JP2001094479A - Echo canceler - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an echo canceler that can stably cancel an echo signal even when there is an excess input by which a reproduction amplifier or a reproduction device causes nonlinear distortion. SOLUTION: In the echo canceler that cancels an echo signal by generating a simulated echo path from a transmission signal led to an echo path and the echo signal resulting from the transmission signal after passing through the echo path and by subtracting a simulated echo signal obtained by regarding the transmission signal as an input signal led to the simulated echo path from the echo signal, when the level of the transmission signal led to the echo path exceeds a threshold set so that no linear distortion is caused in the reproduction amplifier 4 or the reproduction device 5 due to an input signal with an excess level, a clip processing section 10 to decrease the level of the transmission signal led to the echo path and the input signal led to the simulated echo path below the threshold is provided to a pre-stage of the echo canceler 2.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a reverberation canceling apparatus for a voice communication conference, which deletes an indoor reverberation signal which causes howling and impairs hearing in a communication conference system.

[0002]

2. Description of the Related Art An acoustic echo canceller is an indispensable technology for a hands-free conference in a voice communication conference or the like. At this time, the amplifier for the regenerator or the regenerator can reproduce linearly up to a signal of a certain magnitude, but when an excessive signal is input, nonlinear distortion occurs.

FIG. 2 is an explanatory view showing the configuration of a conventional electro-acoustic facility for a voice communication conference. In FIG. 2, 1 is a clip processing unit that performs clipping, 2 is an echo canceller, 3 is a digital / analog (D / A) converter, 4 is an amplifier for a regenerator,
5 is a reproducing device such as a speaker, 6 is a sound collecting device such as a microphone, 7 is a sound collecting amplifier, 8 is an analog / digital (A
/ D) Converters, T1 and T2 are terminals.

That is, a receiving signal of a digital signal from the other party is transmitted to the terminal T 1, and the receiving signal is converted into an analog signal by the digital / analog converter 3 via the echo canceler 2. The received signal from the digital / analog converter 3 is passed through the amplifier 4 for the reproducing device and the clip processing section 1 and then reproduced from the reproducing device 5 to sound. On the other hand, the transmission voice is picked up by the sound pickup 6, converted into a digital transmission signal by the analog / digital converter 8 via the sound pickup amplifier 7, and transmitted from the terminal T 2 via the echo canceler 2 to the other party. Sent. The echo canceller 2 generates a pseudo echo path from a transmission signal to the echo path and an echo signal of the transmission signal after passing through the echo path, and obtains the transmission signal as an input to the pseudo echo path. The reverberation signal is subtracted from the reverberation signal to eliminate the reverberation signal and eliminate the residual signal. Even in conventional electro-acoustic equipment,
In some cases, a clip processing unit 1 for preventing excessive input has been introduced. However, since the purpose is to prevent an excessive input to the reproducing device 5, the clip processing section 1 is arranged in front of the reproducing device 5 as shown in FIG. In many cases, the clip processing unit 1 is built in the amplifier. However, when the clip processing unit 1 is arranged as described above, when an excessive level signal is input, the input signal to the pseudo echo path and the input signal to the echo path are different from each other, and the signal to the pseudo echo path is changed. Therefore, it is not possible to estimate the input signal to the echo path from the input signal of (1), it is not possible to generate a correct pseudo echo signal, and echo cancellation by the echo canceller cannot be performed.

[0005]

In the conventional echo canceller, since the echo path including the amplifier for the regenerator or the regenerator is estimated as a linear function, it cannot be estimated if nonlinear distortion occurs in the middle. There is a problem that a correct pseudo echo signal cannot be generated, and echo / howling cannot be prevented.

The present invention has been made in view of the above circumstances, and provides an echo canceller capable of stably canceling an echo signal even when an amplifier for a regenerator or a regenerator has an excessive input that causes nonlinear distortion. It is intended to do so.

[0007]

According to the present invention, a pseudo echo path is generated from a transmission signal to an echo path and an echo signal of the transmission signal after passing through the echo path.
An echo canceller that eliminates the echo signal by subtracting a pseudo echo signal obtained by inputting the transmission signal to the pseudo echo path from the echo signal,
When the transmission signal to the reverberation path exceeds a threshold set so as not to cause nonlinear distortion due to an excessive level input signal to the regenerator amplifier or regenerator, the transmission signal to the reverberation path and the pseudo reverberation path It is characterized in that a clip processing unit for lowering the input signal below the threshold value is provided at the preceding stage of the echo canceller.

