration diagram showing an example of a code configuration by a multipulse encoding codec.

The numbers shown in Figures 2 and 3 indicate the serial number of bits allocated to each analysis frame of input audio.
Also, S is the synchronization bit, K1 to K12 are the 12th-order K parameters as spectrum envelope information, and the second
The amplitude shown in the figure shows the amplitude information of the sound source, and the pitch shows the pitch period information including voiced and unvoiced. Furthermore, the maximum amplitude shown in FIG. 3 indicates the maximum value of the multi-pulse train for each analysis frame, and amplitude 1 to amplitude 22 are normalized using the maximum amplitude as a reference.
The amplitude values of each multi-pulse made up of 22 impulses, pulse position 1 to pulse position 22 are the multi-pulse positions that indicate the distance between adjacent multi-pulses made up of 22 impulses as a parameter, and only pulse position 1 represents the distance from the analysis frame start time position of the leading pulse as a parameter. In addition, Figure 2 shows that the data rate is
4800bps (bit per second), analysis frame length 20m
SEC, and Figure 3 shows a data rate of 14.4K (Kilometers)
bps, and the analysis frame length is 20 mSEC.

As is clear from Figures 2 and 3, in these two code configurations, the K parameter as spectral envelope information is encoded with the same content using the same allocated bits, but other sound source information is encoded with the same contents. Different contents are encoded. The codec conversion apparatus shown in FIG. 1 converts the code structure shown in FIG. 2 into the code structure shown in FIG. 3, and converts the code structure shown in FIG. 3 into the code structure shown in FIG.

FIG. 4 is a block diagram showing the configuration of a first embodiment of the first code converter of the codec converter according to the present invention, and FIG. 5 is an implementation of the second code converter of the codec converter according to the present invention. FIG. 2 is a block diagram showing an example configuration.

The code converter (1) 11-1 as the first code converter shown in FIG. Correlation coefficient extraction circuit 115, sound source power calculation circuit 11
6, a multiplexer 117, etc.

In FIG. 4, the encoded signal supplied from the multi-pulse encoding type communication network 2 to the demultiplexer 111 has the code structure shown in FIG. This is a multiplexed signal regarding K parameter data, 22 multipulse information indicating the amplitude and position information of each, and maximum amplitude information as the maximum value of the multipulse train, and is demultiplexed by the demultiplexer 111. is performed and output.

The pitch period determination circuit 112 receives the encoded multipulse information and the encoded maximum amplitude information, and determines the pitch period of the input audio signal as follows using the amplitude and position data of 22 multipulses per analysis frame. do.

A multipulse train is a driving sound source sequence that can reproduce the sound closest to the input sound, expressed as a plurality of impulse sequences whose amplitude and temporal position are free.
In the case of pitch period extraction as well, if this multi-pulse train has been determined as a pre-processing, the pitch period can be easily determined using this multi-pulse train.

That is, a method of obtaining pitch period data as a multi-pulse period from the time position of the maximum value obtained by searching the multi-pulse autocorrelation coefficient sequence,
Alternatively, the pitch period can be easily extracted by calculating the sum of absolute values of the differences between multi-pulses, searching for the one that minimizes this value, and finding the pitch period as the time position information of the maximum value. can.

The pitch period data for each analysis frame determined in this way also includes the voiced/unvoiced discrimination circuit 1.
Voiced/unvoiced discrimination data for each analysis frame output from 14 is added to each analysis frame, and these data are supplied to the multiplexer 117 as pitch information of the code structure shown in FIG.

The excitation source amplitude determination circuit 113 also receives encoded multipulse information and encoded maximum amplitude information, decodes them, and reproduces 22 multipulses for each analysis frame. This multi-pulse is a pulse train that has a good approximation to the sound source information obtained by removing the spectral envelope information from the spectral distribution information of the input audio signal, and is used as the excitation sound source pulse of the speech synthesis filter in the case of a multi-pulse vocoder. . The excitation sound source amplitude determining circuit 113 calculates the root mean square value of the amplitude information of the 22 multi-pulse trains, uses this as a representative amplitude value representing the analysis frame, and sends this excitation pulse amplitude/noise amplitude information to the sound source power calculation circuit 116 and the multiplexer. 117.

The voiced/unvoiced discrimination data described above is generated as follows.

In this embodiment, the encoded spectrum envelope information of the common code structure shown in FIGS. 2 and 3 uses the 12th order K parameter.

