JPH0697898A - System and device for voice data communication - Google Patents

Abstract

PURPOSE:To perform normal switching from silence to voice or from voice to silence regardless of a failure in reception of a notice signal with respect to the voice data communication equipment using the VOX control system which detects voice/silence of an audio signal to control the data transfer. CONSTITUTION:A transmission part A is provided with an encoding part 1 which encodes voice data, a discriminating part 2 which discriminates voice/ silence of encoded data and adds notice information to encoded data to output it, and a sending part 3 which receives encoded data with notice information and shortens the burst length of data in the case of silent data to send it to a reception part B, and the reception part B is provided with a substituting part 4 which discriminates silent data by the burst length of data received from the transmission part A and substitutes silent data with encoded data which brings about a CRC error, and a detecting part 5 which receives the output of the substituting part 4 to discriminate notice information included in encoded data and checks the CRC error, and a decoding part 6 which decodes encoded data to voice data in the case of voice and generates noise in the case of silence in accordance with the detection result of the detecting part 5.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice data communication apparatus using a VOX control system for detecting voiced / non-voiced voice signals and controlling data transfer, and a system therefor.

[0002]

2. Description of the Related Art In recent years, voice data communication devices have been required to reduce power consumption in response to demands for cost reduction and device miniaturization. Therefore, there is known a VOX control system in which the presence / absence of a sound signal is detected, and when there is no sound, the CODER or the like on the transmission side is stopped to reduce power consumption.

In the conventional voice data communication system using the VOX control system, the transmitting side detects sound / silence, and when switching from silence to sound / voice to silence
A notice signal is sent before the frame. On the receiving side, the notice signal is received and switching between sound processing and silence processing is performed.

[0004]

However, if this notice signal cannot be received due to noise on the transmission line,
The receiving side cannot switch between voiced processing and silence processing, and even if voiced data is being transmitted, it will perform silence processing, or will perform voiced processing despite the transmission of silence data. There was a problem.

According to the present invention, in addition to the advance notice signal, CRC (Cyclic Redundancy Check) monitoring is added to determine whether the currently received signal is voiced data or silent data,
Provided is a voice data communication device and a method thereof, which can normally switch from silence to sound, or from sound to silence even if a notice signal cannot be received.

[0006]

FIG. 1 is a block diagram showing the principle of the present invention. As shown in FIG. 1, the present invention comprises a transmission unit A and a reception unit B. The transmission unit A discriminates between an encoding unit 1 for encoding audio data and a presence / absence of the encoded data. A discriminating unit 2 that adds the advance notice information to the encoded data and outputs the encoded data, and a transmission unit 3 that receives the encoded data with the advance notice information and shortens the burst length of the data for the silent data and transmits the data to the receiving unit B. Prepare

The receiving unit B determines the silent data from the burst length of the data received from the transmitting unit A, and CR determines the silent data.
A replacement unit 4 for replacing the coded data causing a C error,
A detection unit 5 that receives the output of the replacement unit 4 to discriminate the advance notice information included in the encoded data and check a CRC error; Decoding unit 6 for generating
The present invention provides a voice data communication device provided with.

Further, according to the present invention, in the communication system for transmitting voice data from the transmitting unit A to the receiving unit B, the transmitting unit A encodes the voice data to add the notice information of voiced and voiceless and the voiceless data. With respect to, the encoded data in which the burst length is shortened is sent to the receiving unit B.

The receiving section B replaces the received data having a short burst length with the data which causes a CRC error, checks the advance notice information and the CRC error, and decodes the data when there is sound based on the result. The present invention provides a voice data communication system characterized by generating a predetermined noise when there is no sound.

[0010]

In FIG. 1, in the transmission unit A, the encoding unit 1 encodes the voice data, the discrimination unit 2 discriminates the presence / absence of the encoded data, and outputs the advance notice information to the transmission unit 3.
The sending section 3 outputs the result to the discriminating section 3. The sending unit 3 converts the data from the discriminating unit 3 into a packet, and when the data is silent, does not insert the encoded data into the packet and shortens the burst length and sends the packet to the receiving unit B.

In the receiving section B, the replacing section 4 judges silence data from the burst length of the received data, and if the burst length is short, the coded data causing a CRC error is sent to the detecting section 5, and otherwise. If so, the encoded data is sent as it is. The detection unit 5 detects the advance notice information and determines whether there is sound processing or silence processing.
An error check is performed, and if a CRC error does not occur, control is performed to return to sound processing. Decoding unit 6
Then, in response to the result of the detection unit 5, the voice data is decoded (when there is sound) or noise is generated (when there is no sound).

As described above, the burst length is monitored by the receiving unit, and the data is fixed so that a CRC error is compulsorily generated at the time of silent data (when the burst length is short).
Therefore, if the receiving unit B is controlled so as to return to the sound processing when the data in which the CRC error does not occur is received,
Even if the silence / voice announcement information is not received, the voice processing can be resumed when the voice data is received.

Similarly, when switching from voiced to silence,
Even if the advance notice information is not received, the silent data is converted into the data in which the CRC error occurs, so that the process of reducing the sound level is performed and the silent sound is output as a result.

