JP4138565B2 - Multi-channel audio loss control apparatus, method, and multi-channel audio loss control program - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention is a multi-channel which controls loss given to a speaker reproduction signal and a microphone sound pickup signal in order to prevent echo and howling caused by an acoustic system using one or more microphones from a plurality of speakers in a communication conference or the like. The present invention relates to a voice loss control apparatus, method, and multi-channel voice loss control program.
[0002]
[Prior art]
A communication conference is performed using a multi-channel acoustic system (see FIG. 1) having a speaker group and a microphone group configured by M (≧ 2) channel speakers and N (≧ 1) channel microphones in the same acoustic space. Think about the case. For example, when a call is made by connecting (M + 1) points, there are 1st to Mth communication partners when viewed from each point. The reception signals from the 1st to Mth counterparts are reproduced from the 1st to Mth speakers, respectively, and the N-channel microphones that pick up the own voice are transmitted to the communication counterparts. When N = 1, it is sufficient to send the same sound pickup signal to all the communication partners. When N ≧ 2, the sound pickup signals of different channels are sent for each point, or the sound pickup signals of a plurality of channels are mixed. It may be sent to the other party.
[0003]
Here, in order to prevent echo and howling from deteriorating call quality to each microphone from each speaker, there is a method disclosed in Patent Document 1 conventionally. In Patent Document 1, there is an echo canceling means, and a variable loss means to be inserted into the reception signal path and the transmission signal, and control of the loss that the variable loss means should give to the signal is described. Here, a multi-channel audio loss control apparatus (multi-channel acoustic coupling gain reduction apparatus) 100 as shown in FIG. 2 will be described regardless of the presence or absence of echo cancellation means. In the multi-channel audio loss control device 100 receives the reception signal received from a communication partner received signal input means 101, also collected sound signal input means sound collection signal collected from the microphone 2 ₁ to 2 _N Enter in 104. Based on the received 1 to M channel received signals and the 1 to N channel received signal magnitudes, the first received variable loss means 102 _{1 is} used so as to prevent howling and reduce deterioration in call quality. ˜102 _M , first loss control means 107 for controlling the amount of loss given by transmission variable loss means 105 _{1 to} 105 _N is used. For example, when only the received signal is present and the collected sound signal is not present, the first received variable loss means 102 _{1 to} 102 _M is given a small loss and the transmitted variable loss means 105 _{1 to} 105 _N is given a large loss. To control. The reception signal given the loss by the first reception variable loss means 102 _{1 to} 102 _M is output as a reproduction signal from the reproduction signal output means 103 to the speakers ₁₁ to 1 _M. On the other hand, the collected sound signal that has been lost by the transmission variable loss means 105 _{1 to} 105 _N is transmitted from the transmission signal output means 106 to the communication partner as a transmission signal.
[0004]
[Patent Document 1]
JP 2001-95084 (FIG. 2)
[0005]
[Problems to be solved by the invention]
The multi-channel voice loss control apparatus 100 shown in FIG. 2 in the prior art is based on the received signal and the collected sound signal, and the first received variable loss means 102 _{1 to} 102 _M and the transmitted variable loss means 105 ₁ to Controls the amount of loss _N 105 gives. Therefore, there are two possibilities when the magnitude of the received signal of a certain channel is very small, that is, silent. That is, there are a case where the communication partner connected to the channel is not speaking and a case where the communication partner is not connected in the first place. In these two cases, the first receiving variable loss means 102 _{1 to} 102 _M should be controlled differently. In the first place, when it is not connected to a communication partner, echo and howling do not occur due to acoustic coupling between the speaker and the microphone. Therefore, the loss to be given by the corresponding variable loss means of the first reception variable loss means 102 _{1 to} 102 _M and the transmission variable loss means 105 _{1 to} 105 _N is small. However, if the connected communication partner is not speaking, the acoustic coupling between the corresponding speaker and microphone cannot be estimated unless there is a statement, and if a small loss is given, howling may occur. Increases nature. Therefore, in a state where it is not possible to distinguish between the case where the connected communication partner is not speaking and the case where it is not connected to the communication partner in the first place, a large loss will be given as a whole. Cause the call quality to deteriorate.
