JPH05102942A - Voice coding device - Google Patents

Abstract

PURPOSE:To effectively change the voice coding speed based on the using state of a circuit by selectively switching plural thinning processing parts corresponding to the different coding speeds based on the speed information given from a circuit controller. CONSTITUTION:A thinning processor 4 is provided with the thinning processing parts 41-43 which code the voices at the different coding speeds. When the sound signals are inputted to the processor 4 and a switch 7, the switches S1 and S2 are switched to each other by the speed information. Then a coding operation is carried out based on a thinning pattern of a discrete cosine transformation(DCT) coefficient accordant with the coding speed. Thus it is possible to carry on the communication with no interruption caused by the circuit state by reducing the coding speed with reduction of the assigned capacity of the circuit and then increasing the coding speed with recovery of the assigned capacity of the circuit respectively.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a speech coding apparatus using a waveform coding method which is one of speech signal coding methods, and more particularly to adaptive transform coding with pitch prediction (ATC). ) System speech encoding apparatus.

[0002]

2. Description of the Related Art Conventionally, a speech coding apparatus has been developed which is designed so that the transmission rate of a line is constant and the coding rate is also constant. For example, one of the present inventors applied for a speech coder (Japanese Patent Application No. 1-3335281).
No.) proposed a reduction of distortion by adopting 4.8 kbit / s as a speech coding rate. On the other hand, one of the merits of general digital signal multiplex transmission is that it is possible to effectively use the line by flexibly combining data signals of various speeds from a plurality of transmitters according to the capacity of the system line. Can be mentioned. A speech coder is a speech coder that converts a speech signal into a digital signal and sends it to a line,
It is composed of a voice decoder for converting a digital signal received from a line into an original voice signal. The speech encoder encodes a speech signal into pitch information, DCT decimation information, DCT coefficient, DC
The T position information is analyzed and encoded.

FIG. 6A is a diagram showing a configuration example of a conventional speech coder. In the figure, the input voice signal is divided into frames every 30 ms, and pitch information is detected by the pitch analyzer 1. The pitch predictor 2 predicts the pitch of the input speech signal for each subframe of 15 ms according to the extracted pitch information, performs DCT conversion on the obtained prediction residual signal by the discrete cosine converter (DCT) 3, and then performs a decimation processor. The coefficient is thinned and encoded by 40 to obtain an encoded output (see Japanese Patent Application No. 2-175642). further,
The coefficient obtained by decoding the coded output by the decoder is inverse DCT5.
After inverse DCT conversion by, the pitch synthesizer 6 synthesizes the pitch and supplies it to the pitch predictor 2 for input to the pitch prediction of the next frame (see Japanese Patent Application No. 1-3335281). FIG. 6B shows a configuration example of one frame of coded output having a coding rate of 4.8 kbps. In the figure, A
Is information for frame synchronization and coding rate display, and is for notifying the receiving side. Normally, the reserved 5 bits are used. The decoder switches the coding rate based on this information. B and C are pitch information (9 bits) and decimation information (4 bits), respectively, and D is DCT coefficient related information in which DCT coefficients (92 bits) and DCT position information (34 bits) are combined. That is, the coding rate is 4.8 kbp
The bit allocation per frame when s is as shown in Table 1 below.

[0004]

[Table 1]

[0005]

When attempting to apply a plurality of such speech coding apparatuses to a multi-system line, the speech coding apparatus having a constant coding rate as described above uses a flexible line. I can't. That is, in the case of a fixed-rate voice encoding device, the data rate of another device used at the same time temporarily rises, the occupied capacity of the line increases, and the encoding rate assigned to the voice encoding device is the voice encoding. When the value falls below the specified value of the device, the voice coding device is forced to interrupt the communication. Furthermore, if the data rate of another device decreases and the line has a margin, the voice quality can be expected to be improved because a large amount of voice data can be transmitted if the voice coding rate of the device can be increased. However, this is not possible with the fixed speeds mentioned above. An object of the present invention is to provide a voice code capable of effectively changing the voice coding rate depending on the usage status of the line when a plurality of voice coding apparatuses of the adaptive transform coding system with pitch prediction are used as the voice coding system. To provide a rectification device.

