JPS6218836A - Method and circuit for suppression of howling - Google Patents

Abstract

PURPOSE:To suppress howling and obtain optimum sound volume by using an echo erasing means which inverts the phase of a pseudo echo signal and adds it to a transmission system and a frequency shifting circuit which shifts the frequency component of a signal passed through the signal passage only by a specific value in combination. CONSTITUTION:An echo canceler 17 as an echo adding means which generates the pseudo echo signal of the echo signal reaching the transmission system from a signal arriving at a reception system and adding the pseudo echo signal to the transmission system after inverting its phase and the frequency shifting circuit 21 which shifts the frequency component of the signal passed through the signal passage by the specific value are used in combination. When the frequency shifting circuit 21 is inserted into the signal passage of the reception system, a frequency shifting function is stopped in the operation of the adapting operation function of the echo canceler and started at the stop of the adapting operation function and when the frequency shifting circuit 21 is inserted into the signal passage of the transmission system, the frequency shifting function is started except during transmission. Consequently, the adapting operation and frequency shifting operation of the echo canceler are controlled properly in a no-call state, a transmission state, a reception state, and a duplex call state.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a method and a circuit for preventing howling occurring in a communication system where transmission and reception are separated.

Here, "howling" refers to a situation in which a signal from the receiving system enters the transmitting system for some reason, causing the entire system to oscillate.

〔overview〕

The present invention provides a howling suppression method and circuit for preventing howling occurring in a communication system in which transmitting and receiving are separated, by generating a pseudo-reverberant signal of a reverberating signal that circulates from a signal arriving at a receiving system to a transmitting system. The echo canceling means that inverts the phase of the pseudo echo signal and adds it to the transmitting system, and the frequency shift circuit that shifts the frequency component of the signal passing through the signal path by a predetermined value are used together, and the echo canceling function is installed in the loop. By functioning at least one frequency shift function, it is possible to eliminate the lack of speaker volume that existed in the howling suppression circuit using the echo cancellation method, and to obtain the optimal volume while suppressing howling. .

[Conventional technology]

In order to simplify the explanation, the audio system of the video conference telephone device will be described below. FIG. 4 is a block diagram showing the audio system of the video conference telephone device. FIG. 4 (al indicates an example of two-way communication, and FIG. 4 Fkl) shows an example of four-way communication.

Generally, the audio system of a video conference phone device consists of a speaker (loudspeaker) that allows multiple people to hear the other party's voice at the same time, a microphone that collects the voices generated by multiple people, and an amplifier that drives the speaker and microphone. Ru.

In such a configuration, reverberant sound caused by the sound from the speaker reflecting off the walls of the room may enter the microphone and be inserted as a reverberant signal. Furthermore, a similar phenomenon occurs in the audio system of the other party's video conference phone equipment, and when this echo signal forms a one-round loop, the gain of the amplifier is increased to increase the sound level from the speaker. - It is well known that the gain of a circular loop exceeds 1 and causes howling.

In addition, in a configuration where multiple destinations are connected simultaneously as shown in Fig. 4fb), even if the open-loop gain is less than 1 in opposite communication, multiple echo signals can be generated by connecting multiple destinations via the audio adder circuit. The sum of the loop gains due to the loops exceeds 1, which also causes howling. The details are described in "Design Conditions of Relay Device in Audio Conference Communication System" IEICE Communications System Study Group Material C383-81, August 1983.

As described above, the audio system of a video conference telephone device requires means for preventing howling, regardless of the configuration of the communication system. As a means to prevent howling in the communication system, (a) echo suppression (echo suppressor) method (b)
There is an echo canceller method, and the voice communication terminal device for howling prevention on one side of the opposing video conference telephone device will be described below.

(al) Echo suppression method A device for realizing the anti-Wtffl suppression method includes a voice switch circuit or an echo suppressor, which has been commonly used for a long time.

FIG. 5 is a block diagram showing a howling suppression circuit using an echo suppressor.

Reverberation signal 3 going around from speaker 1 to microphone 2
and an amplifier or attenuator 5.6.7.8 that sets a level diagram so that the speaker's voice signal 4 input directly into the microphone 2 can be distinguished by measuring the power of each signal. This is achieved by inserting them into the transmitting side and the receiving side, respectively. An example of this is described in [“Conference Telephone” Foreign Communication Technology No., January 201969].