Further, the present invention is characterized in that the echo canceller is provided with a setting section for making the threshold of the clip processing section variable.

Further, according to the present invention, in the above echo canceller, the clip processing unit calculates a power of a transmission signal to the echo path, and a transmission signal power and a threshold calculated by the power calculation unit. And a multiplier for multiplying the transmission coefficient to the echo path by the multiplication coefficient calculated by the multiplication coefficient calculation section.

Further, the present invention is characterized in that, in the echo canceller, the threshold set by the threshold setting unit is used as the threshold used by the multiplication coefficient calculation unit.

[0011]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a structural explanatory view showing an embodiment of the present invention. 1, the same parts as those in FIG. 2 are denoted by the same reference numerals, and the description thereof will be omitted. In FIG. 1, reference numeral 10 denotes a clip processing unit that performs clipping.

That is, a pseudo echo path is generated from a transmission signal to the echo path and an echo signal of the transmission signal after passing through the echo path, and the transmission signal is obtained as an input to the pseudo echo path. In the echo canceller 2 for canceling the echo signal by subtracting the pseudo echo signal from the echo signal, the signal sent to the echo path may have a non-linear distortion due to an excessive level input signal to the amplifier 4 for the regenerator or the regenerator 5. Clip processing unit 10 for lowering the transmission signal to the reverberation path and the input signal to the simulated reverberation path to the threshold value or less when the threshold value exceeds a threshold value set so as not to cause the noise.
Is provided before the signal from the other party is input to the echo canceller 2.

Here, the clip processing of the clip processing unit 10 is determined in advance so that the absolute value of the input amplitude or the input power does not cause nonlinear distortion due to excessive input to the amplifier 4 for the regenerator or the regenerator 5. When the signal exceeds the threshold, the signal is set to a signal smaller than the threshold.

By arranging the clip processing section 10 in the preceding stage of the echo canceler 2 as described above, the input signal to the pseudo echo path and the input signal to the echo path can be equalized.
Since the correct pseudo echo signal can be generated, the above-described problem does not occur.

FIG. 3 is a structural explanatory view showing an example of the clip processing unit 10 of FIG. In FIG. 3, reference numeral 11 denotes a power calculator for calculating the power of the received signal, 12 denotes a multiplication coefficient calculator for calculating a multiplication coefficient, and 13 denotes a multiplier.
3, parts corresponding to those in FIG. 1 are denoted by the same reference numerals. Power calculation unit 1 calculates the power of the reception signal from the other party
1, a multiplication coefficient G is calculated based on the received signal power and a predetermined threshold, and the multiplied coefficient G is multiplied by the multiplier 13 to the received signal,
Even if an excessively high level of the received signal is input, the signal does not become nonlinear in the regenerator amplifier 4 or the regenerator 5 at the subsequent stage.

FIG. 4 shows a case where a reception signal exceeding the limit reproduction level of the reproducer is input when clip processing is performed by the clip processing unit in FIG. 1 (with pre-clip) and when clip processing is not performed (without pre-clip). 5 is a characteristic diagram showing the power of a residual signal after echo cancellation. When the maximum volume of the reproducing device 5 is set to 0 dB, the preset threshold value of the clip processing unit 10 is set to about -3 dB. The reverberation canceller 2 having the clip processing unit 10 is about 10 dB lower than the reverberation canceller not having the clip processing unit.
A large amount of erasure is obtained, which indicates the effectiveness of the present invention.

FIG. 5 is a structural explanatory view showing another embodiment of the present invention. 5, the same parts as those in FIG. 1 are denoted by the same reference numerals, and the description thereof will be omitted. In FIG. 5, reference numeral 9 denotes a threshold setting unit for determining a threshold. That is, the threshold value setting unit 9 changes the threshold value of the clip processing unit 10 set so as not to cause non-linear distortion due to an excessive level input signal to the regenerator amplifier 4 or the regenerator 5. When the types of transmission signals are different, it is possible to cope with the difference in the threshold value.

