JP2010183434A - Voice communication system and voice communication apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology for confirming whether voice data transmitted from an own terminal has normally reproduced at an opposite-side terminal, and allowing whether an acoustic state (echo suppression) at each of bases is normal to be confirmed from a remote site, when performing voice communication over a communication network. <P>SOLUTION: An echo cancel section 15 of a terminal 10 uses a filter coefficient stored in a memory 23 to perform echo cancel processing on voice data representing a voice collected by a microphone 11. An adaptive filter of the echo cancel section 15 is initialized by a reset signal supplied from the outside. When the reset signal is received from the outside, a communication/control section 16 rewrites a value of the filter coefficient stored in the memory 23 into a predetermined initial setting value. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a voice communication system and a voice communication apparatus.

  In recent years, a remote conference system that performs a conference using a plurality of communication terminals connected via a communication network has become widespread. In Patent Document 1, in an audio conference system that transmits and receives audio data between a plurality of sites connected via a network, an echo of an input voice transmitted from another site (own site) connected via the network is disclosed. In other words, a technique has been proposed in which it is possible to confirm whether or not the audio signal is being played back with the clear voice quality expected at the other site at the other site sent back.

JP 2007-274176 A

However, in the technique described in Patent Document 1, it is necessary to provide both ON / OFF paths of the acoustic echo canceller, and there is a problem that the apparatus configuration becomes more complicated than that of the conventional apparatus. Also, with the technique described in Patent Document 1, it is possible to confirm the state in which the acoustic echo canceller is ON, but it is not possible to confirm whether it is necessary or not because the process of adaptation cannot be confirmed.
The present invention has been made in view of the above-described background, and confirms whether audio data transmitted from the own terminal is normally reproduced on the counterpart terminal when performing voice communication via the communication network. An object of the present invention is to provide a technology that enables remote confirmation of whether or not the acoustic state (echo suppression) at each site is normal.

  In order to solve the above-described problem, the present invention provides a voice communication system including a first terminal and a second terminal connected to the first terminal via a communication network, wherein the first terminal The terminal includes voice data receiving means for receiving voice data from the second terminal, sound emitting means for emitting voice indicated by the voice data received by the voice data receiving means, and surroundings of the first terminal Sound collecting means for picking up the sound, an adaptive filter for generating a pseudo echo signal based on the sound data received by the sound data receiving means, and sound data representing the sound collected by the sound collecting means. An echo canceling means comprising a subtracting section for subtracting a pseudo echo signal; a filter coefficient storage means for storing coefficients of the adaptive filter; and an echo canceling means by the echo canceling means. Audio data transmitting means for transmitting processed audio data to the second terminal, reset signal receiving means for receiving a reset signal from the second terminal, and resetting by the reset signal receiving means Filter coefficient resetting means for rewriting the value of the filter coefficient stored in the filter coefficient storage means to a predetermined initial set value when a signal is received, and processing of the echo cancellation processing from the second terminal A control signal receiving means for receiving a control signal indicating a mode, and when the control signal is received by the control signal receiving means, the echo is performed so as to perform echo cancellation processing in a processing mode indicated by the received control signal. Echo cancellation control means for controlling the cancellation means, and the second terminal is configured to receive a reverberation state and an echo Correspondence relationship storage means for storing the correspondence relationship with the processing mode of cell processing, reset signal transmission means for transmitting the reset signal to the first terminal, and transmission of voice data to the first terminal The first terminal uses the transmission means, the reception means for receiving the voice data from the first terminal, the voice data received by the reception means, and the voice data transmitted by the transmission means. A reverberation state detection unit for detecting a reverberation state in an installed space, and a processing mode of processing corresponding to the reverberation state detected by the reverberation state detection unit based on the correspondence stored in the correspondence storage unit A specifying means for specifying, and a control signal transmitting means for transmitting a control signal indicating the processing mode specified by the specifying means to the first terminal. A voice communication system is provided.

