JPH06104851A - Digital voice transmitter - Google Patents

Abstract

PURPOSE:To provide the excellent digital voice transmitter in which sound noise is less generated at the time of background noise reproduction for a non-sound period. CONSTITUTION:A voice coder of a digital voice transmitter is provided with a code converter 4 and a multiplexer 5, and when data for background noise reproduction are sent, the code converted by the code converter 4 is given to a multiplexer 6. Thus, the code sent for background noise reproduction is converted into a code at which a noise sound not often heard is not caused and the noise sound is hardly reproduced by the voice decoder.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital voice transmission device used for digital voice communication, digital mobile communication and the like.

[0002]

2. Description of the Related Art Recently, a voice transmission device is used for mobile communication, and it is desired to realize a high-performance VOX function, which is one method for digitization and low power consumption. A conventional example of such a digital audio transmission device will be described below.

FIG. 3 shows a speech encoding apparatus of a conventional digital speech transmission apparatus. In this figure, reference numeral 1
0 is an encoder. Reference numeral 11 is a sound / non-sound determination device, and
2 is a VOX controller. 13 is a multiplexer. Figure 4
Shows a voice decoding device of a conventional digital voice transmission device. In this figure, reference numeral 14 is a separator, and 15
Is a VOX controller. Reference numeral 16 is a code controller, and 17
Is a multiplexer (MUX) and 18 is a decoder.

The operation of the digital audio transmission device configured as described above will be described below. First, the speech coding apparatus will be described. The audio signal is subjected to data compression and encoding by the encoder 10. Further, the sound / silence judging device 11 judges whether the input sound signal is sound or not. VOX
The controller 12 controls the state according to the state transition diagram of FIG. The multiplexer 13 creates transmission data as follows according to the state determined by the VOX controller 12. VCE: The result of the encoder is used as transmission data. PST: A special pattern (postamble) is used as transmission data. BGN: The result of the encoder is used as transmission data. (Data for background noise reproduction) Si: Nothing is transmitted. PRE: A special pattern (preamble) is used as transmission data.

The operation of the speech decoding apparatus shown in FIG. 4 will be described below. First, the separator 14 separates the voice code from the transmission data by detecting the preamble and postamble patterns. The VOX controller 15 determines a voiced section and a silent section based on the preamble and the postamble, and determines the state of the speech decoding apparatus. In the voiced section, the multiplexer 17 connects the voice code from the separator to the decoder 18, and the decoder 18 outputs the voice signal. In the silent section, the code controller 16 causes the decoder 1
8 and the decoder 18 reproduces the background noise. The code controller 16 controls the code using the data of the next frame of the post preamble to reproduce the background noise in the silent section.

[0006]

However, in such a conventional digital voice transmission apparatus, the performance of the voice / sound deciding unit of the voice coder is not sufficient at the current technical level, so that a decision error is generated. May occur. When the voiced section is erroneously determined to be silent, the voice encoding apparatus transmits the voice in the voiced section as data for background noise reproduction after transmitting the post-preamble data. Therefore, since the voice decoding device adopts a method of repeatedly performing background noise reproduction in the silent section based on the data in the voiced section, an unfamiliar noise is repeated in the silent section. Had a problem.

The present invention has been made in view of the above problems, and an object of the present invention is to provide an excellent digital audio transmission device in which abnormal noise is less likely to occur during background noise reproduction in a silent section.

[0008]

In order to achieve the above-mentioned object, the present invention provides a speech coding apparatus with a code converter and a multiplexer, and when transmitting background noise reproduction data, the code converter converts the data. The gist is that the generated code is connected to the multiplexer.

[0009]

With the above structure, the code transmitted for the background noise reproduction is code-converted into a code in which an unfamiliar noise is hardly generated. Therefore, it is possible to make it difficult for the voice decoding device to reproduce the abnormal sound.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. Here, as the encoding / decoding method, "Digital method car telephone system standard RC
The VSELP system defined by RSTD-27 ”will be described as an example.

In FIG. 1, reference numeral 1 is a VSELP for data-compressing and encoding an audio signal by the VSELP method.
A voice encoder, 2 is a voice / sound determiner for determining whether the input voice signal is a voice signal or a silence signal, and 3 is a state control based on the determination result of the voice / sound determiner 2. A VOX controller, 4 is a code converter for converting a coded voice signal into a code in which an unfamiliar noise is hardly generated,
Reference numeral 5 is a multiplexer (MUX) for selecting and outputting a code-converted signal and a code-unconverted signal of the encoded voice signal, and 6 is a multiplexing for creating transmission data according to a state determined by the VOX controller 3. It is a vessel.

