JP2010283483A - Echo canceller, microphone, and method of supporting arrangement of speaker - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To support users so that they dispose speakers and microphones at positions that can sufficiently suppress echoes and howling. <P>SOLUTION: A pseudo-echo generation section 107 estimates an acoustic echo of a far-end sound signal input from a conference terminal via a conference terminal IF section 102 and an FFT section 105 for outputting as a pseudo-echo signal. An echo canceller section 108 subtracts the pseudo-echo signal generated by the pseudo-echo generation section 107 from a near-end sound signal input from the microphone via a microphone connection section 104 and an FFT section 106 to suppress the acoustic echo included in the near-end sound signal. At this time, an arrangement support section 111 determines whether the arrangement of the speaker or microphone is appropriate based on the sound pressure level of the pseudo-echo signal to display the determined result on a display section 1011 of an MMIF section 101. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an echo canceller, and more particularly to a technique for supporting the arrangement of microphones and speakers connected to the echo canceller.

  Patent Document 1 discloses an echo canceller for a communication system using an adaptive filter. This echo canceller generates a pseudo echo signal from an audio signal decoded by a codec and output from a speaker, and subtracts the pseudo echo signal from an audio signal input to a microphone. Thereby, the echo contained in the audio | voice signal input into the microphone is suppressed.

  Patent Document 2 discloses a technique for suppressing howling in a conference system in which microphones and speakers are arranged in two rooms that transmit and receive audio signals via a communication line. In this technique, an echo canceller and a time-varying notch filter are arranged in each room. The echo canceller estimates a sneak component from a speaker arranged in the same room to the microphone, and removes this sneak component from the input signal of the microphone to generate an error signal. The time-variant notch filter monitors the unstable pole of the spectrum of the error signal, and when the unstable pole increases, performs a filtering process to suppress the frequency component.

Japanese National Patent Publication No. 11-513230 JP 2000-92588 A

  Conventionally, a conference system that can use an external speaker and microphone is known. This type of conference system is easy to use because speakers and microphones can be placed in free positions.

  However, on the other hand, a speaker and a microphone may be placed at a position where echo and howling cannot be sufficiently suppressed. In addition, the arrangement of speakers or microphones may be changed during the conference, and echoes and howling that could be suppressed until then may occur. The techniques described in Patent Documents 1 and 2 do not consider these points.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a technique capable of assisting a user to arrange a speaker and a microphone at a position where echo and howling can be sufficiently suppressed.

  In order to solve the above-mentioned problem, the present invention estimates the acoustic echo of the output audio signal of the speaker, and subtracts this estimated acoustic echo from the input audio signal of the microphone, and is included in the input audio signal of the microphone. Suppress acoustic echo. At this time, based on the estimated sound pressure level of the acoustic echo, it is determined whether or not the speaker or microphone is properly arranged, and the determination result is notified to the user.

For example, the present invention is an echo canceller that eliminates an acoustic echo of an output audio signal of a speaker from an input audio signal of a microphone,
A pseudo echo generating means for estimating an acoustic echo of the output audio signal and generating a pseudo echo signal;
Echo canceling means for subtracting the pseudo echo signal from the input voice signal to suppress acoustic echo contained in the input voice signal;
And arrangement support means for determining whether or not the arrangement of the speaker or the microphone is appropriate based on the sound pressure level of the pseudo echo signal and notifying the result.

  Here, a predetermined sample audio signal is output from the speaker and input to the microphone, and the pseudo echo generation unit estimates the acoustic echo of the sample audio signal to generate a pseudo echo signal. Comparing means for comparing the signal and the sampled audio signal input to the microphone may be further provided. Then, the arrangement support means may determine whether the arrangement of the speaker or the microphone is appropriate based on the comparison result of the comparison means, and notify the result.

  ADVANTAGE OF THE INVENTION According to this invention, a user can be assisted to arrange | position a speaker and a microphone in the position which can suppress echo and howling fully.

FIG. 1 is a schematic configuration diagram of an echo canceller according to an embodiment of the present invention. 2A is a flowchart for explaining the echo monitoring process of the echo canceller shown in FIG. 1, and FIG. 2B is a flowchart for explaining the adaptation process of the echo canceller shown in FIG. It is.

  An embodiment of the present invention will be described by taking an echo canceller of a conference terminal used in a conference system as an example. The conference terminal has a function of transmitting and receiving audio signals to and from other conference terminals via a network such as an IP network or PSTN.