  The present invention also provides a correspondence storage unit that stores a correspondence between a reverberation situation and a processing mode of echo cancellation processing, and a reset signal transmission unit that transmits a reset signal to a terminal connected via a communication network. Voice data transmitting means for transmitting voice data to the terminal, voice data receiving means for receiving voice data from the terminal, voice data received by the voice data receiving means and transmitted by the voice data transmitting means Detected by the reverberation status detection means based on the reverberation status detection means for detecting the reverberation status in the space where the terminal is installed and the correspondence relationship stored in the correspondence relationship storage means. Control for specifying a processing mode corresponding to the specified reverberation situation and transmitting a control signal indicating the specified processing mode to the terminal Providing voice communication apparatus characterized by comprising a No. transmitting means.

  In a preferred aspect of the present invention, the correspondence relationship storage means stores a correspondence relationship between a convergence time of the adaptive filter and a processing aspect of echo cancellation processing, and the reverberation state detection means is received by the voice data reception means. You may measure the time required for the convergence of the adaptive filter in the said terminal using audio | voice data and the audio | voice data transmitted by the said audio | voice data transmission means.

  Further, in a further preferred aspect of the present invention, the correspondence relationship storage means stores a correspondence relationship between a reverberation situation and a filter coefficient, and the control signal transmission means is based on the correspondence relationship stored in the correspondence relationship storage means. Then, the filter coefficient corresponding to the reverberation situation detected by the reverberation situation detection means may be specified, and the specified filter coefficient may be transmitted to the terminal.

  Further, in a further preferred aspect of the present invention, the correspondence relationship storage means stores a correspondence relationship between a reverberation situation and a suppressor parameter, and the control signal transmission means has a correspondence relationship stored in the correspondence relationship storage means. Based on this, a parameter corresponding to the reverberation status detected by the reverberation status detection means may be specified, and the specified parameter may be transmitted to the terminal.

  In a further preferred aspect of the present invention, the information processing apparatus may further comprise notification means for notifying the reverberation situation detected by the reverberation situation detection means.

Further, in a further preferred aspect of the present invention, the audio data storing means for storing the sound data is provided, and the sound data transmitting means transmits the sound data stored in the sound data storing means to the terminal. Also good.
The voice data transmitting unit may transmit voice data representing the voice collected by the sound collecting unit to the terminal.

  The present invention also provides filter coefficient storage means for storing filter coefficients used when performing echo cancellation processing, voice data receiving means for receiving voice data from a terminal connected via a communication network, and the voice data A sound emitting means for emitting the sound indicated by the sound data received by the receiving means, a sound collecting means for collecting the surrounding sound, and a pseudo echo signal is generated based on the sound data received by the sound data receiving means And an echo canceling means provided with a subtracting section for subtracting the pseudo echo signal from the sound data representing the sound collected by the sound collecting means, and the sound subjected to echo cancellation processing by the echo canceling means Voice data transmitting means for transmitting data to the terminal, and receiving a reset signal from the terminal Reset signal receiving means and filter coefficient resetting means for rewriting the value of the filter coefficient stored in the filter coefficient storage means to a predetermined initial set value when the reset signal is received by the reset signal receiving means A control signal receiving means for receiving a control signal indicating a processing mode of the echo cancellation processing from the terminal, and a process indicated by the received control signal when the control signal is received by the control signal receiving means There is provided an audio communication apparatus comprising echo cancellation control means for controlling the echo cancellation means so as to perform echo cancellation processing in a mode.

  In a preferred aspect of the present invention, the echo cancellation control means may rewrite the filter coefficient value stored in the filter coefficient storage means to a filter coefficient value indicated by the control signal received by the control signal receiving means. Good.

  Furthermore, in a further preferred aspect of the present invention, there is provided a suppressor that performs acoustic processing on the audio data subjected to the echo cancellation processing according to a set parameter, wherein the echo cancellation control means is the control signal reception means. The parameter indicated by the control signal received by may be set in the suppressor.

  ADVANTAGE OF THE INVENTION According to this invention, when performing voice communication via a communication network, it can be confirmed whether the audio | voice data transmitted from the own terminal are normally reproduced | regenerated by the other party terminal, and the acoustic state of each base Whether or not (echo suppression) is normal can be confirmed from a remote location.

It is a block diagram which shows an example of a structure of a remote conference system. It is a block diagram which shows an example of a structure of a terminal. It is a figure which shows an example of the content of a parameter table. It is a flowchart which shows the flow of the process which a terminal performs.