The present invention provides feedback by giving a predetermined amount of loss to the reception signal of the reception channel until the connection is confirmed (after the device is started or restarted, until the reception signal is detected). It is an object of the present invention to provide a multi-channel audio loss control apparatus that can reduce voice discontinuity while avoiding it.
[0006]
[Means for Solving the Problems]
In order to solve the above-mentioned problems, the present invention includes a plurality of channels of reception signal input means, a first reception variable loss means for giving variable loss to each of the input reception signals of the plurality of channels, and the variable loss. Reproduction signal output means for outputting reproduced signals of a plurality of channels, sound pickup signal input means for one or more channels, transmission variable loss means for giving variable loss to each of the sound pickup signals of one or more channels, and the one channel Transmission signal output means for outputting a transmission signal in which a variable loss is given to the collected sound signal, the first reception variable loss means and the transmission variable based on the presence / absence or magnitude relationship of the reception signal and transmission signal In a voice loss control device having first loss control means for controlling a loss amount to be given by the loss means,
A second reception variable loss means for giving a variable loss different from the first reception variable loss means to each of the input reception signals of a plurality of channels; and each input channel of the reception signal input means and a reception signal source. The incoming call connection detecting means for detecting the connection state and the incoming call connection detecting means give a large loss to the channel of the received signal that is determined not to be connected to the received signal source, and is connected to the received signal source. Second loss control means for controlling the amount of loss given to the received signal of each channel by the second received variable loss means so as to give a small loss to the channel of the received signal determined to be It is characterized by.
[0007]
(Function)
The present invention includes a case where connected communication partner to each receiver channel is not speaking, in any of the case is not connected with the communication partner is also a Ison loss amount such to generate howling each variable loss means Giving (that is, giving a loss amount such that the gain from the input terminal of the receiving variable loss means to the output terminal of the transmitting variable loss means is less than 0 dB) and reducing howling while avoiding howling it can. In other words, in any case, a large loss is caused in the corresponding received signal path without clarifying the distinction between the case where the connected communication partner is not speaking and the case where the communication partner is not connected. I made it.
[0008]
DETAILED DESCRIPTION OF THE INVENTION
A configuration example (1) of the multi-channel audio loss control apparatus of the present invention is shown in FIG.
In the configuration of FIG. 3, in addition to the configuration of the conventional multi-channel audio loss control apparatus shown in FIG. 2, second received variable loss means 108 _{1 to} 108 _M , received connection detection means 109, and second loss control means 110 are provided. Have Here, the incoming call connection detection means 109 determines for each channel whether the input of the incoming call signal input means 101 is connected to the communication partner. The determination is made based on whether or not a received signal is present in each channel after the multi-channel audio loss control apparatus 100 of the present invention is started or restarted. After the activation or restart of the multi-channel audio loss control apparatus 100 of the present invention, a channel for which the presence of a received signal (the presence of a signal larger than a certain threshold) has been confirmed even once is determined to be connected to a communication partner. After starting or restarting the multi-channel audio loss control apparatus 100 of the present invention, a channel for which the presence of a received signal (the presence of a signal larger than a preset threshold value) has not been confirmed is connected to a communication partner. It is determined that there is no. Based on this determination result, the second loss control means 110 determines the amount of loss that the second received variable loss means 108 _{1 to} 108 _M should give the received signal of each channel. The second reception variable loss means 108 _{1 to} 108 _M of the channel determined not to be connected to the communication partner is given a large loss amount, and the second reception variable loss of the channel determined to be connected to the communication partner The means 108 _{1 to} 108 _M are given a small loss amount or no loss at all.
