JPH02305153A - Automatic gain controller for voice transmitter-receiver - Google Patents

Abstract

PURPOSE:To obtain optimum talking state by detecting a level of noise included in transmission reception sound and transmission respectively busy tone and employing fuzzy control so as to adjust the gain of a transmission reception tone amplifier circuit depending on each obtained level and a desired gain adjustment degree to be inputted. CONSTITUTION:A reception level extraction section 11 extracts a level of a received signal tone itself, a difference extraction section 13 extracts a change in the inputted received level signal x5 and the change is sent to a fuzzy deduction section 9 as a signal x8. A noise extraction section 12 extracts a level of noise such as a background tone included in the inputted reception signal and sends it to the fuzzy inference section 9 as a signal x6. The fuzzy inference section 9 calculates an optimum output based on a preset fuzzy rule according to the fuzzy inference and converts the result into a non-fuzzy signal and sends the gain change for transmission reception to a gain control section 15 as output signals y1, y2. Thus, the excellent talking state is always maintained corresponding to the change in the environment quickly.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to an automatic gain control device used in audio transceivers such as video conferences and hands-free telephones.

(Summary of the Invention) This invention detects the levels of transmitted and received sounds and noise contained in the transmitted and received sounds in an audio transmitter/receiver, and uses the obtained levels and the input desired degree of gain adjustment. This system uses fuzzy control to adjust the gains of the transmitting and receiving sound amplification circuits to obtain the optimal call state.

(Prior art and its problems) Conventionally, the gain in this type of audio transmitter/receiver has been controlled manually by adjusting the volume or the like, or by automatically controlling the transmission gain according to the strength of the line current.

In the former case, the gain is simply adjusted, and it is impossible to perform sophisticated control that takes into consideration the presence or absence of ambient noise, intensity, etc.

In the latter case, there are many factors to consider when controlling the gain in consideration of the surrounding sound environment, resulting in a complicated and large circuit configuration and a slow processing speed.

Furthermore, even if the gain obtained in this manner is the same value, there is a problem in that the perceived volume differs depending on individual differences between callers.

(Purpose of the Invention) This invention was made to solve the above problems, and its purpose is to provide a simple configuration that can quickly respond to changes in the environment and always maintain good communication quality. An object of the present invention is to provide an automatic gain control device for an audio transmitter/receiver that can maintain the state.

(Structure and Effects of the Invention) In order to achieve the above object, the present invention includes a level detection circuit that detects a transmitted voice and a noise level in the transmitted voice, and a level detection circuit that detects a received voice and a noise level in the received voice, respectively. A means for inputting a desired degree of adjustment for the gain of the level detection circuit to be detected and the transmitting/receiving sound amplifying circuit, and a means for inputting the desired degree of adjustment for the gain of the detected level detection circuit and the transmitting/receiving sound amplification circuit, and member simple calculation of the control amount of the gain of the transmitting sound and the receiving sound from each detected level and the desired adjustment degree. The present invention is characterized by comprising an inference means for calculating by fuzzy control using a function, and a means for increasing or decreasing the gain of the transmitting/receiving sound amplification circuit based on the calculated gain control mfl.

In this way, the present invention detects the level of the transmitted/received sound and the level of noise contained in the transmitted/received sound in the audio transmitter/receiver, and performs fuzzy calculation based on the obtained level values and the input desired degree of gain adjustment. By adjusting the gains of the transmitting/receiving sound amplification circuit using the control, it is possible to always maintain an optimal call state.

In particular, if the level of harmful background sound contained in the speaker's sound fluctuates, the gain can be quickly adjusted to always minimize the influence of the background sound.

Furthermore, the user's subjective preference for the call state can be arbitrarily set by inputting the desired degree of adjustment.

Furthermore, since this device uses transmitting and receiving calls and the noise level contained therein as input, it is highly compatible with circuit configurations such as video conferences and hands-free telephones that require controlling the transmitting and receiving gains. Since the current is high, there is an advantage that the circuit can be shared.

(Explanation of Examples) Next, embodiments of the present invention will be described based on the drawings.

FIG. 1 is a block diagram showing the configuration when the device of the present invention is applied to a telephone.

In the figure, the handset l has a microphone 2 and a speaker 3.
The transmitted speech sound inputted into the microphone 2 is converted into an audio signal and then sent to the transmitting amplifier 4, the transmitted speech level extraction section 6, and the noise extraction section 7.

The transmission amplifier 4 amplifies the audio signal and then sends it to the outside as a transmission signal.

On the other hand, a received signal sent from an external device (not shown) is received by a receiving amplifier 5, a reception level extractor 11, and a noise extractor 1.
Sent to 2.

The receiving amplifier 5 amplifies the received signal and then transmits it to the speaker 3.
and convert it into audio.

The transmission level extraction unit 6 extracts the level of the transmission sound itself of the input transmission sound signal, and outputs the level of the transmission sound itself as a signal xi to the difference extraction unit 8,
Send it to the fuzzy inference section 9.

The difference extracting unit 8 extracts the amount of change in the input transmission level signal X1, and sends the amount of change to the fuzzy inference unit 9 as a signal x4.

The noise extractor 7 extracts the level of noise such as background sound contained in the input transmission sound signal and sends it to the fuzzy inference section 9 as a signal x2.