<A: Configuration>
FIG. 1 is a block diagram showing a configuration of a remote conference system 1 according to an embodiment of the present invention. The remote conference system 1 is configured by connecting a plurality of terminals 10a, 10b, 10c... Installed in various places to a communication network 100 such as the Internet. In the following description, for convenience of description, when it is not necessary to distinguish the terminals 10a, 10b, 10c..., These will be referred to as “terminal 10”. A remote conference is realized by a voice conference performed by a participant of the remote conference using the terminal 10.

  FIG. 2 is a diagram illustrating an example of the configuration of the terminal 10. In the figure, microphones 11-1, 11-2,..., 11-16 each pick up sounds of persons participating in the conference (hereinafter referred to as “participants”), and sound signals representing the picked up sounds. Sound collecting means for outputting (analog signal). The audio signals output from the microphones 11-1, 11-2,..., 11-16 are output to the amplifiers 12-1, 12-2,. The amplifiers 12-1, 12-2,..., 12-16 amplify the audio signals output from the microphones 11-1, 11-2,. 2,..., 13-16. The A / D converters 13-1, 13-2,..., 13-16 convert the audio signals output from the amplifiers 12-1, 12-2,. In the following description, for convenience of description, when it is not necessary to distinguish the microphones 11-1, 11-2,..., 11-16, these will be referred to as “microphones 11”. Similarly, when it is not necessary to distinguish between the amplifiers 12-1, 12-2,..., 12-16, these will be referred to as “amplifiers 12”. Further, when it is not necessary to distinguish the A / D converters 13-1, 13-2,..., 13-16, these will be referred to as “A / D converter 13”.

  The microphone signal processing unit 14 performs, for example, delay processing on the audio signals output from the A / D converters 13-1, 13-2,. By processing, a microphone signal MB indicating directivity characteristics is generated. The microphone signal processing unit 14 transmits the generated microphone signal MB to the echo canceling unit 15.

  The echo cancellation unit 15 has a function for canceling the acoustic echo. The acoustic echo is an echo generated by acoustic feedback in which sound emitted from a speaker goes around a microphone. In the propagation of echoes from the speaker to the microphone, there are paths that are reflected by indoor walls and speakers in addition to the paths that propagate directly. The acoustic echo is characterized in that its characteristics are rapidly changed by a change in acoustic space such as movement of a speaker or a microphone with a long duration. The echo cancellation unit 15 resets the learning result of the adaptive filter 151 by a reset signal controlled by the communication / control unit 16. At this time, the communication / control unit 16 operates as a reset signal receiving unit and a filter coefficient reset unit.

The communication / control unit 16 is connected to the communication network 100 and exchanges voice data with other terminals 10 at remote locations. That is, the communication / control unit 16 also has a function as a voice data receiving unit and a voice data transmitting unit. The operation unit 17 outputs a signal corresponding to the operation content of the user. The communication / control unit 16 performs a control operation in accordance with a signal supplied from the operation unit 17.
When the communication / control unit 16 receives an eco-can control signal from another terminal 10, the communication / control unit 16 updates the gain of the echo suppressor 22 or the filter coefficient of the adaptive filter 151 according to the received parameter. At this time, the communication / control unit 16 operates as a control signal receiving unit and an echo cancellation control unit. Whether the data received from the other terminal 10 is audio data, a reset signal, or an eco-can control signal is determined by a header included in the received data.

  The speakers 18-1, 18-2,..., 18-16 are sound emitting means for emitting sound according to the supplied speaker driving signal. The speaker signal processing unit 19 generates each speaker drive signal from an audio signal transmitted from a remote place. When directivity control is performed, optimum delay processing is performed for each speaker, and output to each D / A converter 20-1, 20-2,... 20-16. The D / A converters 20-1, 20-2,..., 20-16 convert the audio data supplied from the speaker signal processing unit 19 into analog signals, and the amplifiers 21-1, 21-2,. To supply. The amplifiers 21-1, 21-2,..., 21-16 amplify the audio signals supplied from the D / A converters 20-1, 20-2,. ,..., 18-16. In the following description, for convenience of explanation, when it is not necessary to distinguish the speakers 18-1, 18-2,..., 18-16, these will be referred to as “speakers 18”. Similarly, when it is not necessary to distinguish the amplifiers 21-1, 21-2,..., 21-16, these will be referred to as “amplifiers 21”. Further, when it is not necessary to distinguish each of the D / A converters 20-1, 20-2,..., 20-16, these will be referred to as “D / A converter 20”. In the example shown in FIG. 2, an array microphone including an array microphone having 16 microphones and 16 speakers is shown, but the number of speakers and microphones is not limited to 16, and more than this. May be less.