The echo suppressor 9 shown in FIG.
．． 11 and a control circuit 12. The signal power at the received signal input terminal Rin 13 and the transmitted signal input terminal Sin 15 is measured, and the signal power at the received signal input terminal Rin 15 is measured.
Reception level LRin and transmission signal input terminal Sin at 13
The transmission level L Sin at 15 is L Rin >
When L 5inO, the attenuation amount of the variable attenuation circuit 10 is minimized,
The attenuation amount of the variable attenuation circuit 11 is set to the maximum, and LRin=
When LSin, the attenuation amount of the variable attenuation circuit 10 and the attenuation amount of the variable attenuation circuit 11 are each set to 6 dB, and LR4n<
When LSin, the attenuation amount of the variable attenuation circuit 10 is maximized,
Control circuit 1 that sets the attenuation amount of the variable attenuation circuit 11 to the minimum
2, the loop gain is controlled to be 1 or less,
Prevents howling.

(b) Echo Cancellation Method An echo canceller is a well-known device that implements the echo cancellation method.

FIG. 6 is a block diagram showing a howling suppression circuit using an echo canceller.

The echo canceller 17 generally includes an adaptive control filter 18
, a subtraction circuit 19, and a detection circuit 20 that instructs to start or stop the adaptive control operation.

The detection circuit 20 determines that there is an audio signal input to the reception signal input terminal R4n 13 by comparing the reception level L Rin at the reception signal input terminal Rin 13 and the transmission level L Sin at the transmission signal input terminal Sin 15. When it can be determined that there is no voice signal input from the speaker to the input terminal Sin 15, the output of the subtraction circuit 19 (residual signal)
The transfer function of the adaptive control filter 18 is controlled so that the power is always close to O.

If it is determined that there is an audio signal input from the caller,
The operation of the adaptive control filter 18 is stopped.

At this time, the received signal becomes an echo signal 3 via the echo path and is mixed with the speaker's voice signal 4, but the adaptive control filter 18 outputs a pseudo echo signal generated from the received signal. The speaker's voice signal is extracted from the output of the subtraction circuit 19 from the subtraction circuit 19 .

In such an echo prevention method or an echo cancellation method, it is necessary to identify whether the current communication system is always in a transmitting state or a receiving state.

On the other hand, in addition to preventing howling in the communication system, there is a frequency shift method as a means for preventing howling in the sound field system caused by microphones and speakers in a performance venue.

Devices that implement this method are called frequency shifters, pinch controllers, or harmonizers.

Harmonizers are used at concerts and lectures to enhance the acoustic effect, and their operating principle is to memorize the input signal for a preset amount of time, and then store the stored information at a slower speed than the time rate at which the input signal was captured. Alternatively, the frequency is shifted by speeding up the output.

In general, if the pinch ratio, which is the ratio of the changed frequency to the original frequency, is increased, the characteristics of the audio will change, but if the pitch ratio is set to a level that does not change the audio quality, there will be no problem, and the reverberation signal in the room will change. Wavy frequency characteristics can be improved.

[Problem that the invention seeks to solve]

However, such conventional howling suppression methods have the following problems.

In the echo suppression method, the echo suppressor 9 inserts and removes the amount of attenuation depending on the level of the transmitting signal and the receiving signal in order to suppress the echo, and it takes an operating time of several m5ec to insert and remove the amount of attenuation. This has the drawback that it takes time to recover, and as a result, there are interruptions in the beginning and end of words or in the beginning and end of sentences, making it difficult to speak.

This drawback can be improved to some extent by setting the maximum insertion attenuation to a small value, but it also has the drawback that the volume from the speaker cannot be increased, resulting in a situation where the other party's voice cannot be heard. .

In the echo cancellation method, unlike the echo suppressor 9, the echo canceller 17 does not insert or remove the amount of attenuation, but instead creates a pseudo echo signal and subtracts this pseudo echo signal from the echo signal to cancel the echo signal. Therefore, unlike the echo suppressor 9, difficulty in speaking due to insertion and removal of the attenuation amount does not occur, but it has the following drawbacks.