The listening level extraction unit 11 extracts the level of the listening sound itself of the input reception signal, and sends it to the fuzzy inference unit 9 and the difference extraction unit 13 as a signal x5.

The difference extraction unit 13 extracts the amount of change in the input listening level signal x5, and sends the amount of change to the fuzzy inference unit 9 as a signal x8.

The noise extraction section 12 extracts the level of noise such as background sound contained in the input received signal, and sends it to the fuzzy inference section 9 as a signal x6.

When the desired gain adjustment degree input section 14 receives the content of the gain adjustment degree desired by the user, it sends it to the fuzzy inference section 9 as a signal x9.

The fuzzy inference unit 9 calculates the goodness of fit (membership (II)) of the membership functions of the input signals x1 to x9, and then
After calculating the optimal output value based on preset fuzzy rules by fuzzy inference and defuzzifying the result, the output signal! As 1.12, the amount of change in gain for transmitting and receiving is sent to the gain control unit 15.

Here, the fuzzy rule stored in the fuzzy inference unit 9 has the input detection level, current gain, and desired gain adjustment degree as the antecedent part, and the amount of change in gain as the consequent part. , as an example, transmitting level x1. The rule for determining the amount of change y1 in the transmission gain from the noise level X2 of the transmission sound and the current transmission gain x3 is as follows.

if xl=Ps x2-ZRx3=Psthen
yl-PL i[xi-PS x2-ZRx3-PMthen
yl=PM if xi-PS x2=ZRx3-PLthen
yl-PS if xi-PS x2=Ps x3-P
Then yl=PL if xl=Ps x2=PM x3-P
Then yl=Ps if xi-PS x2-PL x3=P
then yl-ZR jf xi-PM x2-PL x3=P
LLhen yl-NM if xi-PL x2=PL x3-P
Lthen yl-NL if xl=PL x2-PM x3=P
Mthen yl-NS Furthermore, the rule for determining the change amount y2 of the listening gain may be obtained by replacing the antecedent parts x1 to x3 with x5 to x7.

Next, the gain control unit 1 receives the output signals yl and y2.
5 changes the gains held until then based on the amount of change in the transmitting and receiving gains indicated by the signals yl and y2, and sends the new gains to the transmitting and receiving amplifiers 4.5, respectively, and The current gain is also sent to the fuzzy inference unit 9 by x3. Send as x7.

The transmitting/receiving amplifiers 4.5 each have a gain set to a value output by the gain control section 15.

The transmitted/received signals thus input are maintained in a good conversation state by passing through the transmitting/receiving amplifier 4.5 whose gain is adjusted to the optimum gain for the current conversation state.

FIG. 2 shows an input signal X1 that is manually input to the fuzzy inference section 9.
This is a graph showing a membership function used when fuzzifying ~X9, and the degree of suitability of each fuzzy label is determined from the level of each input signal.

FIG. 3 shows the calculation results of the fuzzy inference unit 9 as concrete outputs y1. It is a graph showing a membership function used when defuzzifying y2.

Further, since this device is equipped with a desired gain adjustment degree manual control unit 14, it is possible to specify the degree of gain adjustment so as to obtain desired characteristics in accordance with the auditory characteristics that differ for each individual user.

Further, when this device is used for video conferences and hands-free telephone calls, adding a switching function for sending and receiving calls in addition to the above-mentioned control contents can prevent howling from occurring.

Furthermore, the amount of level change x4 extracted by the difference extraction unit 8.13. Regarding x8, for example, if xi becomes positive and the transmitting level x1 is increasing, but the receiving level X5 does not change, it is determined that it is difficult to hear, and control content is added, such as increasing the receiving gain a little. You can also do that.

As described above, this embodiment uses fuzzy control for gain control for an audio transmitter/receiver, which conventionally has a complicated circuit configuration, and it is possible to quickly respond to various situations with a simple circuit configuration. Therefore, it is possible to realize a high-performance automatic gain control device that can perform high-performance automatic gain control, and the range of use of this type of audio transceiver can be greatly expanded.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of an embodiment in which the inventive device is applied to a telephone, and FIGS. 2 and 3 are graphs showing membership functions set in the fuzzy inference section. l... Handset 2... Microphone 3... Speaker 4... Transmitting amplifier 5... Receiving amplifier 6... Transmitting level extraction section 7... Noise extraction section 8... Difference Extraction unit 9... Fuzzy inference unit 11... Reception level extraction unit 12... Noise extraction unit 13... Difference extraction unit 14... Desired gain adjustment degree input unit 15... Gain control unit

Claims (1)

[Claims] 1. A level detection circuit that detects the transmitted voice and the noise level in the transmitted voice, a level detection circuit that detects the received voice and the noise level in the received voice, and the transmitted and received voices. A means for inputting a desired degree of adjustment for the gain of the amplifier circuit, and an inference for calculating the control amount of the gain of the transmitted voice and the received voice from each detected level and the desired degree of adjustment by fuzzy control using a membership function. An automatic gain control device for an audio transmitter/receiver, comprising: means for increasing or decreasing the gain of the transmitting/receiving sound amplifying circuit according to the calculated gain control amount.