  The echo suppressor 22 performs acoustic processing on the signal output from the echo cancellation unit 15 according to the set parameters and outputs the processed signal to the communication / control unit 16. The memory 23 stores test audio data, filter coefficients, and a parameter table. The test voice data is voice data used when the terminal 10 performs communication / reverberation confirmation processing. The sound represented by the test sound data preferably has a high frequency (for example, 3 kHz), but it only needs to have an acoustic characteristic suitable for measuring the reverberation characteristic described later.

  The filter coefficient is a parameter used when the terminal 10 performs an echo cancellation process. The echo cancellation unit 15 generates a pseudo echo Z by filtering audio data received from another terminal 10 using a filter coefficient, and performs echo cancellation processing by subtracting the generated pseudo echo Z from the microphone signal MB. .

  Next, the parameter table stored in the memory 23 will be described. As shown in FIG. 3, parameters for optimizing the acoustic echo cancellation function are written in the parameter table in association with the convergence time. In the present embodiment, gain spectrum data is associated as a parameter for optimizing the acoustic echo cancellation function. That is, the gain spectrum A (f) is associated when the convergence time t is 0 ≦ t <Ta, the gain spectrum B (f) is associated when Ta ≦ t <Tb, and the gain spectrum when Tb ≦ t <Tc. The spectrum C (f) is associated. However, Ta <Tb <Tc. The gain spectrum parameter is a parameter used when the echo suppressor 22 removes the acoustic echo component. The gain spectrum is a gain spectrum defined in a specific frequency band. In the gain spectrum, the input signal has a high value for a frequency component derived from a signal to be passed (speaker's speech) and a low value for a signal to be removed (a signal derived from an echo). Is set. The echo suppressor 22 multiplies the FFT-processed audio signal (spectrum) and the set gain spectrum to change the frequency characteristics. The gain spectrum data in FIG. 3 satisfies A (f) ≧ B (f) ≧ C (f) at an arbitrary frequency f within a defined specific frequency band. Therefore, in the case of FIG. 3, the longer the convergence time, the smaller the gain of the echo suppressor 22 and the greater the attenuation of the audio signal. As a result, an effect of attenuating the unerased echo by the echo canceler 15 by the echo suppressor 22 is achieved.

  Hereinafter, the flow of the audio signal in the echo cancellation unit 15 will be described with reference to FIG. Voice data transmitted from another terminal 10 via the communication network 100 is received by the communication / control unit 16. The received audio data is supplied to the speaker signal processing unit 19 and also supplied to the adaptive filter 151. The adaptive filter 151 generates pseudo echo Z by filtering the received audio data using the filter coefficient, and outputs the generated pseudo echo Z to the adder 152. The adaptive filter 151 is formed by combining a plurality of structural units (taps), and each tap includes a unit delay element and a multiplier. The unit delay element delays the supplied audio signal by one unit time. Each multiplier is set with a time-varying filter coefficient Wn (k) (in parentheses indicates time k; the same applies hereinafter). The input audio signal is subjected to a product-sum operation by an adaptive filter, and the operation result is generated as a pseudo echo Z. That is, the pseudo echo Z (k) is generated by a product-sum operation using each filter coefficient of the reference signal input before time k and time k.

  The echo cancellation unit 15 feeds back the audio signal output from the adder 152, that is, the processing result of the echo cancellation processing, and thereby the filter coefficient until the processing result satisfies a predetermined condition (until the adaptive filter 151 converges). Echo cancellation processing is performed while correcting the above.

  Now, the audio signal input from the microphone 11 is output to the adder 152 via the amplifier 12, the A / D converter 13, and the microphone signal processing unit 14. The adder 152 receives the audio signal (microphone signal MB) from the microphone signal processing unit 14, subtracts the pseudo echo Z received from the adaptive filter 151, and outputs the result to the echo suppressor 22. The echo suppressor 22 performs acoustic processing on the audio signal received from the adder 152 according to the set parameters, and outputs the result to the communication / control unit 16. The audio signal is output toward the communication network 100 via the communication / control unit 16.