  FIG. 1 is a schematic configuration diagram of an echo canceller according to an embodiment of the present invention.

  As illustrated, the echo canceller 1 includes an MMIF unit 101, a conference terminal IF unit 102, a speaker connection unit 103, a microphone connection unit 104, FFT units 105 and 106, a pseudo echo generation unit 107, and an echo canceller. Unit 108, IFFT unit 109, adaptation processing unit 110, and placement support unit 111.

  The MMIF unit 101 is a man-machine interface including a display unit 1011 such as an LCD and an LED, and an operation unit 1012 such as a switch and a button.

  The conference terminal IF unit 102 is an interface for transmitting and receiving audio signals to and from the conference terminal. The conference terminal IF unit 102 receives the voice signal of the other party from the conference terminal (referred to as a far-end voice signal), passes it to the speaker connection unit 103 and the FFT unit 105, and receives the user voice signal received from the IFFT unit 109. (Referred to as near-end audio signal) is transmitted to the conference terminal.

  The speaker connection unit 103 is an interface for connecting an external speaker. The speaker connection unit 103 outputs the far-end audio signal received from the conference terminal IF unit 102 from the speaker connected to itself.

  The microphone connection unit 104 is an interface for connecting an external microphone. The microphone connection unit 104 passes the near-end audio signal collected by the microphone connected to the microphone connection unit 104 to the FFT unit 106.

  The FFT unit 105 performs FFT (Fast Fourier Transform) on the far-end audio signal received from the conference terminal IF unit 102 and outputs the frequency domain information of the far-end audio signal.

  The FFT unit 106 performs FFT on the near-end audio signal received from the microphone connection unit 104 and outputs frequency domain information of the near-end audio signal.

  The pseudo echo generation unit 107 estimates the acoustic echo component of the far end speech signal using an adaptive filter based on the frequency region information of the far end speech signal output from the FFT unit 105, and the frequency domain information of the pseudo echo signal. Is generated.

  An adaptive filter is a change that occurs when a reference signal (here, a far-end audio signal) is input to this filter and that reference signal is output from the speaker and transmitted as a sound wave to reach the microphone (possible (As long as it is) a filter created to output the result (signal) of applying an equivalent change to its reference signal. Such an adaptive filter is created as a result of acoustic environment learning performed in accordance with the call situation. Various existing techniques including the techniques described in Patent Documents 1 and 2 can be used for the pseudo echo generation unit 107 and the adaptive filter.

  The echo canceller 108 is included in the near-end audio signal by subtracting the frequency domain information of the pseudo echo signal output from the pseudo echo generator 107 from the frequency domain information of the near-end audio signal output from the FFT unit 106. Suppresses the acoustic echo that is present.

  The IFFT unit 109 performs IFFT (Inverse FFT) on the frequency domain information of the near-end audio signal output from the echo canceller unit 108 and outputs a near-end audio signal.

  The adaptation processing unit 110 outputs a predetermined sample audio signal from a speaker connected to the speaker connection unit 103 and collects sound on a microphone connected to the microphone connection unit 104 in accordance with an instruction from the arrangement support unit 111. At the same time, this sample audio signal is input to the FFT unit 105 and the pseudo echo generation unit 107 generates a pseudo echo signal. Then, the pseudo echo signal is compared with the sample audio signal output from the speaker and collected by the microphone, and the difference (phase difference, sound pressure level difference, etc.) between them is notified to the placement support unit 111.

  The placement support unit 111 performs the echo monitoring process illustrated in FIG. 2A when the conference terminal IF unit 102 transmits and receives audio signals to and from the conference terminal, that is, when the conference terminal is communicating with the other party. To do.

  That is, the arrangement support unit 111 sequentially acquires the pseudo echo signal generated by the pseudo echo generation unit 107 (S101), and measures the sound pressure level of the pseudo echo signal (S102). Then, it is checked whether or not the sound pressure level is lower than a predetermined reference level T1 (S103).

  If the sound pressure level of the pseudo echo signal is less than the reference level T1 (YES in S103), the microphone and speaker settings (setting of arrangement, volume, etc.) are appropriate, that is, the microphone and It is determined that the speaker is set, and a display “OK display” to that effect is displayed on the display unit 1011 of the MMIF unit 101 (S104).