It is generally known that the frequency spectrum of voice and music has a distribution in which the power density is high in the low range and low in the high range. Details are described in [``Sound'', published by the Japanese Society for Sound and Education, published by Coronasha Publishing, 1978]. echo canceller 1
7 indicates that no matter what frequency characteristics a signal is input to the reception signal input terminal R1n13, the transmission signal output terminal 5
Out 16 has the property that the frequency characteristics of the echo-cancelled signal are flat.

FIG. 7 is an oscilloscope waveform diagram showing the frequency characteristics of an echo canceller that receives white noise as input.

FIG. 8 is an oscilloscope waveform diagram showing the frequency characteristics of an echo canceller that receives Hoss noise as input.

White noise is input to the received signal input terminal Rin 13, and the white noise is input from the received signal output terminal Rout14 to the transmitted signal input terminal 5i.
The frequency characteristics at the transmission signal output terminal Sout16 are shown when the echo canceller 17 is disabled or enabled when a resistive attenuator (7 dB attenuation) is inserted up to n15.

In the oscilloscope waveform diagrams shown in FIGS. 7 and 8, one division of the vertical axis is 10 dB, one division of the horizontal axis is Q, 5kHz, the upper waveform is when disabled, and the lower waveform is is enabled.

In this way, even if the received input noise has different frequency characteristics, the frequency characteristics after echo cancellation are flattened, which has been observed in experiments.

Furthermore, the frequency characteristics (echo path characteristics) of the transfer function of the echo signal 3 that goes around from the speaker 1 to the microphone 2 are as follows:
As shown in FIG. 9, since the absorption rate of the walls in the room is lower than 1, it generally exhibits low-pass filter characteristics, but its frequency characteristics exhibit large changes in echo path loss with respect to minute changes in each frequency. This is a phenomenon that occurs because the sound emitted from the speaker 1 is reflected by the wall and inserted into the microphone 2.

Furthermore, it has been reported that the echo path loss changes greatly with small changes in each frequency, exhibiting wavy frequency characteristics. Therefore, if the gain of the amplifier in the echo path is increased by 10 dB so that the volume from the speaker 1 is optimized by the amount of echo cancellation obtained by the echo canceller 17, the O dB loss of the echo path shown in FIG. The reference point moves to the level shown in A in Figure 9, and even if the reverberating signal that goes around from speaker 1 to microphone 2 is a system with a power loss, a gain occurs at a certain frequency and causes howling. Sometimes. Therefore, if you try to increase the gain of an amplifier that is in the echo path and has a flat frequency response,
The gain can only be increased up to level point B shown in FIG. Furthermore, due to the characteristics of the echo canceller 17, the amount of echo cancellation cannot be made larger in the high range of the audio frequency spectrum than in the low range. As described above, the echo canceller 17 has the drawback that the amplifier cannot increase the gain by an amount equal to the absolute value of the amount of echo cancellation, and the effect of improving the insufficient sound volume is reduced.

Further, in the frequency shift method, although the wavy frequency characteristics of the echo path characteristics are corrected, the amount of smoothing that is corrected is about 6 to 10 at most, and cannot be said to be decisive. Therefore, implementing each method alone is not sufficient to prevent howling in a communication system that requires an amplifier gain higher than this value.

The present invention has been made by focusing on such conventional problems, and is aimed at preventing howling in a communication system caused by a reverberant signal generated by a received signal via an echo path being inserted into a transmitted signal. It is an object of the present invention to provide a howling suppression method and circuit that can increase the gain of an amplifier to enable comfortable voice communication.

[Means for solving problems]

The first aspect of the present invention is to suppress howling by generating a pseudo echo signal of the echo signal that goes around to the transmitting system from a signal arriving at the receiving system, and inverting the phase of this pseudo echo signal and adding it to the transmitting system. The method is characterized in that it also uses a method of shifting the frequency components of the signal passing through the signal path by a predetermined value.

The second aspect of the present invention is a circuit means for predicting and calculating a reverberation signal that goes around to the transmitting system using a signal arriving at the receiving system as an input, and adding the output of this circuit means to the signal of the transmitting system with an inverted phase. The howling suppression circuit includes a frequency shift circuit for shifting a frequency component of a signal passing through a signal path by a predetermined value.

Preferably, the signal path is a receiving system signal path and/or a transmitting system signal path.