<B: Operation>
Next, the operation of the terminal 10 will be described with reference to the drawings.
<B-1: Echo cancellation processing>
First, echo cancellation processing performed by the terminal 10 will be described. The adaptive filter 151 of the terminal 10 is initialized by a reset signal supplied from the outside. When receiving the reset signal from the outside, the communication / control unit 16 initializes the adaptive filter 151. That is, when receiving a reset signal from another terminal 10, the communication / control unit 16 rewrites the filter coefficient value stored in the memory 23 to a predetermined initial setting value. The initial setting value differs depending on the acoustic environment of the room where the terminal 10 is placed. That is, it is desirable that the adaptive filter 151 is in a state where it does not converge and that the acoustic echo generated when the adaptive filter 151 is not converged exhibits an echo cancellation capability that does not cause excessive acoustic interference. This value can be determined experimentally after the terminal 10 is installed. However, if the acoustic echo is not a major problem, such as when there is little noise and the reverberation time is short, all the filter coefficients may be set to 0 as the initial setting values.

  Next, the communication / control unit 16 causes the speaker 18 to emit sound according to the audio data received from the other terminal 10. The sound emitted from the speaker 18 generates an acoustic echo in the acoustic space where the terminal 10 is installed. The microphone 11 collects an acoustic echo. The microphone signal processing unit 14 generates a delay sum (hereinafter referred to as “echo Y”) of an audio signal representing an acoustic echo picked up by the microphone 11. The echo cancellation unit 15 uses this echo Y to update the filter coefficient. In brief, the adaptive filter 151 generates the pseudo echo Z from the audio data received from the other terminal 10 and the filter coefficient. The echo cancellation unit 15 calculates the difference between the generated pseudo echo Z and echo Y, and updates the filter coefficient according to a predetermined algorithm based on the calculation result. For this algorithm, the LMS (Least Mean Square) method is used. In brief, the echo cancellation unit 15 performs filtering so that the calculated differential signal level approaches the minimum audio signal, that is, the pseudo echo Z generated by the filter processing of the echo cancellation unit 15 is closest to the echo Y. Update the coefficient. By updating the filter coefficient in this way, the filter coefficient optimized for generating the pseudo echo Z imitating the echo Y is determined. Note that the algorithm used by the echo cancellation unit 15 may be an adaptive algorithm other than the LMS method.

  Further, when receiving the eco-can control signal from the counterpart terminal 10, the communication / control unit 16 controls the echo cancel unit 15 to perform the echo cancel process in the processing mode indicated by the received eco-can control signal. In this embodiment, the communication / control unit 16 sets a parameter indicated by the received eco-can control signal in the echo suppressor 22.

<B-2: Communication connection / acoustic confirmation processing>
Next, communication connection / acoustic confirmation processing performed by the terminal 10 will be described. FIG. 4 is a flowchart showing a flow of communication connection / acoustic confirmation processing performed by the terminal 10. First, the terminal 10 starts connection with another terminal 10 (step S1). The communication / control unit 16 of the terminal 10 determines whether or not a connection has been established (step S2). Here, when the connection is not established due to a communication failure or the like (step S2; NO), the process is terminated without performing the processes after step S3. On the other hand, if it is determined that the establishment of the connection has been completed (step S2; YES), the communication / control unit 16 starts transmitting test voice data and transmits a reset signal to the counterpart terminal 10 (Step S3).

  Next, the communication / control unit 16 uses the audio data received from the other terminal 10 and the audio data transmitted from the own terminal 10 to the other terminal 10, so that the sound in the space where the other terminal 10 is installed is used. Detect state. In this operation example, the communication / control unit 16 determines whether or not the appearance of an acoustic echo has been confirmed in the received audio data (step S4). At this time, when there is an abnormality in the sound emission system or the sound collection system of the counterpart terminal, the echo appearance may not be confirmed. If it is determined that the echo appearance cannot be confirmed (step S4; NO), the communication / control unit 16 ends the process without performing the processes after step S5.