These encoded spectral envelope information are supplied as they are to the multiplexer 117 as well as to the primary autocorrelation coefficient determining circuit 115.

The first-order autocorrelation coefficient extraction circuit 115 decodes this input and extracts the first-order autocorrelation coefficient ρ ₁ , that is, K ₁ using an arithmetic operation circuit. It is clear from many materials that this ρ ₁ has a fairly good correspondence with the first formant of the input speech, and that the value of ρ ₁ shows clear changes depending on voiced and unvoiced sounds. many.

In the case of this embodiment, since the 12th order K parameter is used as encoded spectrum envelope information, direct decoding may be performed to obtain _ρ1 , but
If the spectral envelope information has been converted into an expression format other than LPC depending on the operational purpose, it is first converted into LPC expression and then decoded to obtain the first-order autocorrelation coefficient ρ ₁ . ρ ₁ thus obtained is supplied to the voiced/unvoiced discrimination circuit 114.

The sound source power calculation circuit 116 receives excitation pulse amplitude or noise amplitude information from the excitation sound source amplitude determination circuit 113, and determines that each analysis frame is a voiced or unvoiced sound section whose effective value is the amplitude value based on the amplitude information. Based on this amplitude information, the power of each analysis frame is calculated, and this is taken as the short-time average sound source power. When comparing the sound source power of voiced and unvoiced sounds on an analysis frame basis, voiced sounds generally have higher energy and higher sound source power, and therefore the magnitude of the short-term average sound source power also differs between voiced and unvoiced sounds. It can be used as a judgment condition.

The voiced/unvoiced discrimination circuit 114 compares the input primary autocorrelation coefficient and short-term average sound source power with respective predetermined decision threshold values, and determines whether each analysis frame is voiced or unvoiced.

The multiplexer 117 multiplexes each encoded information input in a predetermined format.
Send to LPC vocoder communication network 3,
In this way, it becomes possible to transmit data by direct code conversion from the codec having the code structure shown in FIG. 3 to the codec having the code structure shown in FIG.

A code converter (2) 12-1 as an embodiment of the second code converter shown in FIG. 5 is a demultiplexer 1.
211, pitch period decoder 1212, pitch pulse train generator 1213, switch 1214, noise generator 1215, clipper 1216, pitch period code conversion circuit 1217, and multiplexer 121
It is composed of 8 mag.

The code conversion unit (2) 12-1 receives the encoded spectrum envelope information, encoded amplitude information, and encoded pitch period information multiplexed from the LPC vocoder communication network 3 to the demultiplexer 1211, and
These are multiplexed and separated. This information is encoded based on the code structure shown in Figure 2,
The encoded spectrum envelope information and sound source amplitude information for each analysis frame using the 12th-order K parameter are supplied as they are to the multiplexer 1218.

Now, the encoded pitch period information is supplied to the pitch period decoder 1212 based on the code structure shown in FIG. 2, and as content including voiced/unvoiced data, and decoded. The voiced/unvoiced data is separated and supplied to a switch 1214.

The switch 1214 switches the pitch pulse train generator 1 when the input voiced/unvoiced data specifies voiced data.
213 to the pitch periodic code conversion circuit 1217.
Also, when the voiced/unvoiced data specifies unvoiced, the output of the clipper 1216 is switched to be supplied to the pitch cycle code conversion circuit 1217.

The pitch pulse train generator 1213 generates a pitch pulse train of pulses having the repetition number of the pulse interval corresponding to the pitch period data decoded by the pitch period decoder 1212. is supplied to the pitch cycle code conversion circuit 1217 via the switch 124.

The pitch period code conversion circuit 1217 converts the input pitch pulse train into the code configuration shown in FIG. In the case of 22 pulses, the equal pulse intervals are converted to the contents of pulse positions 1 to 22 in FIG. 3.

The pitch cycle code conversion circuit 1217 also receives encoded amplitude information for each analysis frame, and uses this as the amplitude information of the multi-pulse train for each analysis frame. That is, the encoded amplitude information is decoded and expressed in normalized units, and these are the contents corresponding to amplitudes 1 to 22 in FIG. 3.

The encoded amplitude information described above is used as the maximum amplitude in FIG. 3. Therefore, when converting the code structure from FIG. 2 to the format shown in FIG. The extracted amplitude information for each analysis frame will be used as the maximum amplitude in FIG. 3 and as the multi-pulse amplitude information.