The circuit means for predictive calculation is of an adaptive operation type, and the frequency shift circuit includes control means for controlling the operating state and non-operation state, and this control means generates a control signal in response to the adaptive operation of the circuit means for predictive calculation. It is preferable to have a configuration that sends out.

[For production]

The present invention provides an echo canceling means for generating a pseudo echo signal of an echo signal that goes around to a transmitting system from a signal arriving at a receiving system, inverting the phase of this pseudo echo signal, and adding it to the transmitting system; When a frequency shift circuit that shifts the frequency component of a signal passing through the path by a predetermined value is used in combination, and the frequency shift circuit is inserted into the signal path of the receiving system, the adaptive operation function of the echo canceling means starts. When the adaptive operation function is stopped, the frequency shift function is stopped, and when the adaptive operation function is stopped, the frequency shift function is started.If a frequency shift circuit is inserted into the signal path of the transmitter system, the frequency shift function is disabled except when transmitting a voice. By starting the function, the adaptive operation and frequency shift operation of the echo canceling means are appropriately controlled in each of the no-call state, the transmitting state, the receiving state, and the simultaneous call state, suppressing howling and suppressing the signal on the calling side. The frequency can be prevented from being shifted.

Note that the frequency shift function corrects the wave-like frequency characteristics of the echo path characteristics, thereby increasing the gain of the amplifier.

Therefore, the echo canceling and frequency shifting functions function in the loop for at least -1 or more times, and it is possible to solve the problem of insufficient volume of the speaker, which was caused by the howling suppression circuit using the echo canceling method.

〔Example〕

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

The howling suppression circuit of the present invention is characterized in that it uses echo canceling means and a frequency shift circuit together. In this embodiment, an echo canceller will be used as the echo canceling means.

1 to 3 are block diagrams showing first to third embodiments of the howling suppression circuit of the present invention. In FIGS. 1 to 3, the echo canceller 17
is a detection circuit 20 that inputs signals from a reception signal input terminal Rin 13 and a transmission signal input terminal Sin 15.
, control output of the detection circuit 20, received signal input terminal Rin
13 and the output of the subtraction circuit 19 as inputs,
Adaptive control filter 1 connecting the output signal to subtraction circuit 19
8. It is composed of a subtraction circuit 19 that adds the output signal of the adaptive control filter 18 to the signal of the transmission signal input terminal Sin 15 with an inverted phase and sends it to the transmission signal output terminal 5out16.

The circuit according to the first embodiment of the present invention shown in FIG. 1 is characterized in that a frequency shift circuit 21 is inserted into the receiving path between the received signal input terminal Rin 13 and the received signal output terminal Rout 14.

In FIG. 1, the frequency shift circuit 21 is inserted between the received signal input terminal R4n 13 and the echo canceller 17, and the received signal input terminal R4n input to the detection circuit 20.
The signal from Rin 13 is taken from the input side of the frequency shift circuit 21, and the signal from the received signal input terminal Rin 13, which is input to the adaptive control filter 18, is taken from the output side of the frequency shift circuit 21. The frequency shift circuit 21 includes a detection circuit 20
The control output from is connected.

Table 1 is a table showing operating conditions of the detection circuit 20 of the first embodiment.

Below, the operation of the circuit of the first embodiment will be explained in Fig. 1 and Fig. 1.
Explain with reference to the table.

First, the detection circuit 20 detects the reception level LRin and the transmission level LSin of the signals input to the reception signal input terminal Rin 13 and the transmission signal input terminal Sin 15, respectively.
It is determined whether or not the power exceeds a set level (identification level) Lth that can be recognized as voice power. Transmission level L
Both Sin and reception level LRin are at identification level Lt.
h), or the transmission level LSin exceeds the identification level Lth and the reception level L
When Rin does not exceed the discrimination level Lth (sending state), the operation of the adaptive control filter 18 is stopped and the operation of the frequency shift circuit 21 is started. Reception level LRi
When n exceeds the identification level Lth and the transmission level LSin does not exceed the identification level Lth (receiving state), the operation of the adaptive control filter 18 is started, and the frequency shift circuit 2
Stop operation 1.