  On the other hand, if it is determined in step S4 that the appearance of an echo has been confirmed (step S4; YES), the communication / control unit 16 determines whether or not the adaptive filter 151 in the echo cancellation unit 15 has converged. (Step S5). If it is determined that the adaptive filter 151 has converged (step S5; YES), the communication / control unit 16 ends the process as it is. Whether the adaptive filter 151 has converged is determined by determining whether the audio data received from the counterpart terminal 10 is at a predetermined level for a certain period of time with respect to the signal level of the audio data transmitted from the terminal 10 to the counterpart terminal 10. It is done depending on whether it is below. For example, it is determined that the received audio data has converged when the state of the level within ± 1% of the transmitted audio data continues for 1 second. However, this level ratio (1%) and duration (1 second) may be changed according to the installation environment of the terminal 10. For example, in a situation where there is a large amount of human movement, the level ratio may be ± 5% or less, and the duration may be 0.5 seconds.

  On the other hand, when it is determined in step S5 that the adaptive filter has not converged (step S5; NO), the communication / control unit 16 determines whether or not a predetermined time T1 has elapsed ( Step S6). If the time T1 has not elapsed (step S6; NO), the communication / control unit 16 returns to the process of step S5 and determines the convergence of the adaptive filter. On the other hand, if it is determined in step S6 that the predetermined time has elapsed (step S6; YES), the communication / control unit 16 refers to the parameter table stored in the memory 23 for the parameter corresponding to the elapsed time. To identify. In this embodiment, B (f) is specified as the gain spectrum of the echo suppressor 22 when T1 = Ta, and the gain spectrum C (f) of the echo suppressor 22 is specified when T1 = Tb. Next, the communication / control unit 16 changes the parameter of the counterpart terminal 10 by transmitting the specified parameter to the counterpart terminal 10 (step S7), and returns to the process of step S3.

  That is, in this operation example, the terminal 10 measures the time required for convergence of the adaptive filter in the counterpart terminal 10 using the received voice data and the transmitted voice data, and sets the parameter according to the measurement result to the counterpart. To the terminal 10.

  As described above, according to the present embodiment, the function of resetting the acoustic echo cancellation function (adaptive filter) by remote operation is provided, and the normal state is confirmed by checking the convergence state after the reset. Can do. Thus, by confirming the appearance of the echo, it can be confirmed whether the voice has reached the other party (communication OK / NG).

  Furthermore, according to the present embodiment, it is possible to confirm whether the echo cancellation is operating correctly by confirming the convergence state of the echo cancellation. In this way, it is possible to confirm from a remote location whether or not the acoustic state (echo suppression) at each site is normal when performing a voice conference. By doing so, it is possible to realize an audio conference in an optimal acoustic parameter environment. Moreover, according to this embodiment, since the acoustic condition of the other party's room can also be grasped, the parameter value can be adjusted according to the other party's environment.

  Note that the acoustic processing performed by the echo suppressor 22 is not limited to the above-described processing, and may be acoustic processing that removes components that cannot be removed by the echo canceling processing using the above-described adaptive filter. . When performing other acoustic processing, the parameter relating to the processing may be associated with the parameter table stored in the memory 23 instead of or in combination with the gain spectrum.

<Modification>
As mentioned above, although embodiment of this invention was described, this invention is not limited to embodiment mentioned above, It can implement with another various form. An example is shown below. In addition, you may combine each following aspect suitably.
(1) In the above-described embodiment, the case where a remote conference is performed using the terminal according to the present invention has been described. However, the present invention is not limited to this, and for example, when a lecture or lecture is performed via a communication network. The present invention can be applied.

(2) In the above-described embodiment, the convergence check of the adaptive filter is performed using the test sound data. However, instead of using the test sound data, the adaptation is performed using the sound data representing the sound collected by the microphone 11. You may make it perform the convergence check of a filter. Further, the present invention is not limited to this, and for example, a dial tone, a ring tone, or the like generated during communication with the counterpart terminal 10 may be used.

(3) Although the case where the terminal 10 is connected to the network by the wired communication network 100 has been described in the above embodiment, each device may be connected wirelessly. The communication network 100 may be a local area network (LAN) in addition to the Internet.

(4) In the above-described embodiment, the terminal 10 detects the convergence time of the adaptive filter, but the reverberation situation detected by the terminal 10 is not limited to this, and the reverberation in the space where the counterpart terminal 10 is installed. What is necessary is just to detect the reverberation situation in the space where the counterpart terminal 10 is installed using the received voice data and the transmitted voice data.