Now, when the voiced/unvoiced data specifies unvoiced, the switch 1214 is connected to supply the output of the clipper 1216 to the pitch cycle code conversion circuit 1217.

The white noise output from the noise generator 1215 is subjected to so-called center clipping in which the center portion of its amplitude is clipped by a clipper 1216 and only the portions remaining above and below the clipped portion are used. The range of this center clipping is set for each analysis frame so that the total number of noise pulses existing on the upper and lower sides left through this clipping is equal to the number of multipulses. This is carried out in the form of counting the number of pulses remaining after clipping, and the pulse positions of the 22 pulses with the set random occurrence probability are determined by the distance between the pulses, and the first pulse is determined by the distance from the starting point of the analysis frame. It is expressed and converted into the code structure of pulse position 1 to pulse position 22 shown in FIG.

Further, the amplitude information in this case is converted based on the code structure of amplitude 1 to amplitude 22 shown in FIG. 3 using encoded amplitude information for each analysis frame, as in the animate case.

In this way, the transmission signal of the LPC vocoder communication network 3 formed based on the code structure shown in FIG. 2 is code-converted in accordance with the code structure shown in FIG. Supplied to network 2.

Next, a second embodiment of the first code conversion section will be described.

The second embodiment of the first code conversion section is such that the pitch period determination circuit shown in FIG. 4 determines the pitch period through the maximum value of the pseudo-autocorrelation coefficient sequence of the multi-pulse train. .

FIG. 6 is a block diagram showing a second embodiment of the first code conversion section.

The pitch period determining circuit 112 shown in FIG. 4 decodes the input encoded multi-pulse information to reproduce multi-pulse trains, and then searches the autocorrelation coefficient sequences of these multi-pulse trains to determine the pitch period from the time position of the maximum value obtained. One of the basic methods is to find the period, but the code conversion section shown in Figure 6
(1) The pitch period determination circuit 118 of 11-2 determines the pitch period using the following pseudo-autocorrelation coefficient.

The pseudo-autocorrelation coefficient is an autocorrelation coefficient obtained over a finite interval, and is intended to reduce the amount of calculation required.

The pseudo-autocorrelation coefficient is expressed by the following equation (1).

In equation (1), j is the corresponding multipulse number of the multipulse extraction window position, n is the number of multipulses in a finite interval from j, i is the position shift amount W _j from j,
W _j+i are the multipulses at pulse positions j and j+i, respectively.

Equation (1) means that the pseudo-autocorrelation coefficient _i is calculated for a finite interval of multipulses by taking the cross-correlation of n multipulse trains whose pulse positions are separated by i. The pitch period determination circuit 118 calculates this pseudo-autocorrelation coefficient _i , searches for its maximum value, and attempts to determine the pitch period from that time position.

FIG. 7 is a block diagram showing the structure of a third embodiment of the first code conversion section of the codec conversion apparatus according to the present invention.

The configuration of the code conversion unit (1) 11-3 shown in FIG.
Pseudo autocorrelation coefficient calculator 119, maximum value searcher (1)
120, Maximum value searcher (2) 121, Maximum value searcher (3)
122 and the pitch period determination circuit 123 are all the same as the configuration of the first embodiment shown in FIG. 4, so a detailed explanation thereof will be omitted.

The pseudo autocorrelation coefficient calculator 119 shown in FIG. 7 calculates the pseudo autocorrelation coefficient in the second embodiment described above, and decodes the encoded multipulse information and then calculates the pseudo autocorrelation function. and convert this into maximum value searchers (1) 120, (2) 121 and (3) 1
22 respectively. These three maximum value searchers each have a search interval that divides the pitch cycle search range to be searched into three through the maximum value of the pseudo-autocorrelation coefficient sequence, and The maximum values of the pseudo-autocorrelation coefficient sequences are searched for, and the time delays up to these maximum values are supplied to the pitch cycle determining circuit 123.

The pitch period determination circuit 123 further determines the maximum value of the pitch periods found from the time-delayed data up to the maximum value of the three pseudo-autocorrelation coefficients in each input search interval, and then once determines the maximum value of the pitch period. Temporarily determined as a temporary pitch period. Then, if the maximum pitch period of another search interval exists at a time position within one integer of this provisional pitch period, the pitch period is recalculated by taking into account the weighting coefficient set in advance based on audio materials, etc. Determine. The pitch period determination circuit 123 has a determination logic circuit that reliably determines the pitch period in this way, and determines the hit period and encodes it while being supplied with voiced/unvoiced discrimination data for each analysis frame from the voiced/unvoiced discrimination circuit 114. and supplies it to multiplexer 117.