The operation of the digital voice transmission apparatus configured as described above will be described with reference to FIG. First, the voice input is encoded by the VSELP voice encoder 1. At the same time, the sound / sound determination unit 2 determines whether the sound input is sound or not, and the VOX controller 3 determines the state according to the state transition diagram of FIG. The state of the multiplexer 5 is "
Only in the case of BGN ″, the output of the code converter 4 is connected to the multiplexer 6. The multiplexer 6 creates transmission data by the conventional method. The code converter 4 sets the parameters of the VSELP encoder 1. Of these, only the four parameters of GSP0_1 to GSP0_4 are converted.
It takes an integer value of 127 and is a common index for the GS value table and the PO value table, which are parameters in VSELP. The code converter 4 converts each of the four values of GSP0 into a value of GSPO that indicates a value with a small PO value (0.1 or less) and a GO value that does not change much. This can be easily realized by preparing a conversion data table.

The structure of the speech decoding apparatus is the same as that shown in FIG. However, in this embodiment, the decoder 1
A VSELP decoder is used for 8. The code controller 16 holds the coded data next to the postamble,
Only four values of CODE_1 to CODE_4 are randomly changed.

FIG. 2 is a block diagram showing a VSELP decoder. Long-term filter state 7 and probability codebook 8
There are two codebooks, the data is selected from the respective codebooks by the codes of LAG and CODE and added, and the speech is decoded by the synthesis filter 9. LAG_1 to LAG_4, CODE_1 to CO
Each of the four codes of DE_4 represents one audio frame.
The code for each divided sub-frame is shown. The input of the past synthesis filter 9 is fed back and held in the long-term prediction filter. The probability codebook 8 is fixed and stores white noise. VSEL
PO, which is a parameter in P, takes a value of 0 to 1, and
Determine the ratio of β to γ. When PO is small, γ is small and β is small. The voice encoder controls the GSPO code so that the PO for background noise reproduction becomes small. Therefore, in the voice encoding device, PO is a small value during the background noise reproduction in the silent section, so that the input of the synthesis filter 9 becomes white noise and the decoded voice also becomes white noise.

As described above, according to the above-described embodiment, the VSE is used in the voice decoding device during the background noise reproduction in the silent section.
Since the PO of the LP parameter is made small, it is possible to avoid the reproduction of an abnormal sound that is unfamiliar to the decoded sound, which is like white noise.

Although the VSELP system is used for the encoder and the decoder in the above-mentioned embodiment, other voice encoding and decoding systems may be used.

[0017]

As described above, according to the present invention,
By providing a code converter and a multiplexer in the speech coding apparatus, when an error occurs in the voice / sound deciding unit of the speech coding apparatus, an unfamiliar difference occurs when the speech decoding apparatus reproduces background noise in a silent section. The effect that an excellent digital audio transmission device capable of reducing the generation of sound can be realized is obtained.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of a digital audio transmission device according to the present invention.

FIG. 2 is a diagram for explaining the operation of the VSELP decoder in the above embodiment.

FIG. 3 is a block diagram showing a conventional digital audio encoding device.

FIG. 4 is a block diagram showing a conventional digital speech decoding apparatus.

FIG. 5 is a diagram for explaining the operation of a VOX controller.

[Explanation of symbols]

 1 Encoder (VSELP) 2 Voice / Silence Determinator 3, 15 VOX Controller 4 Code Converter 5, 17 Multiplexer 6 Multiplexer 7 Long-term Predicted State 8 Probability Codebook 9 Synthesis Filter 10 Encoder 14 Separator 16 Code controller 18 Decoder

Claims (1)

[Claims] 1. A speech coder for encoding a speech signal,
A code converter for converting the coded voice signal into a code in which abnormal noise is unlikely to occur, a voice / sound determiner for determining whether the input voice signal is a voice signal or a silence signal, and VO for controlling the state based on the judgment result of the sound silence judgment device
An X controller, a multiplexer for selecting and outputting a coded raw voice signal and a code-converted signal based on a control signal from the VOA controller, and transmission data according to a state determined by the VOA controller. A digital audio transmission device having a multiplexer for creating a processing device.