FIG. 6 is an explanatory diagram showing an example of the clip processing section and threshold setting section shown in FIG. 6, the same parts as those in FIG. 5 are denoted by the same reference numerals, and description thereof will be omitted. The power of the reception signal from the other party is calculated by the power calculation unit 11, and a threshold value Th set so as not to cause nonlinear distortion due to an excessive level of input signal to the regenerator amplifier 4 or the regenerator 5 is input from the threshold value setting unit 9. Then, a multiplication coefficient G is calculated based on the reception signal power and the threshold value, and the multiplication coefficient G is multiplied by the multiplier 13 with the reception signal, whereby
Even if an excessively high level signal is input, the signal does not become nonlinear in the regenerator amplifier 4 or the regenerator 5 at the subsequent stage.

It is generally said that a music signal has a larger peak factor than an audio signal. If the type of signal is known in advance, the threshold is adjusted with a dial volume or the like. The threshold value can be automatically set by analyzing the frequency components, time characteristics, and the like of each signal in the clip processing unit 10.

[0021]

As described above, according to the present invention, the reverberation signal can be stably eliminated even if the amplifier for the regenerator or the regenerator has an excessive input enough to cause nonlinear distortion.

[Brief description of the drawings]

FIG. 1 is a configuration explanatory diagram showing an embodiment of the present invention.

FIG. 2 is a configuration explanatory view showing a conventional electro-acoustic facility for a voice communication conference.

FIG. 3 is a configuration explanatory diagram illustrating an example of a clip processing unit in FIG. 1;

FIG. 4 is a diagram showing a case where a reception signal exceeding a limit reproduction level of a reproducing device is input when clip processing is performed (with pre-clip) and when clip processing is not performed (without pre-clip) in the clip processing unit in FIG. FIG. 9 is a characteristic diagram illustrating power of a residual signal after echo cancellation.

FIG. 5 is a configuration explanatory view showing another embodiment of the present invention.

FIG. 6 is a configuration explanatory diagram showing an example of a clip processing unit and a threshold setting unit of FIG. 5;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Clip processing part which performs clip 2 Echo canceller 3 Digital / analog (D / A) converter 4 Amplifier for regenerator 5 Regenerator such as speaker 6 Sound pickup device such as microphone 7 Amplifier for sound pickup 8 Analog / Digital (A / D) converter 9 threshold setting unit 10 clip processing unit that performs clipping T1, T2 terminals

Continued on the front page (72) Inventor Masafumi Tanaka 2-3-1 Otemachi, Chiyoda-ku, Tokyo Japan Telegraph and Telephone Corporation (72) Inventor Shigeaki Aoki 2-3-1 Otemachi, Chiyoda-ku, Tokyo Date Within the Telegraph and Telephone Corporation (72) Inventor Shoji Makino 2-3-1 Otemachi, Chiyoda-ku, Tokyo F-term within the Telegraph and Telephone Corporation (reference) 5K027 AA07 BB03 DD10 KK04 LL03 5K046 AA01 BB01 HH13 HH18 HH29 HH79

Claims (4)

[Claims] 1. A pseudo echo path is generated from a transmission signal to an echo path and an echo signal of the transmission signal after passing through the echo path, and is obtained by inputting the transmission signal to the pseudo echo path. An echo canceller for canceling the echo signal by subtracting a pseudo echo signal from the echo signal, wherein the output signal to the echo path does not cause nonlinear distortion due to an excessive level input signal to the amplifier for the regenerator or the regenerator. And a clip processing unit for lowering the transmission signal to the reverberation path and the input signal to the pseudo reverberation path below the threshold value when the threshold value set in the above manner is exceeded. apparatus. 2. The echo canceller according to claim 1, wherein:
A reverberation canceling device, comprising: a setting unit that makes the threshold value of the clip processing unit variable. 3. The echo canceller according to claim 1, wherein:
A clip processing unit for calculating a power of the transmission signal to the echo path, a power calculation unit, a multiplication coefficient calculation unit for calculating a multiplication coefficient based on the transmission signal power and the threshold calculated by the power calculation unit, A reverberation canceling device comprising a multiplier for multiplying a multiplication coefficient calculated by a multiplication coefficient calculation unit with a signal transmitted to an echo path. 4. The echo canceller according to claim 3, wherein:
An echo canceller characterized by using a threshold set by a threshold setting unit as a threshold used by a multiplication coefficient calculation unit.