JPH01245660A - Houling suppressor - Google Patents

Abstract

PURPOSE:To suppress houling automatically by inferring, in the case there is a difference between and output signal from a first subtractor and that from a second subtractor, the respective impulse responses by a first and a second impulse response inferring circuit so as to make said difference come to zero, and generating a pseudo echo. CONSTITUTION:If the ambient temperature varies or preliminarily-inferred impulse responses H1, H2 varies because of the moving of the positions of microphones 1, 2 and a speaker 6, the output level of the subtractor 3a changes. The impulse response inferring circuits 3, 4 newly speculates the impulse responses H1, H2 between the microphones l, 2 and the speaker 6 in a learning identification method so as to make the output level of the subtractor 3a zero. Accordingly, houling can be suppressed by means of a pseudo echo generated with the new impulse responses H1, H2. In such a way, houling can be automatically suppressed even if the ambient temperature varies or the positions of the microphones and the speaker moves.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a howling suppression device used in a public address system equipped with a microphone, an amplifier, and a speaker.

Conventional technology FIG. 2 shows the configuration of a conventional howling suppression device.

In FIG. 2, reference numeral 12 denotes a microphone for manually inputting the voice 7 etc. from the speaker. There is a possibility that it will wrap around and cause howling.

11 is a circuit for outputting a sample signal such as white noise; 13 is for estimating the impulse response H between the microphone 12 and the suspension speaker 6; The changeover switch 5 is an amplifier that amplifies the signal from the changeover switch and outputs it to the speaker 6.

Next, the operation of the above conventional example will be explained.

In FIG. 2, first, the positions of the microphone 12 and the speaker 6 are set, and the sample signal output circuit 11 is
The changeover switch 10 is switched so that the amplifier 5 is connected to the amplifier 5.

Therefore, the sample signal is amplified and widened by the amplifier 5 and reproduced by the speaker 6, and is routed to the microphone 12 according to the positions of the microphone 12 and the speaker 6.
The surrounding voice 14 is input.

In this case, the microphone 12 is set so that no other voice is input.

In this state, the impulse response estimation circuit 13 estimates the impulse response H between the microphone 12 and the speaker 6 using the learning identification method so that the difference between the output signal of the microphone 12 and the pseudo echo becomes "0".

Next, when the selector switch 10 is changed so that the microphone 12 side is connected to the amplifier 5 and used as a public address system, the impulse response estimation circuit 13 convolves the estimated impulse response H and the input signal of the amplifier 5. This generates a false echo, causing microphone 1 to
By subtracting this pseudo echo from the output signal of No. 2, the nowing is suppressed.

Problems to be Solved by the Invention However, in the conventional howling suppression device described above, howling is suppressed after estimating the impulse response between the microphone and the speaker in advance. This causes a problem in that the impulse response estimated in advance changes, making it impossible to suppress howling.

SUMMARY OF THE INVENTION In view of the above-mentioned problems of the prior art, it is an object of the present invention to provide a howling suppression device that can automatically suppress howling even when the ambient temperature changes or the positions of a microphone and a speaker move. .

Means for Solving the Problems In order to achieve the above object, the present invention includes first and second microphones arranged so that the sound from the speaker is input at the same level, and an amplifier that amplifies the audio signal and outputs it to the speaker;
first and second impulse response estimating circuits that estimate an impulse response using the output signal of the microphone and the input signal of the amplifier, respectively, and generate a pseudo echo based on the estimated impulse response; first and second subtracters that subtract pseudo echoes from the first and second impulse response estimation circuits from the output signal, and when the difference between the output signals of the first and second subtracters is different; In addition, the first and second impulse response estimating circuits estimate their respective impulse responses so that the difference becomes rOJ, thereby generating a pseudo echo.

Effect The present invention has the above-mentioned structure. J+W+ degree changes 6, -7
, the impulse responses to the second microphone are different, and the difference between the output signals of the first and second subtracters becomes different.

Therefore, when the difference between the output signals of the first and second subtracters is different, the first and second impulse response estimation circuits re-estimate their respective impulse responses so that the difference becomes "0". However, by generating a pseudo echo from this re-estimated impulse response C2, howling can be automatically suppressed even if the ambient temperature changes or the positions of the microphone and speaker move.

EXAMPLES Hereinafter, the present invention will be described in detail with reference to the drawings. FIG. 1 is a schematic block diagram showing an embodiment of a howling suppression device according to the present invention, and the same components as those shown in FIG. 2 are given the same reference numerals.

In FIG. 1, 1 and 2 are voices 7 from the speaker, respectively.
It is a microphone for inputting information such as 77.

The microphones 1 and 2 are arranged so that the voice 7 from the speaker is output at the same level.