Furthermore, the detection circuit 20 detects the gain of the amplifier and the minimum loss E of the reverberation characteristics that circulate from the speaker to the microphone.
RL is set in advance, and the difference from the minimum loss amount ERL via the echo path from the received signal output terminal Rout I4 to the transmitted signal input terminal Stn 15, that is, the received signal input terminal R
Take the difference between the reception level LRin at in 13 and the transmission level L Sin at transmission signal input terminal Sin 15,
This amount of difference is compared with the minimum loss amount ERL to determine whether it is an adaptive operation state or a simultaneous call state. When both the transmission level LSin and the reception level LRin exceed the identification level Lth, the reception level LR4n and the transmission level LSin
If the difference is larger than the minimum loss amount ERL, the receiving state is established, and the operation of the adaptive control filter 18 is started and the operation of the frequency shift circuit 21 is stopped. If the difference between the reception level LRin and the transmission level LSin is smaller than the minimum loss amount ERL, it is a simultaneous call state, the operation of the adaptive control filter 18 is stopped, and the operation of the frequency shift circuit 21 is started.

In this way, the detection circuit 20 sends control operation commands to the adaptive control filter 18 and the frequency shift circuit 21.

Currently, a control method called a learning identification method is adopted as an embodiment of the adaptive control filter 18, and the received human power signal ER4n from the received signal input terminal Rin 13, which is input to the adaptive control filter 18, and the input of the subtraction circuit 19 are A signal, that is, a transmission input signal ESin from the transmission signal input terminal Sin 15, and an output signal of the N arithmetic circuit 19, that is, a transmission output signal E sent to the transmission signal output terminal 5out15.
Using the three signals Sout and Sout, in the operation start state of the adaptive control filter 18 shown in Table 1, the transmission output signal E
The power of Sout is automatically controlled to always be 0 to eliminate echo signals. The stopped state of the adaptive control filter 18 means that the adaptive operation is stopped, a pseudo echo signal is sent to the subtraction circuit 19, and the echo signal cancellation operation is performed.

Furthermore, the frequency shift circuit 21 has a memory approximately equal to the audio pinch time interval, samples the input signal at a constant time interval, stores the sample value in the memory,
Next, the accumulated sample values are read out at intervals longer than the sampling interval at which they were input, generating an output signal whose frequency is isometrically shifted to the lower frequency side. Here, the stopping operation of the frequency shift circuit 21 means stopping the frequency shift function, and is equivalent to allowing the input signal to pass through as is.

Generally, in the audio system of a video conference telephone device, the devices shown in FIG. is stopped and the frequency shift circuit 21 starts operating, but the device on the opposite side operates in the opposite manner, and no frequency distortion occurs on the calling side. Also,
The howling margin in the four-wire loop system via the communication path is the sum of the amount of echo cancellation and the amount of smoothing due to frequency shift, making it possible to further increase the gain of the amplifier.

The circuit according to the second embodiment of the present invention shown in FIG. 2 is characterized in that a frequency shift circuit 22 is inserted in the transmission path between the transmission signal input terminal Sin 15 and the transmission signal output terminal 5out16.

In FIG. 2, the frequency shift circuit 22 is inserted between the echo canceller 17 and the transmission signal output terminal 5out16, and the control output from the detection circuit 20 is connected to the frequency shift circuit 22.

Table 2 is a table showing operating conditions of the detection circuit 20 of the second embodiment.

Below, the operation of the second embodiment circuit will be explained in Fig. 2 and 2.
Explain with reference to the table.

As in the first embodiment, the detection circuit 20 first includes a reception signal input terminal Rin 13 and a transmission signal input terminal 5in15.
It is determined whether the receiving level L Rin and the transmitting level LSin of the signals inputted to each exceed a set level (discrimination level) Lth that can be recognized as voice power. Transmission level LSin and reception level LR
When both in do not exceed the identification level Lth (no call state B), the operation of the adaptive control filter 18 is stopped and the operation of the frequency shift circuit 22 is started. Transmission level LS
in exceeds the identification level Lth, and the reception level L Rin
does not exceed the discrimination level Lth (speech state), the operation of the adaptive control filter 18 and the frequency shift circuit 22 are stopped. When the reception level LRin exceeds the identification level Lth and the transmission level L Sin does not exceed the identification level Lth (receiving state), the operation of the adaptive control filter 18 and the frequency shift circuit 22 are started.

