JPS62180622A - Echo preventing device for telephone - Google Patents

Abstract

PURPOSE:To improve the voice quality of a telephone call of an overseas communication system where voice is delayed largely especially and to reduce the cost by providing a voice/silence recognizing function and a voice volume comparison function to a voice digital converter so as to stop the voice conversion of either a caller or a called party. CONSTITUTION:A synthesis processing module 11 applies a frame reception processing according to the synthesis processing flow, decides the frame form, applies format conversion and stores the result in a buffer, and sends discrimination information being synthesized at present to an analysis processing module 8. The analysis processing module 8 receives a parameter from an analysis LSI 7 according to the analysis processing flow to decide whether the voice exists or not and applies initial value setting of a sound source strength A and a pitch T. At the time of the present voice is being synthesized, the echo is suppressed by attenuating or stopping the sound source strength A at the analysis side of the voice.

Description

[Detailed Description of the Invention] [Industrial Field of Application] - The present invention relates to the prevention and prevention of echoes caused by echoing of voices, and is particularly applicable to applications where voice delays are large and international lines are used. The present invention relates to a telephone echo prevention device suitable for overseas communication systems. ... [Conventional technology] Conventionally, in overseas communication systems using international lines with large voice delays, echo suppressors and echo cancellers were installed on the lines to prevent echoes. However, no consideration was given to the mounting space and cost increase.

For related technical issues in this field, see ``Speech Synthesis and Its Applications'' published in the February 1980 issue of the Journal of the Institute of Electrical Engineers of Japan.
etc.

[Problem that the invention seeks to solve]

As described above, the conventional workpiece prevention device requires a large mounting space and has the problem of high cost.
□ An object of the present invention is to provide an echo suppress function without increasing the size of the device, and to improve the voice quality of telephone calls, especially in overseas communication systems where voice delay is large, and to reduce costs.

[Means for solving problems]

The above purpose is achieved by providing an echo suppression function in the voice analysis and synthesis section, so that the voice analysis side attenuates and stops the echo generated by the loop from the transformer based on the voice discrimination information from the synthesis side. be done.

[For production]

The speech analysis side identifies whether speech is currently being synthesized or silence is being synthesized based on the discrimination information from the synthesis side. Thereby,
If the voice is currently being synthesized, the sound source strength A is used on the voice analysis side.
It suppresses echoes by attenuating or stopping them, so it works to eliminate echoes during calls.

〔Example〕

An embodiment of the present invention will be described below with reference to the drawings. 1st
The figure shows a block diagram of the speech analysis and synthesis section. In a configuration in which the voice analysis and synthesis units are connected to each other via a dedicated line or an international line, 1 is a telephone, 2 is a line circuit, a hybrid transformer (HYB) 4 performs 2-to-4 line conversion, and a C0DEC3 converts the voice. Performs digital/analog conversion. Reference numeral 5 denotes an audio analysis and synthesis section, in which the audio signal input from the CODEC 3 is converted into a linear signal by an expander 6, and is input to the analysis LSI 7. As shown in FIG. 4, the analysis processing module 8 receives parameters necessary for audio processing in response to a transfer request from the analysis LSI 7, creates an audio frame, and sends it to the line. On the other hand, the synthesis processing module 11 receives the voice frame from the line and sends information according to the data format shown in FIG. 5 to the synthesis LSI 10. In this data format, A is the sound source strength, K1- is 1
o is the partial autocorrelation coefficient and T is the pitch period. The linear signal sent out from the composite LSI 1.0 is compressed to 8 bits by the compressor 9 and input to the COIEC3. Due to the impedance mismatch of the HY B 4, the analysis/synthesis processing delay, and the propagation delay, interference occurs in the audio output from the CODEC 3, resulting in an echo. The synthesis processing module 11 performs frame reception processing according to the synthesis processing flow shown in FIG. 3, determines the frame format, performs format conversion, stores it in a buffer, and sends discrimination information currently being synthesized to the analysis processing module 8. do. The analysis processing module 8 receives parameters from the analysis LSI 7 as shown in the analysis processing flow shown in FIG.

Set initial values for degree A and pitch T. Further, the sound source intensity A is corrected (attenuated) based on the synthetic sound discrimination information received from the synthesis processing module 11 to perform echo suppression. The correction method is to use the synthetic voice discrimination information and the echo duration of 20 milliseconds XJ (frames) to calculate the sound source strength correction value A' = A ÷
Let it be 2J. A represents the amplitude intensity of the sound source using a two's complement number.

If the bit length of A - 1 frame echo continues, the transmission stops completely. According to this embodiment, since the voice analysis and synthesis processing section has an echo suppression function, the echo
It has the effect of removing echoes without providing a suppressor or echo canceller. When shown in this example, the quantity relationships and data formats are merely examples.

〔Effect of the invention〕

-According to the present invention, in an overseas communication system that accommodates telephone calls using dedicated lines or international lines, it is possible to eliminate echoes that occur due to voice interference in a hybrid transformer due to voice analysis and synthesis processing delays and line transmission delays. Therefore, by providing an echo suppression function in the microprogram of the speech analysis and synthesis section, it is possible to reduce the noise caused by echoes and to have a smooth conversation.

[Brief explanation of drawings]

FIG. 1 is a block diagram of a speech analysis and synthesis section according to an embodiment of the present invention, FIG. 2 is a flowchart showing analysis processing, FIG. 3 is a flowchart showing synthesis processing, FIG. 4 is a time chart, and FIG. 5 is a flowchart showing analysis processing. FIG. 3 is a diagram showing a composite LSI interface data format. 1...Telephone, 2...Line circuit,'□3...C
0DEC, 4...HYB, 5...Speech analysis and synthesis section,
6... Stretcher, 7... Analysis LSI', 8...□
Analysis processing module, 9...Compressor, 10...-Synthesis LSI, 11...Synthesis processing module. □ □・

Claims (1)

[Claims] In a voice communication system consisting of a leased line, a public network, and a voice digital converter, the voice digital converter is equipped with a voice/silence recognition function and a volume comparison function to stop voice conversion for either the sender or the receiver. By doing so, the said leased line,
A telephone echo prevention device characterized by reducing noise caused by echoes generated on public networks.