  On the other hand, when the sound pressure level of the pseudo echo signal is equal to or higher than the reference level T1 (NO in S103), the microphone and the speaker are improperly set, that is, the microphone and the speaker are set so as not to generate echo and howling. It is determined that there is not, and a display “NG display” to that effect is displayed on the display unit 1011 of the MMIF unit 101 (S105).

  Here, the predetermined reference level T1 may be appropriately set by the user in consideration of the performance of the echo canceller 1 or the like so that “NG display” is obtained when the echo and howling cannot be sufficiently suppressed. When a situation occurs in which echo and howling cannot be sufficiently suppressed during use of the echo canceller 1, the sound pressure level of the pseudo echo signal at that time is stored, and these values are presented as reference values for the reference level T1. May be.

  In addition, the arrangement support unit 111 receives an adaptation instruction from the user when the conference terminal IF unit 102 does not transmit / receive audio signals to / from the conference terminal, that is, when the conference terminal is before starting communication with the communication partner. When accepted via the operation unit 1012, the adaptation processing unit 110 is activated. As a result, the following adaptation process shown in FIG.

  The activated adaptation processing unit 110 outputs a predetermined sample audio signal (S121). The sample audio signal is output from the speaker via the speaker connection unit 103 and then collected by the microphone. And it inputs into the FFT part 106 via the microphone connection part 104, and is converted into frequency domain information. The sample audio signal is also input to the FFT unit 105, converted into frequency domain information, and then input to the pseudo echo generation unit 107. Then, the pseudo echo generation unit 107 generates and outputs pseudo echo frequency domain information from the frequency domain information of the input sample sound.

  Next, the adaptation processing unit 110 acquires the frequency domain information of the sample audio signal collected by the microphone output from the FFT unit 106 and generates the sample audio signal output from the pseudo echo generation unit 107. The frequency domain information of the pseudo echo signal thus obtained is acquired (S122). Then, the difference (phase difference, sound pressure level difference, etc.) between the frequency domain information of the sample audio signal collected by the microphone and the frequency domain information of the pseudo echo signal generated from the sample audio signal is measured. Is notified to the placement support unit 111 (S123).

  In response to this, the arrangement support unit 111 checks whether or not the difference between the two received from the adaptation processing unit 110 is within the predetermined allowable range T2 (S124).

  If the difference between the two is within the allowable range T2 (YES in S124), the microphone and speaker are set properly (setting, arrangement, volume, etc.), that is, the microphone and the speaker are set so that no echo or howling occurs. The display “OK display” to that effect is displayed on the display unit 1011 of the MMIF unit 101 (S125).

  On the other hand, if the difference between the two is outside the allowable range T2 (NO in S124), it is determined that the setting of the microphone and the speaker is inappropriate, that is, the microphone and the speaker are not set so that no echo or howling occurs. Then, a display “NG display” to that effect is displayed on the display unit 1011 of the MMIF unit 101 (S126). Here, when the placement support unit 111 receives a re-adaptation instruction from the user via the operation unit 1012 of the MMIF unit 101 (NO in S127), the adaptation processing unit 110 is activated again to perform the processing after S121. To do.

  Here, the predetermined allowable range T2 may be appropriately set by the user in consideration of the performance of the echo canceller 1 or the like so that “NG display” is obtained when the echo and howling cannot be sufficiently suppressed. Differences when echo and howling can be sufficiently suppressed as a result of the adaptation process may be stored and presented as reference values for determining the allowable range T2.

  The embodiment of the present invention has been described above.

  In the echo canceller 1 of the present embodiment, the pseudo echo generation unit 107 estimates the acoustic echo of the far-end audio signal input from the conference terminal via the conference terminal IF unit 102 and the FFT unit 105, and uses this as the pseudo echo. Output as a signal. The echo canceller unit 108 subtracts the pseudo echo signal generated by the pseudo echo generation unit 107 from the near end audio signal input from the microphone via the microphone connection unit 104 and the FFT unit 106 and includes it in the near end audio signal. Suppresses the acoustic echo that is present. At this time, the arrangement support unit 111 determines whether or not the speaker or the microphone is properly arranged based on the sound pressure level of the pseudo echo signal, and displays the determination result on the display unit 1011 of the MMIF unit 101. .