11 is a circuit that outputs a sample signal such as white noise, and 3 and 4 are impulse response estimation circuits that estimate impulse responses H1 and l-12 between the microphone 1゜2 and the speaker 6, respectively, and generate pseudo echoes. It is.

1a and 2a are subtractors that subtract pseudo echoes from impulse response estimation circuits 3 and 4 from the output signals of microphones 1 and 2, respectively; 3a is a subtractor that subtracts the signal from subtracter 2a from the signal from subtractor 1a, and calculates the impulse response This is a subtracter that outputs to the response estimation circuits 3 and 4.

10 selectively selects the difference between the audio signal from the microphone 1 and the pseudo echo from the impulse response estimation circuit 3 or the signal from the sample signal output circuit 11. Next, the operation of the above embodiment will be explained.

In FIG. 1, first, the changeover switch 10 is switched so that the sample signal output circuit 11 is connected to the amplifier 5.

Therefore, the sample signal is amplified by the amplifier 5 and reproduced by the speaker 6, and is routed to the microphones 1, 2 and 2 according to the positions of the microphones 1, 2 and the speaker 6.
The surrounding voices 8 and 9 are respectively input. In this case, the microphones 1 and 2 are set so that no other voices are input.

In this state, the impulse response estimation circuits 3 and 4 calculate the difference between the output signals of the microphones 1 and 2 and the pseudo echo as "
The impulse response Hh between the microphone 1.2 and the loudspeaker 6 is 0'.

H2 is estimated using the learning identification method.

Next, when the selector switch 10 is changed so that the microphone 1 side is connected to the amplifier 5 and used as a public address system, the impulse response estimating circuit 3.4 calculates the above estimated impulse responses H1 and H2 and the input signal of the amplifier 5. A pseudo echo is generated by performing a convolution operation.

9.

Therefore, the pseudo echo from the impulse response estimating circuit 3 is subtracted from the output signal of the microphone 1 to suppress howling, and if the same level of sound 7 is input to the microphones 1 and 2, the output of the subtracter 3a The level is "0".

In this case, if the ambient temperature changes or the microphones 1 and 2
When the position of the speaker 6 moves and the pre-estimated impulse responses H1 and H2 change, the output level of the subtractor 3a changes.

The impulse response estimation circuits 3 and 4 operate the microphones 1 and 2 so that the output level of the subtracter 3a becomes "0".
Impulse responses H1 and H2 between the speaker 6 and the speaker 6 are newly estimated using a learning identification method.

Therefore, howling can be suppressed by the pseudo echo created by the new impulse responses H2 and H2, and howling can be automatically suppressed even if the ambient temperature changes or the positions of the microphone and speaker move.

DETAILED DESCRIPTION OF THE INVENTION As described, the present invention
．． −.

first and second microphones arranged so that voices are input at equal levels; an amplifier that amplifies the audio signal from the first microphone and outputs it to the speaker;
The first and second microphones each estimate an impulse response using the output signal of the second microphone and the input signal of this amplifier, and generate a pseudo echo based on the estimated impulse response.
an impulse response estimation circuit, and first and second subtracters that subtract pseudo echoes from the first and second impulse response estimation circuits from the output signals of the first and second microphones, 1. When the difference between the output λ1 output signals of the second subtractor is different, the first and second impulse response estimating circuits estimate their respective impulse responses so that the difference becomes 10'', and generate a pseudo echo. Since this method generates howling, it is possible to automatically suppress howling even if the ambient temperature changes or the positions of the micropon and speaker move.

[Brief explanation of the drawing]

FIG. 1 shows 1.1 of the howling suppression device according to the present invention.
,,,-. A schematic block diagram showing one embodiment. FIG. 2 is a schematic block diagram showing a conventional howling suppression device. 1.2... Microphone la, 2a, 3a
・・Subtractor, 3.4・・Impulse response estimation circuit, ・
Amplifier, 6...Speaker, 10. Selector switch, 11...Sample signal output circuit.

Claims (1)

[Claims] first and second microphones arranged so that voices from the speaker are input at equal levels; an amplifier that amplifies the voice signal from the first microphone and outputs it to the speaker; first and second impulse response estimating circuits each estimating an impulse response based on the output signal of the second microphone and the input signal of the amplifier, and generating a pseudo echo based on the estimated impulse response; From the output signal of the microphone, the first and second
and first and second subtracters for subtracting a pseudo echo from the impulse response estimation circuit, and when the difference between the output signals of the first and second subtracters is different, the first and second A howling suppression device characterized in that each of the impulse response estimating circuits estimates the respective impulse responses so that this difference becomes "0", and generates a pseudo echo.