[0006]

A speech coding apparatus according to the present invention comprises pitch information obtained by pitch-analyzing a speech signal divided into predetermined frames, and a prediction residue obtained by performing pitch prediction of the speech signal according to the pitch information. An adaptive transform coding system with pitch prediction, which is configured to transform the difference signal by discrete cosine transform, then perform thinning coding of the discrete cosine transform coefficient by the thinning processor, and send it to the line controller as a coded output together with the pitch information. In the voice encoding device, the decimation processor includes a plurality of decimation processing units that perform encoding with a decimation pattern of discrete cosine transform coefficients corresponding to a plurality of different encoding speeds, and speed information from the line control device. A switch for selectively switching the thinning-out processing unit corresponding to the corresponding encoding speed, and using the line detected by the speed information. State by and is characterized by being configured to deliver an encoded output by changing the encoding rate.

The speech coder according to the present invention comprises a speech coder for converting a speech signal into a digital signal and transmitting the speech signal, and a speech decoder for converting a received digital signal into an original speech signal as in the conventional case. It The speech coder converts the speech signal into pitch information, DCT decimation information, DCT coefficient, DCT
The position information is analyzed and encoded. The characteristic of the adaptive transform coding system with pitch prediction is that the bit rate of the basic coding rate (here, 4.8 kbit / s) causes the DC
By increasing (decreasing) the amount of information regarding the T coefficient, it is possible to improve (decrease) the voice quality. Therefore, by decreasing the coding rate when the allocated capacity of the line decreases and increasing the coding rate when it recovers, the communication can be continued without interruption due to the status of the line. Next, the speech coding apparatus of the present invention capable of changing the coding speed will be specifically described.

[0008]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1A is a diagram showing an example of a circuit configuration of a speech coder of the present invention in which a coding rate is variable. The difference from the conventional configuration of FIG. 6 is that the method of thinning processing is changed depending on the speed information. That is, in the figure, the pitch analyzer 1, the pitch predictor 2, the DCT 3, the pitch synthesizer 6, and the inverse DCT 5 are the same as the conventional one shown in FIG. 6, and the thinning processor 4 and the switching device 7 are different from the conventional one. It is a thing. The thinning-out processing unit 4 is provided with a plurality of thinning-out processing units for coding at different coding rates, which will be described later, and the switching unit 7 is an on / off switching unit for not outputting pitch information when not transmitting. Is. FIG. 5 is a diagram showing a configuration example of a multiple system to which the present invention is applied. As shown in the figure, a plurality of speech coding devices 50 are connected to the line control device 53. Each speech encoding device 50 has an encoding unit 51 and a decoding unit 52 installed therein. Speed information is given from the line control device 53 to each speech encoding device 50, and
(A) is input to the decimation processor 4 and the switch 7. As a variable example of the coding rate, a standard coding rate of 4.8 kbp
2. When the line is busy depending on the usage status of the line.
A case will be described in which the speed is lowered to 6 kbps and the speed is raised to 6.0 kbps when it is idle. Table 2 shows an example of bit allocation per frame when the coding rates are 3.6 kbps and 6.0 kbps, corresponding to Table 1.

[0009]

[Table 2]