[0009]
As described above, the incoming connection detection means 109 solves the problem described in the problem to be solved by the invention, regardless of whether the connection with the communication partner is determined based on a very simple determination. Can do. The incoming connection detection means 109 erroneously determines that the communication partner is not connected even when the communication partner does not speak at all after the multi-channel audio loss control device 100 of the present invention is started or restarted. The second received variable loss means 108 _{1 to} 108 _M can be compensated by giving a large loss, and the amount of loss determined by the first loss control means 107 can be led to be small without causing howling. . That is, a large amount of loss is given to the second received variable loss means 108 _{1 to} 108 _M both in the case where the connected communication partner is not speaking and the case where the communication partner is not connected in the first place. Does not affect call interruption. In addition, as long as it is based on the acoustic coupling between the corresponding speaker and the microphone for the reception channel in which the reception signal exists and the connection is confirmed, the first reception variable loss means 102 _{1 to} 102 _M and the transmission variable loss means Even if the first loss control means 107 judges and controls that howling does not occur even if a small loss is given to 105 _{1 to} 105 _N , there is no received signal, and the receiving channel for which the acoustic coupling cannot be evaluated does not exist for the receiving channel. Since a large loss is inserted by the variable loss means 108 _{1 to} 108 _M , howling does not occur.
[0010]
A configuration example (2) of the multi-channel audio loss control apparatus of the present invention is shown in FIG.
The first received variable loss means 102 _{1 to} 102 _M and the second received variable loss means 108 _{1 to} 108 _{M in} FIG. 3 were put together by the third received variable loss means 111 _{1 to} 111 _M as shown in FIG. Even in this case, the same effect can be obtained. At this time, the loss given by the third reception variable loss means 111 _{1 to} 111 _M is given by the first reception variable loss means 102 _{1 to} 102 _M and the second reception variable loss means 108 _{1 to} 108 _M in each reception channel. It is equal to the product of losses (sum in decibel values (logarithmic values)).
[0011]
Also, the incoming connection detection means 109 in FIGS. 3 and 4 determines the connection state based on the magnitude of the reception signal of each reception channel after activation or reactivation, but when implemented as hardware, etc. In some cases, pulsed noise or the like due to activation upon power-on occurs in the reception channel. In order to avoid such misjudgment due to noise, it is possible to forcibly determine that all reception channels are not connected to the communication partner in a short period after activation or reactivation.
[0012]
FIG. 5 is an example in which the multi-channel audio loss control apparatus of the present invention corresponding to three speakers (M = 3) and one microphone (N = 1) is applied to three-point communication. In FIG. 4, the device of the present invention introduced at point A will be described. Point A, point B, is connected to the point C, respectively reproduce the sound from the point B the voice from the speaker 1 _2, the point C from the speaker 1 _3. Speaker 1 ₁ is not connected to the communication partner. In such circumstances, the present invention apparatus having the configuration shown in FIG. 3 is not affected by the acoustic coupling between the loudspeaker 1 ₁ and the microphone 2 _1, the transmitting variable loss means and the first receiver variable loss means Can be controlled. Therefore, the intermittent call can be reduced.
[0013]
The multi-channel audio loss control apparatus of the present invention can be composed of a computer having a CPU, a memory and the like, a terminal used by a user, and a recording medium such as a CD-ROM, a magnetic disk device, and a semiconductor memory. A multi-channel audio loss control program recorded on a recording medium or a multi-channel audio loss control program transmitted via a communication line is read by a computer to control the operation of the computer. Implement each process.
[0014]
【The invention's effect】
By using the device of the present invention, voice can be interrupted while avoiding howling in both cases where the communication partner connected to each receiving channel is not speaking and when it is not connected to the communication partner. Can be reduced.
[Brief description of the drawings]
FIG. 1 is a diagram showing a schematic configuration of a conventional multi-channel acoustic system.
FIG. 2 is a diagram showing a configuration of a conventional multi-channel audio loss control apparatus.
FIG. 3 is a diagram showing a configuration example (1) of a multi-channel audio loss control apparatus according to the present invention.
FIG. 4 is a diagram showing a configuration example (2) of the multi-channel audio loss control apparatus according to the present invention.