Furthermore, the detection circuit 20 detects the gain of the amplifier and the minimum loss 1ER of the reverberation characteristics from the speaker to the microphone.
L is set in advance, and the difference from the minimum loss amount ERL via the echo path from the received signal output terminal Rou t14 to the transmitted signal input terminal Sin 15, that is, the received signal input terminal R
Take the difference between the reception level L Rin at in 13 and the transmission level LSin at transmission signal input terminal Sin 15,
This amount of difference is compared with the minimum loss amount ERL to determine whether it is an adaptive operation state or a simultaneous call state. When both the transmission level LSin and the reception level LRin exceed the identification level Lth, the reception level LR4n and the transmission level LSi
If the difference with n is larger than the minimum loss amount ERL, the receiving state is established, and the operation of the adaptive control filter 18 and the frequency shift circuit 22 are started.

If the difference between the reception level L Rin and the transmission level L Sin is smaller than the minimum loss amount ERL, it is a simultaneous call state,
The operation of the adaptive control filter 18 is stopped and the operation of the frequency shift circuit 22 is started.

In this way, the detection circuit 20 sends commands for control operations to the adaptive control filter 18 and the frequency shift circuit 22.

In general, in the audio system of a video conference telephone device, the devices shown in FIG. is stopped, the operation of the frequency shift circuit 22 is also stopped, and even if the device on the opposite side performs the opposite operation, no frequency distortion will occur on the calling side. In addition, the howling margin in a four-wire loop system via a communication path is the sum of the amount of echo cancellation and the amount of smoothing due to frequency shift.
It becomes possible to further increase the gain of the amplifier.

The circuit according to the third embodiment of the present invention shown in FIG.
13 and the reception signal output terminal Rout14, and a transmission path between the transmission signal input terminal Sin 15 and the transmission signal output terminal 5out16.

In FIG. 3, the first frequency shift circuit 21 is inserted between the received signal input terminal Rin 13 and the echo canceller 17, and the second frequency shift circuit 22 is inserted between the echo canceller 17 and the transmitted signal output terminal 5out16. received signal input terminal Rin which is input to the detection circuit 20.
The signal from Rin 13 is taken from the input side of the first frequency shift circuit 21, and the signal from the received signal input terminal Rin 13, which is input to the adaptive control filter 18, is taken from the output side of the first frequency shift circuit 21. A control output from the detection circuit 20 is connected to the first and second frequency shift circuits 21,22.

Table 3 is a table showing operating conditions of the detection circuit 20 of the third embodiment.

The operation of the circuit of the third embodiment will be explained below with reference to FIGS. 3 and 3.
Explain with reference to the table.

As in the first embodiment and the second embodiment, first, the detection circuit 2
0 indicates whether the reception level LR4n and the transmission level L Sin of the signals input to the reception signal input terminal R4n 13 and the transmission signal input terminal Sin 15, respectively, exceed the set level (identification level) LLh that can be recognized as audio power. identify. When both the transmission level L Sin and the reception level LRin do not exceed the identification level Lth (no call state), the adaptive control filter 18
stops the operation of the first and second frequency shift circuits 2.
1. Start the operation of 22.

When the transmission level L Sin exceeds the identification level Lth and the reception level LRin does not exceed the identification level Lth (sending state), the operation of the adaptive control filter 18 is stopped, and the operation of the first frequency shift circuit 21 is stopped. , and the operation of the second frequency shift circuit 22 is stopped. The reception level LRin exceeds the identification level Lth and the transmission level LSi
When n does not exceed the identification level Lth (receiving state), the operation of the adaptive control filter 18 is started, the operation of the first frequency shift circuit 21 is stopped, and the operation of the second frequency shift circuit 22 is started. do.

Further, the detection circuit 20 is in a listening state if the difference between the reception level L Rin and the transmission level LSin is greater than the minimum loss 11ERL when both the transmission level LSin and the reception level LR4n exceed the identification level Lth,
The operation of the adaptive control filter 18 is started, the operation of the first frequency shift circuit 21 is stopped, and the operation of the second frequency shift circuit 22 is started. If the difference between the reception level L Rin and the transmission level LSin is smaller than the minimum loss amount URL, it is a simultaneous call state, the operation of the adaptive control filter 18 is stopped, and the operation of the first and second frequency shift circuits 21 and 22 is stopped. Start operation.

In this way, the detection circuit 20 sends control operation commands to the adaptive control filter 18, the first frequency shift circuit 21, and the second frequency shift circuit 22.