In this case, 3.
Although the case of changing to 6 kbps and 6.0 kbps is targeted, it is also possible to change to a speed other than this. FIG. 2 is a block diagram showing the details of the thinning-out processor 4 of FIG. 1A, in which 41, 42 and 43 are each 3.6 k.
It is a decimation processing unit for bps, 4.8 kbps, and 6.0 kbps, and switches S1 and S2 depending on speed information.
Are switched. Reference numeral 44 is an encoder, and 45 is a decoder. FIG. 3 is an example of a bit configuration of speed information given from the line control device 53. The rate information is assigned to 2 bits of the synchronization / encoding rate information A (5 bits) in FIG. 1B, and the type of encoding rate is represented by a combination of low level (L) and high level (H). .. FIG. 1B is a diagram showing a configuration example of one frame of encoded output according to the present invention. In the figure, A, B and C are the same information as in FIG. 6B, and the coding rates are 3.6 kbps, 4.8 kbps and 6.0 kb.
Each ps has the same format code. On the encoder side, the information of the coding rate is shown in A and 3.6 kbps.
Then, in D1, DCT coefficient and DCT position information (68 + 22
Bit) is set and D2 and D3 are deleted. 4.8 kb
ps uses D1 and D2 (92 + 34 bits) to remove D3. At 6.0 kbps, D1, D2, D3
By using (122 + 40 bits) to allocate information, it is possible to respond to changes in the amount of information. The decoder determines the coding speed from the information of A and decodes according to the format.

The following Table 3 is a table showing a coefficient thinning method. The maximum absolute value is encoded by 4 bits to normalize all the coefficients, and the coefficient having the larger absolute value is preferentially selected and encoded. To do. The number of selected coefficients can be changed according to the speed information.

[0012]

[Table 3]

FIG. 4 is a block diagram showing an example of the configuration of the decoder of the device of the present invention. Pitch information of the received input signal from the line is input to the pitch synthesizer 10 and other information is input to the decoder 8. It is input and decoded at the coding rate according to the rate information from the line controller, and the inverse DCT 9 performs the inverse DCT.
It is converted and input to the pitch synthesizer 10.

[0014]

As a result of implementing the present invention, when the capacity allocated to the voice coder of the transmission line using the voice coder increases, the voice coder transmits high quality voice to the receiving side and reproduces it. You can do it. In addition, when the capacity of the transmission line decreases, the line can be maintained by lowering the voice coding speed and slightly sacrificing the voice quality, and the call is not interrupted. Since the ATC speech coding system with pitch prediction, which is the object of the present invention, has an algorithm that can easily cope with a change in the coding rate, it is possible to realize a speech coding apparatus capable of changing the coding rate. There are no particular restrictions on production.

[Brief description of drawings]

FIG. 1 is a circuit configuration diagram and an encoded output frame configuration example diagram showing an embodiment of a speech encoder of the present invention.

FIG. 2 is a diagram showing a circuit configuration example of a thinning-out processor which is a main part of the present invention.

FIG. 3 is a diagram showing an example of a bit configuration of A information shown in FIG.

FIG. 4 is a block diagram showing an embodiment of a speech decoder of the present invention.

FIG. 5 is a diagram showing a system configuration example to which the present invention is applied.

FIG. 6 is a circuit configuration diagram of a conventional speech encoder and 4.8 kb.
It is a ps coded output frame structural example figure.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Pitch analyzer 2 Pitch predictor 3 DCT 4 Decimation processor 5 Inverse DCT 6 Pitch synthesizer 7 Switcher 8 Decoder 9 Inverse DCT 10 Pitch synthesizer 40 Decimation processor 41, 42, 43 Decimation processing unit 44 Encoder 45 Decoder 50 Speech Encoder 51 Encoding Unit 52 Decoding Unit 53 Line Control Device

Claims (1)

[Claims] 1. Pitch information obtained by pitch analysis of a voice signal divided into predetermined frames, and a prediction residual signal obtained by performing pitch prediction of the voice signal according to the pitch information are discrete cosine transformed and then discretized by a decimation processor. In a speech coding apparatus of an adaptive transform coding system with pitch prediction, which is configured to perform thinning-out coding of cosine transform coefficients and send out to the line controller as a coded output together with the pitch information, the thinning-out processor includes A plurality of decimation processing units that perform decimation with discrete cosine transform coefficient decimation patterns corresponding to a plurality of different coding rates; and a decimation processing unit that corresponds to the coding rates according to the rate information from the line control device. With a switch for selectively changing, depending on the usage state of the line detected by the speed information,
A speech coding apparatus configured to send a coded output with a different coding rate.