The above-mentioned weighting coefficients are set in advance for the purpose of distinguishing double pitch, half pitch, etc. appearing in the pitch period and extracting the pitch period with high accuracy.

In this embodiment, the search range of the pitch period is divided into three parts, but this can be arbitrarily set in consideration of the operational purpose and the like.

FIG. 8 is a block diagram showing the configuration of a fourth embodiment of the first code conversion section of the codec conversion apparatus of the present invention.

Code converter (1) 11 of the fourth embodiment shown in FIG.
4 differs from the third embodiment shown in FIG. 7 only in that a pitch memory 124 is added.

The pitch memory 124 includes the pitch cycle determination circuit 1
Approximately several frames of the pitch period outputted from 22 are stored and supplied to the pitch period determination circuit 123 to facilitate the pitch period determination process, and furthermore, the pitch search range is expanded/contracted based on these past pitch periods. The maximum value searchers (1) 120, (2) 12
1 and (3) 122 to improve search efficiency.

Next, a fifth embodiment of the first code converter of the codec converter according to the present invention will be described.

The code conversion unit (1) 11-5 of this fifth embodiment is based on the first self-correlation obtained from the encoded spectrum envelope information by the voiced/unvoiced discriminator in the first embodiment explained with reference to FIG. The determination factors are the sound source power obtained by directly using the amplitude information of the sound source encoded by the multipulse encoding method, After decoding and reproducing, the short-term average sound source power for each analysis frame is determined through the effective value of the sound source waveform determined from this multi-pulse train.

FIG. 9 is a block diagram showing the configuration of a fifth embodiment of the first code conversion section.

The sound source power calculation circuit 125 shown in FIG. 9 decodes the encoded multi-pulse to reproduce a multi-pulse train, reproduces a sound source waveform based on this multi-pulse train, and calculates the short-term average sound source for each analysis frame from this effective value level. The power is calculated and supplied to the voiced/unvoiced discrimination circuit 114.

Next, a sixth embodiment of the first code conversion section will be described.

FIG. 10 shows the code conversion unit (1) 11- of the sixth embodiment.
FIG. 6 is a block diagram showing the configuration of FIG.

This sixth embodiment differs from the first embodiment shown in FIG. The sound source power calculated based on the sum of squares is used as the voiced/unvoiced discrimination condition, and it reflects a more accurate sound source power than the first embodiment, and therefore the discrimination accuracy is also improved. It can be improved.

In FIG. 10, after the encoded maximum amplitude information is decoded by the maximum amplitude decoding circuit 126,
The square value is calculated by the square calculation circuit 127 and used as the excitation sound source by the LPC synthesis filter 128.
supply to. The excitation sound source supplied in this way is the square value of the maximum multipulse value for each analysis frame by the multipulse encoding method, and the maximum amplitude multipulse is converted into a positive pulse by this squaring process, and then each It is supplied to the LPC synthesis filter 128 as corresponding to the sound source power of the analysis frame. Use all-pole digital filter
The LPC synthesis filter 128 is also supplied with the 12th-order K parameter from the first-order autocorrelation coefficient extraction circuit 115 and is driven by the excitation sound source using this as a filter coefficient. An impulse response waveform driven by is output from the LPC synthesis filter 128.

The impulse response waveform output in this way is then supplied to an impulse response power calculation circuit 129, and its power is calculated. This power is supplied to the voiced/unvoiced discrimination circuit 114 as being close to the power of the composite waveform.

FIG. 11 is a block diagram showing the configuration of a seventh embodiment of the first code conversion section. Next, a seventh embodiment will be described with reference to FIG.

The seventh embodiment is the same as the sixth embodiment shown in FIG. 10, in which weighting corresponding to the average value of the sum of squares of the normalized multi-pulse amplitude is applied to the sound source power.