(5) In the above-described embodiment, the parameters to be set in the echo suppressor 22 are stored in the parameter table. However, the parameters stored in the parameter table are not limited to this, and may be filter coefficients of an adaptive filter, for example. Stores the correspondence between the reverberation status and the processing mode of the echo cancellation processing in advance in the terminal 10, and the terminal 10 specifies the processing mode corresponding to the detected reverberation status based on the stored correspondence. Then, a control signal indicating the specified processing mode may be transmitted to the counterpart terminal 10. For example, when using a filter coefficient as a parameter, the terminal 10 reads out the filter coefficient corresponding to the detected reverberation situation from the parameter table, and transmits the read filter coefficient to the counterpart terminal 10. In this case, when receiving the filter coefficient from the counterpart terminal 10, the terminal 10 rewrites the value of the filter coefficient stored in the memory 23 with the value of the received filter coefficient, and performs echo cancellation using the rewritten filter coefficient. Process. The number of taps of the adaptive filter may be used as a parameter instead of the filter coefficient. As a result, the number of taps of the adaptive filter can be set in accordance with the length of the reverberation time.

(6) In the above-described embodiment, the terminal 10 may notify the detected reverberation situation. Specifically, for example, an image representing the detected reverberation situation may be displayed on a display device such as a liquid crystal display, or for example, the detected reverberation situation may be simulated.

(7) Each unit of the terminal 10 in the above-described embodiment may be configured as hardware, and a control unit such as a CPU (Central Processing Unit) executes a computer program stored in a storage unit such as a hard disk. It may be realized as software. In this case, the program executed by the control unit is recorded on a computer-readable recording medium such as a magnetic recording medium (magnetic tape, magnetic disk, etc.), an optical recording medium (optical disk, etc.), a magneto-optical recording medium, or a semiconductor memory. Can be provided in the state. It is also possible to download to the terminal 10 via a network such as the Internet.

DESCRIPTION OF SYMBOLS 1 ... Remote conference system, 10 ... Terminal, 11 ... Microphone, 12 ... Amplifier, 13 ... A / D converter, 14 ... Microphone signal processing part, 15 ... Echo cancellation part, 16 ... Communication / control part, 17 ... Operation part, DESCRIPTION OF SYMBOLS 18 ... Speaker, 19 ... Speaker signal processing part, 20 ... D / A converter, 21 ... Amplifier, 22 ... Echo suppressor, 23 ... Memory, 100 ... Communication network, 151 ... Adaptive filter, 152 ... Adder.

Claims (11)