FIG. 5 is a diagram showing a system configuration example of multipoint communication using the multichannel audio loss control apparatus of the present invention.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Speaker, 2 ... Microphone, 100 ... Multichannel sound loss control apparatus, 101 ... Received signal input means, 102 ... First received variable loss means, 103 ... Playback signal output Means, 104 ... Collected sound signal input means, 105 ... Transmission variable loss means, 106 ... Transmission signal output means, 107 ... First loss control means, 108 ... Second reception variable Loss means, 109 ... receive connection detection means, 110 ... second loss control means, 111 ... third receive variable loss means

Claims (3)

Multi-channel received signal input means, first received variable loss means for giving variable loss to each of the input multi-channel received signals, and a reproduced signal output for outputting the multi-channel reproduced signal subjected to the variable loss Means, a sound pickup signal input means for one or more channels, a transmission variable loss means for giving a variable loss to each of the sound pickup signals of one or more channels, and a transmission for giving a variable loss to the sound pickup signals of one or more channels. The transmission signal output means for outputting the speech signal and the presence or absence or magnitude relationship of the reception signal and the transmission signal are detected. When only the reception signal is detected, the amount of loss given to the first reception variable loss means is transmitted. In a voice loss control device having a first loss control means for allocating a loss amount determined based on an acoustic coupling amount so as to be smaller than a loss amount given to a speech variable loss means ,
Provided before the first reception variable loss means, each of the input reception signals of a plurality of channels is given a variable loss different from the first reception variable loss means, and the signal given the variable loss is A second received variable loss means for inputting to the first received variable loss means;
After the voice loss control device is started or restarted, the channel in which the received signal is confirmed once is connected to the received signal source depending on whether or not a received signal exists in the received signal input means of each input channel. it is determined that the receiving connection detection means is channel also once not confirmed the reception signal determined not to be connected with the received signal source,
The received connection detection means gives a large loss to the channel of the received signal determined not to be connected to the received signal source, and the received signal channel determined to be connected to the received signal source A multi-channel audio loss control apparatus comprising: second loss control means for controlling a loss amount that the second reception variable loss means gives to the reception signal of each channel so as to give a small loss. A multi-channel received signal is input, a first variable loss is given to each of the input multi-channel received signals, and a multi-channel reproduced signal subjected to the first variable loss is output and reproduced. Input a sound collection signal of channels or more, output a transmission signal giving variable loss to each of the sound collection signals of one channel or more, and based on the presence / absence or magnitude relationship of the reception signal and the transmission signal , In the voice loss control method of allocating the amount of loss determined based on the amount of acoustic coupling so that the first received variable loss is smaller than the transmitted variable loss in the case,
A second variable loss different from the first received signal variable loss is given to each of the input reception signals of the plurality of channels, and the first received signal variable is applied to the received signal given the second variable loss. Give a loss,
After starting or restarting the voice loss control device, it is determined that the channel for which the received signal has been confirmed is connected to the received signal source even if there is a received signal on each input channel. A channel in which the received signal is not confirmed is determined not to be connected to the received signal source, and the channel of the received signal determined not to be connected to the received signal source gives a large loss and is connected to the received signal source. A multi-channel audio loss control method, comprising: controlling a variable loss amount of the second received signal so as to give a small loss to the channel of the received signal determined to be. A process of inputting a reception signal of a plurality of channels, a process of giving a first variable loss to each of the input reception signals of a plurality of channels, and outputting a reproduction signal of the plurality of channels subjected to the first variable loss A process of reproducing, a process of inputting a sound pickup signal of one or more channels, a process of giving a variable loss to each of the sound pickup signals of one or more channels, and a transmission of giving a variable loss to the sound pickup signals of one or more channels Based on the process of outputting the speech signal and the presence or absence or magnitude relationship of the received signal and the transmitted signal, in the case of only the received signal, acoustic coupling is performed so that the first received variable loss is smaller than the transmitted variable loss. In a multi-channel audio loss control program for causing a computer to execute a process of allocating a loss determined based on a volume ,
A second variable loss different from the first received signal variable loss is given to each of the input reception signals of the plurality of channels, and the first received signal variable is applied to the received signal given the second variable loss. a process of Ru given the loss,
After starting or restarting the voice loss control device, it is determined that the channel for which the received signal has been confirmed is connected to the received signal source even if there is a received signal on each input channel. every time a process channels not confirmed the reception signal you determined not to be connected with the received signal sources is also
A large loss is given to the channel of the received signal determined not to be connected to the received signal source, and a small loss is given to the channel of the received signal determined to be connected to the received signal source. A multi-channel audio loss control program including a process for controlling a loss amount of the second received signal.

Images