The code conversion unit (1) 11-7 shown in FIG. 11 supplies encoded multi-pulse information and encoded maximum amplitude information to a multi-pulse decoding circuit 130 to reproduce a multi-pulse train. When reproducing this multipulse, the amplitude of the multipulse is normalized, and all multipulses including the maximum value for each analysis are output as century values, and then the sum of squares average calculation circuit 1
31, the average value of the sum of squares is calculated. The sum-of-squares average value for each analysis frame calculated in this way has a value corresponding to the sound source power level for each analysis frame, and by weighting the sound source power, the sound source power can be calculated with even higher accuracy. This can be used for voiced/unvoiced discrimination.

After performing such weighted multiplication, the multiplication circuit 132 supplies the weighted sound source power to the voiced/unvoiced discrimination circuit 114 to perform highly accurate voiced/unvoiced discrimination.

Next, an eighth embodiment of the first code conversion unit according to the present invention will be described.

FIG. 12 is a block diagram showing the configuration of an eighth embodiment of the first code conversion section.

The code conversion unit (1) 11-8 shown in FIG.
In the first embodiment shown in the figure, the sound source power calculated as the sum of squares of the impulse response waveform whose impulse amplitude is the maximum multipulse value included in the code structure of the multipulse encoding method is expressed as voiced/
This is used for voiceless determination, specifically the first
In the sixth embodiment shown in Figure 0, the square value of the maximum multipulse value included in the code structure of the multipulse encoding method is used as the excitation impulse of the LPC synthesis filter, whereas the maximum multipulse value is used as the excitation impulse. As shown in FIG. 12, the impulse generator 133 generates an impulse with an amplitude corresponding to the maximum value of the multi-pulse, and supplies this as an excitation sound source to the LPC synthesis filter 128.

Finally, a ninth embodiment of the first code conversion section of the codec conversion apparatus of the present invention will be described.

FIG. 13 is a block diagram showing the structure of a ninth embodiment of the present invention.

The ninth embodiment shown in FIG. 13 weights the sound source power in the eighth embodiment shown in FIG. 12 by the average value of the sum of squares of the normalized multipulse amplitude. A multiplication circuit 13 uses the output of the sum-of-squares average calculation circuit 131 as a weighting factor for the output of the impulse response power calculation circuit 129.
2, which enables highly accurate voiced/unvoiced discrimination.

The object of the present invention is that it has a basic feature in that different types of codecs can be connected through direct conversion of encoding parameters, and various modifications of the first to ninth embodiments described above are possible. .

For example, the pitch period determination circuit 112 in the ninth embodiment shown in FIG. It may be replaced with the pitch period determining means shown in the fourth embodiment, and in this case, the characteristics of each pitch period determining means are taken into consideration. Also, in this way, the first
Various other modifications by combinations of the configurations from to to ninth can be easily considered, and any of these can be easily implemented without impairing the gist of the present invention.

〔Effect of the invention〕

As explained above, according to the present invention, in a codec conversion device that exchanges the code structure of each communication network in order to interface an LPC vocoder type communication network and a multipulse coding type communication network, By connecting different codecs with a code conversion means that directly converts the coding feature parameters of
This has the advantage that it is possible to realize a codec conversion device that fundamentally eliminates the increase in deterioration in communication quality, significantly shortens processing time, and significantly simplifies the hardware configuration.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the basic configuration of the codec conversion apparatus of the present invention, FIG. 2 is a block diagram of the LPC vocoder code used in the present invention, and FIG. 3 is a block diagram of the multipulse code used in the present invention.
FIG. 4 is a block diagram showing the configuration of a first embodiment of the first code converter of the codec converter according to the present invention, and FIG. 5 is an implementation of the second code converter of the codec converter according to the present invention. Blog diagrams showing examples, Figures 6, 7, 8, 9, and 10
11, 12 and 13 show the second, third, fourth, fifth, sixth, seventh and eighth code conversions, respectively, of the first code conversion of the codec conversion apparatus according to the present invention.
and FIG. 9 is a block diagram showing the configuration of a ninth embodiment. DESCRIPTION OF SYMBOLS 1... Codec converter, 2... Multi-pulse code system exchanger, 3... LPC vocoder system communication network, 11, 11-1, 11-2, 11-
3, 11-4, 11-5, 11-6, 11-7,
11-8, 11-9... code conversion unit (1), 12, 1
2-1... Code converter (2), 21-1 to 21-N...
...Terminal device, 22-1 to 22-N...Multi-pulse coding system codec, 23...Multi-pulse coding system transmitting/receiving station, 31-1 to 31-N...Terminal device, 32-1 to 32- N...LPC vocoder,
33... LPC vocoder system exchange, 111... Demultiplexer, 112... Pitch cycle determining circuit, 113... Excitation sound source amplitude determining circuit, 114...
...voiced/unvoiced discrimination circuit, 115...primary autocorrelation coefficient extraction circuit, 116...sound source power calculation circuit, 1
17... Demultiplexer, 118... Pitch period determination circuit, 119... Pseudo autocorrelation coefficient calculator, 120... Maximum value searcher (1), 121... Maximum value searcher (2), 122... Maximum value searcher (3), 123
... Pitch period determination circuit, 124 ... Pitch memory, 125 ... Sound source power calculation circuit, 126 ... Maximum amplitude decoding circuit, 127 ... Square calculation circuit, 1
28...LPC synthesis filter, 129...Impulse response power calculation circuit, 130...Multipulse decoding circuit, 131...Sum of squares average calculation circuit, 1
32...Multiplication circuit, 133...Impulse generator, 1211...Demultiplexer, 1212...
...Pitch period decoder, 1213...Pitch pulse train generator, 1214...Switcher, 1215...
Noise generator, 1216...Clippa, 1217...
...Pitch periodic code conversion circuit, 1218...Multiplexer.