[0014]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail with reference to the embodiments shown in the drawings. This does not limit the invention. FIG. 2 is a block diagram showing an embodiment of the present invention, in which A is a transmitter and B is a receiver.

In the transmission section A, 11 is an encoding section for encoding input voice data, 12 is a voice / silence determination section for determining whether the encoded data is voiced or silenced,
Reference numeral 13 is a first data generation unit that generates a switching notice signal from sound to silence, 14 is a second data generation unit that generates a switching notice signal from silence to sound, and 15 is a sound / silence determination unit 12
The encoding unit 11 and the first data generation unit 1 according to the determination result of
3, or a selector for selecting and outputting the output of the second data generating unit 14, and 16 is a line interface for sending the output of the selector 15 to the receiving unit B.

In the receiving section B, 17 is a line interface for extracting the encoded data from the transmitting section A, 18 is a burst length monitoring section for monitoring the length of the communication burst and detecting silent data, and 19 is a CRC error. A third data generation unit 20 for generating data for generating a line is generated by the line interface 17 according to the detection result of the burst length monitoring unit 18.
Alternatively, it is a selector that selects and outputs the output of the third data generation unit 19.

Reference numeral 21 is a warning signal detecting portion for detecting the warning signal generated by the first data generating portion 13 or the second data generating portion 14, and 22 is a CRC error for detecting the data generated by the third data generating portion 19. A check unit, 23 is a fourth data generation unit that generates noise generation data, and 24 is a detection result of the notice signal detection unit 21 and a CRC error check unit 2
A selector for selecting and outputting coded data or noise generation data based on the detection result of 2, and a decoding unit 25 for decoding the output of the selector into audio data.

In such a structure, the transmitter A
The voice data encoded by the encoding unit 11 is determined by the voice / non-voice determination unit 12 as voice / non-voice based on its power parameter, and the selector 15 encodes the encoded data, a voice advance notice signal for switching from voice to silence. Any one of the announcement signals for switching from silence to sound is selected, converted (packetized) into line data by the line interface unit 16 and sent to the receiving side. However, here, the burst length is shortened without inserting the encoded data into the packet when there is no sound.

On the receiving side B, the line interface 17
The encoded data is extracted in step S1, and the burst length monitor 18 monitors the length of the communication burst to detect silent data. Based on the detection result, the selector 20
Coded data is selected when there is sound, and data that causes a CRC error when there is no sound is selected and output.

The warning signal detecting section 21 detects a warning signal for switching from silence to sound and to switch from sound to silence, and CR
The C error check unit 22 detects a CRC error in the input data, and the CRC is continuously detected a predetermined number of times or more when there is no sound.
When no error is detected, a voiced switching signal is sent to the selector 24. The selector 24 includes the warning signal detector 21 and C
Coded data or noise generation data is selected from the detection result of the RC error check unit 22. The output of the selector 24 is decoded by the decoding unit 25 and output as audio data.

[0021]

According to the present invention, in voice data communication using the VOX control system, even if the advance notice signal for switching from silence to sound from the transmission unit cannot be received due to noise on the transmission path or the like, the reception unit. Can properly process the voiced data received thereafter.

[Brief description of drawings]

FIG. 1 is a block diagram showing the principle of the present invention.

FIG. 2 is a block diagram showing an embodiment.

[Explanation of symbols]

 A transmitter B receiver 11 encoding unit 12 voice / silence determination unit 13 first data generation unit 14 second data generation unit 15 selector 16 line interface 17 line interface 18 burst length monitoring unit 19 third data generation unit 20 selector 21 advance notice signal detection section 22 CRC error check section 23 fourth data generation section 24 selector 25 decoding section

Claims (2)

[Claims] 1. A transmission section (A) and a reception section (B), wherein the transmission section (A) discriminates between a voiced part and a silenced part of the coded data and a coding part (1). A discriminator (2) that adds the advance notice information to the encoded data and outputs the encoded data.
And a sending section (3) for receiving the encoded data with advance notice information and reducing the burst length of the silent data and sending the data to the receiving section (B). A) determining a silent data from the burst length of the data received from A) and replacing the silent data with encoded data that causes a CRC error; and a replacement unit (4) output to obtain encoded data. A detection unit (5) for discriminating the advance notice information included therein and for checking a CRC error, and a decoding unit (6) for decoding voice data in the presence of voice and generating noise in the absence of voice according to the detection result of the detection unit (5). ) Voice data communication device equipped with. 2. In a communication system for transmitting voice data from a transmitter (A) to a receiver (B), the transmitter (A) comprises:
The voice data is encoded to add the announcement information of voiced and silent voices, and for the voiceless data, the encoded data of which the burst length is shortened is sent to the reception unit (B), and the reception unit (B) receives the received data. For data with a short burst length, data with a CRC error is replaced, and then the advance notice information and the CRC error are checked. Based on the result, the data is decoded when there is sound and a predetermined noise is generated when there is no sound. A voice data communication method characterized in that