Generally, in the audio system of a video conference telephone device, the devices shown in FIG. 3 are arranged opposite to each other via a communication path, and when one of them is making a call (during a one-way call), the adaptive control filter I8 operates. is stopped, and the operation of the frequency shift circuits 21, 22 is started or stopped, but the device on the opposite side operates in the opposite manner, and no frequency distortion occurs on the calling side. Furthermore, the howling margin in the four-wire loop system via the communication path is the sum of the amount of echo cancellation and the amount of smoothing due to frequency shift, making it possible to further increase the gain of the amplifier.

In the above specific example, the explanation was given using a system with a speaker and a microphone for four-wire audio of a video conference telephone device, but it can also be applied to cases where the cause of howling is not necessarily an acoustic system but reflections. It is also applicable to a four-wire section system of a telephone network that connects one or more ground stations and has a two-wire/four-wire conversion circuit.

(Hereinafter, the margin of this page) [Effects of the Invention] As explained above, the present invention provides a system in which, during one-way communication, a frequency shift is not inserted into the signal of the calling side, but a frequency shift is inserted into the signal of the non-calling side. Therefore, there is no deterioration in call quality, and when considered in a four-wire loop system, at least one echo cancellation and frequency shift function function in the loop, and therefore the amount of echo cancellation and the amount of smoothing due to frequency shift are Since the sum of is the howling margin,
This has the effect of increasing the gain of the amplifier. That is, there is an effect that the insufficient volume of the speaker, which was caused by the conventional howling suppression circuit based on the echo cancellation method, can be solved, and the optimal volume can be obtained.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a first embodiment of the present invention. FIG. 2 is a block diagram showing a second embodiment of the present invention. FIG. 3 is a block diagram showing a third embodiment of the present invention. FIG. 4 is a block diagram showing an example of the configuration of the audio system of a conventional video conference telephone device. FIG. 5 is a block diagram of a howling suppression circuit using an echo subreflexor. FIG. 6 is a block diagram of a howling suppression circuit using an echo canceller. Figure 7 is an oscilloscope waveform diagram showing the frequency characteristics of an echo canceller using white noise as input. Figure 8 is an oscilloscope waveform diagram observing the frequency characteristics of an echo canceller using Hoss noise as input. FIG. 9 is a diagram showing an example of indoor reverberation characteristics. 1...Speaker, 2...Microphone, 3...
Echo signal, 4...Speaker's audio signal, 5.6.7.8
... Amplifier or attenuator, 9... Echo suppressor, 10.11... Variable attenuation circuit, 12... Control circuit, 13... Received signal input terminal Rin, 14...
Reception signal output terminal Rout, 15... Transmission signal input terminal Sin, 16... Transmission signal output terminal Sout
, 17...Echo canceller, 1st...Adaptive control filter, 19...Subtraction circuit, 20...Detection circuit, 2
1.22...Frequency shift circuit.

Claims (6)

[Claims] (1) In a howling suppression method in which a pseudo-reverberation signal is generated from a signal arriving at the receiving system to a reverberation signal that goes around to the transmitting system, and the phase of this pseudo-echo signal is inverted and added to the transmitting system, the signal path A method for suppressing howling, characterized in that it also uses a method of shifting a frequency component of a signal passing through by a predetermined value. (2) Howling comprising circuit means for predicting and calculating the reverberation signal that goes around to the transmitting system using the signal arriving at the receiving system as input, and means for adding the output of this circuit means to the signal of the transmitting system with an inverted phase. A howling suppression circuit, characterized in that the suppression circuit includes a frequency shift circuit that shifts a frequency component of a signal passing through a signal path by a predetermined value. (3) The howling suppression circuit according to claim (2), wherein the signal path is a receiving system signal path. (4) The howling suppression circuit according to claim (2), wherein the signal path is a transmission system signal path. (5) The howling suppression circuit according to claim (2), wherein the signal path is a receiving system signal path and a transmitting system signal path. (6) The circuit means for performing predictive calculation is of an adaptive operation type, and the frequency shift circuit includes control means for controlling the operating state and the non-operating state, and this control means corresponds to the adaptive operation of the circuit means for performing predictive calculation. The howling suppression circuit according to claim 2, wherein the howling suppression circuit is configured to send out a control signal.