Claims (1)

[Scope of Claims] 1. A codec conversion device for interfacing an LPC vocoder-based communication network and a multi-pulse coding system communication network by converting the code structures of the communication networks, comprising:
pitch period determining means for determining the pitch period of a sound source from a multi-pulse train received and encoded by a signal from the multi-pulse coding system communication network; excitation source amplitude determining means for determining the excitation pulse amplitude or excitation noise amplitude to be excited; and a first-order autocorrelation relationship obtained by decoding encoded spectral envelope information input from the multi-pulse encoding system communication network; a voiced/unvoiced discriminator for determining the presence or absence of input speech or whether it is unvoiced based on the sound source power calculated based on the amplitude information of the encoded input speech; a first code conversion section that directly converts the code structure of the information into a sound source information code structure in the LPC vocoder communication network; a pitch period code conversion means for converting the pitch period information into the code format of the multi-pulse sequence and outputting the converted pitch period information; 1. A codec conversion device comprising: a second code conversion section that converts the code configuration of sound source information into the code structure of the sound source information. 2. Claim 1, characterized in that the pitch period determining means determines the pitch period via the maximum value of pseudo-autocorrelation coefficients calculated for a preset finite section of the multi-pulse train. The codec conversion device described. 3 When determining the pitch period using the maximum value of the pseudo-autocorrelation coefficient, search for the maximum value of the pitch period in each search range obtained by dividing the pitch period search range into at least three parts, and then search for the maximum value of the pitch period in each search range, and then calculate the maximum value of each maximum value The maximum pitch period is tentatively determined by searching for the maximum value, and if the maximum value of the search range exists near an integer fraction of the tentatively determined pitch pitch period, a preset pitch pitch determination weight is assigned. 3. The codec device according to claim 2, further comprising means for determining the pitch period again. 4. The codec conversion apparatus according to claim 3, further comprising means for storing pitch cycles calculated over several past frames of analysis frames and controlling a pitch search range based on the pitch cycles. 5. Calculating the sound source power in the voiced/unvoiced discriminating means by providing multipulse decoding means for calculating the sound source power based on the effective value of the sound source waveform obtained from the multipulse train that is reproduced by decoding the encoded multipulse train. A codec conversion device according to claim 1, characterized in that: 6. The sound source power in the voiced/unvoiced discrimination means is calculated based on the square value of the maximum multipulse value included in the code configuration of the multipulse encoding method and the sum of squares of the impulse response waveform of the speech synthesis filter. A codec conversion device according to claim 1, characterized in that: 7. The codec conversion device according to claim 6, further comprising means for applying weighting to the sound source power corresponding to the average value of the sum of squares of the normalized multi-pulse amplitudes. 8. The voice power calculated as the sum of squares of impulse response waveforms with the maximum multipulse value included in the code configuration of the multipulse encoding method as the impulse amplitude is used as the sound source power in the voiced/unvoiced discrimination means. A codec conversion device according to claim 1. 9. The codec conversion apparatus according to claim 8, further comprising means for imparting weighting to the sound source power corresponding to the average value of the sum of squares of the normalized multi-pulse amplitudes.