  Therefore, in the present embodiment, when the arrangement of speakers or microphones is changed during a conference, and there is a possibility that echoes and howlings that could be suppressed until then occur, the user is immediately notified of this. it can. As described above, according to the present embodiment, it is possible to assist the user to arrange the speaker and the microphone at a position where echo and howling can be sufficiently suppressed during the conference.

  Further, in the echo canceller 1 of the present embodiment, the adaptation processing unit 110 outputs a predetermined sample audio signal from the speaker via the speaker connection unit 103 and collects the sound on the microphone, and also causes the pseudo echo generation unit 107 to Then, a pseudo echo signal is generated from the sample audio signal. Then, the pseudo echo signal is compared with the sample audio signal input via the microphone connection unit 104 and the FFT unit 106. Then, the arrangement support unit 111 determines whether or not the speaker or microphone is properly arranged based on the comparison result of the adaptation processing unit 110, and displays the determination result on the display unit 1011 of the MMIF unit 101.

  Therefore, in the present embodiment, if there is a possibility that the arrangement of the speaker or microphone may cause echo or howling before the start of the conference, it can be promptly notified to the user. Thus, according to the present embodiment, it is possible to assist the user to arrange the speaker and the microphone at a position where echo and howling can be sufficiently suppressed before the start of the conference.

  In addition, this invention is not limited to said embodiment, Many deformation | transformation are possible within the range of the summary.

  For example, in the above embodiment, the determination result of whether or not the speaker or microphone is properly arranged is displayed on the display unit 1011 of the MMIF unit 101. However, the present invention is not limited to this. You may make it notify a user of the judgment result whether the arrangement | positioning of a speaker or a microphone is appropriate by audio | voice output.

  Further, in the above embodiment, the function of the echo canceller 1 shown in FIG. 1 may be executed by hardware by an integrated logic IC such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array). Alternatively, it may be realized by software by a computer such as a DSP (Digital Signal Processor). Alternatively, a PC (Personal) including an auxiliary storage device such as a CPU, memory, HDD, DVD-ROM, a communication interface such as a NIC (Network Interface Card), a display device such as an LCD or CRT, and an input device such as a keyboard or a mouse. In a general-purpose computer such as a computer, the CPU may be realized by loading a predetermined program from an auxiliary storage device onto a memory and executing it.

  DESCRIPTION OF SYMBOLS 1: Echo canceller 101: MMIF part 102: Conference terminal IF part 103: Speaker connection part 104: Microphone connection part 105: FFT part 106: FFT part 107: Pseudo echo generation part 108: Echo canceller 109: IFFT unit 110: Adaptation processing unit 111: Placement support unit 1011: Display unit 1012: Operation unit

Claims (4)

An echo canceller that eliminates the acoustic echo of the speaker output voice signal from the microphone input voice signal,
A pseudo echo generating means for estimating an acoustic echo of the output audio signal and generating a pseudo echo signal;
Echo canceling means for subtracting the pseudo echo signal from the input voice signal to suppress acoustic echo contained in the input voice signal;
An echo canceller comprising: arrangement support means for determining whether or not the speaker or the microphone is properly arranged based on a sound pressure level of the pseudo echo signal and notifying the result. The echo canceller according to claim 1,
A predetermined sample audio signal is output from the speaker and input to the microphone, and a pseudo echo signal is generated by estimating an acoustic echo of the sample audio signal in the pseudo echo generation unit, and the pseudo echo signal, Comparing means for comparing the sample audio signal input to the microphone;
The placement support means includes
An echo canceller that determines whether or not the speaker or the microphone is properly arranged based on a comparison result of the comparison means and notifies the result. A microphone and speaker placement support method connected to an echo canceller,
During the communication, the echo canceller
The acoustic echo contained in the voice signal of the other party is estimated by subtracting the estimated acoustic echo from the voice signal of the user input to the microphone. While suppressing the echo,
Based on the estimated sound pressure level of the acoustic echo, it is determined whether or not the speaker or the microphone is properly arranged, and the result is notified. A microphone and speaker arrangement support method according to claim 3,
Before the echo canceller starts communication,
A predetermined sample audio signal is output from the speaker and input to the microphone.
Estimating the acoustic echo of the sample audio signal, comparing the acoustic echo with the sample audio signal input to the microphone,
Based on the comparison result, it is determined whether or not the arrangement of the speaker or the microphone is appropriate, and the result is notified.

Images