An audio communication system having a first terminal and a second terminal connected to the first terminal via a communication network,
The first terminal is
Voice data receiving means for receiving voice data from the second terminal;
Sound emitting means for emitting the sound indicated by the sound data received by the sound data receiving means;
Sound collecting means for collecting sound around the first terminal;
An adaptive filter that generates a pseudo echo signal based on the voice data received by the voice data receiving unit; and a subtracting unit that subtracts the pseudo echo signal from the voice data representing the voice collected by the sound collecting unit. Echo canceling means,
Filter coefficient storage means for storing coefficients of the adaptive filter;
Audio data transmitting means for transmitting the audio data subjected to echo cancellation processing by the echo canceling means to the second terminal;
Reset signal receiving means for receiving a reset signal from the second terminal;
Filter coefficient resetting means for rewriting the value of the filter coefficient stored in the filter coefficient storage means to a predetermined initial setting value when the reset signal is received by the reset signal receiving means;
Control signal receiving means for receiving a control signal indicating a processing mode of the echo cancellation processing from the second terminal;
Echo cancellation control means for controlling the echo cancellation means so as to perform echo cancellation processing in a processing mode indicated by the received control signal when the control signal is received by the control signal receiving means,
The second terminal is
Correspondence storage means for storing the correspondence between the reverberation situation and the processing mode of echo cancellation processing;
Reset signal transmitting means for transmitting the reset signal to the first terminal;
Transmitting means for transmitting voice data to the first terminal;
Receiving means for receiving audio data from the first terminal;
Reverberation condition detecting means for detecting a reverberation condition in a space where the first terminal is installed, using the voice data received by the receiving means and the voice data transmitted by the transmitting means;
Identification means for identifying a processing mode of processing corresponding to the reverberation situation detected by the reverberation situation detection means based on the correspondence relation stored in the correspondence relation storage means;
A voice communication system comprising: a control signal transmitting unit that transmits a control signal indicating the processing mode specified by the specifying unit to the first terminal. Correspondence storage means for storing the correspondence between the reverberation situation and the processing mode of echo cancellation processing;
Reset signal transmitting means for transmitting a reset signal to a terminal connected via a communication network;
Voice data transmitting means for transmitting voice data to the terminal;
Voice data receiving means for receiving voice data from the terminal;
Reverberation condition detecting means for detecting a reverberation condition in the space where the terminal is installed, using the sound data received by the sound data receiving means and the sound data transmitted by the sound data transmitting means;
Based on the correspondence stored in the correspondence storage unit, the processing mode corresponding to the reverberation situation detected by the reverberation situation detection unit is specified, and a control signal indicating the specified processing mode is transmitted to the terminal. A voice communication apparatus comprising: a control signal transmission unit. The correspondence storage means stores the correspondence between the convergence time of the adaptive filter and the processing mode of echo cancellation processing,
The reverberation state detecting means measures the time required for convergence of the adaptive filter in the terminal using the voice data received by the voice data receiving means and the voice data transmitted by the voice data transmitting means. The voice communication apparatus according to claim 2. The correspondence storage means stores a correspondence between a reverberation situation and a filter coefficient,
The control signal transmission means identifies a filter coefficient corresponding to the reverberation situation detected by the reverberation situation detection means based on the correspondence relation stored in the correspondence relation storage means, and sends the identified filter coefficient to the terminal. The voice communication apparatus according to claim 2, wherein the voice communication apparatus transmits the voice communication apparatus. The correspondence storage means stores a correspondence between a reverberation situation and a suppressor parameter,
The control signal transmission unit identifies a parameter corresponding to the reverberation situation detected by the reverberation situation detection unit based on the correspondence stored in the correspondence storage unit, and transmits the identified parameter to the terminal. The voice communication apparatus according to any one of claims 2 to 4, wherein the voice communication apparatus is characterized. The voice communication apparatus according to any one of claims 2 to 5, further comprising notification means for notifying a reverberation situation detected by the reverberation situation detection means. Voice data storage means for storing voice data;
The voice communication apparatus according to any one of claims 2 to 6, wherein the voice data transmitting unit transmits the voice data stored in the voice data storage unit to the terminal. The voice communication apparatus according to claim 2, wherein the voice data transmission unit transmits voice data representing the voice collected by the voice collection unit to the terminal. Filter coefficient storage means for storing a filter coefficient used when performing echo cancellation processing;
Audio data receiving means for receiving audio data from a terminal connected via a communication network;
Sound emitting means for emitting the sound indicated by the sound data received by the sound data receiving means;
Sound collecting means for collecting surrounding sounds;
An adaptive filter that generates a pseudo echo signal based on the voice data received by the voice data receiving unit; and a subtracting unit that subtracts the pseudo echo signal from the voice data representing the voice collected by the sound collecting unit. Echo cancellation means;
Audio data transmitting means for transmitting the audio data subjected to echo cancellation processing by the echo canceling means to the terminal;
Reset signal receiving means for receiving a reset signal from the terminal;
Filter coefficient resetting means for rewriting the value of the filter coefficient stored in the filter coefficient storage means to a predetermined initial setting value when the reset signal is received by the reset signal receiving means;
Control signal receiving means for receiving a control signal indicating a processing mode of the echo cancellation processing from the terminal;
Echo cancellation control means for controlling the echo cancellation means to perform echo cancellation processing in the processing mode indicated by the received control signal when the control signal is received by the control signal receiving means. A voice communication apparatus characterized by the above. The echo cancellation control means rewrites the filter coefficient value stored in the filter coefficient storage means to a filter coefficient value indicated by a control signal received by the control signal receiving means. The voice communication device according to the description. A suppressor for performing acoustic processing on the audio data subjected to the echo cancellation processing according to a set parameter;
The voice communication apparatus according to claim 9 or 10, wherein the echo cancellation control unit sets a parameter indicated by a control signal received by the control signal reception unit in the suppressor.

Images