JPS59152771A - Conference communication control system - Google Patents

Abstract

PURPOSE:To smooth the progress of a conference by setting optionally a threshold value of presence of sound volume corresponding to a participant in conference communication so as to place the priority over the right of speaking. CONSTITUTION:A voice signal of conference participants incoming to an incoming highway via a time division exchange is discriminated for a speaker and a non-speaker corresponding to the participants at a device comprising a sound volume detecting section 8, a control section 3, a threshold value of presence sound volume setting memory 5 and a control bus 4 by comparing the detected sound volume with the threshold value of presence of sound volume and the voice of participants regarded as the speaker only is synthesized as the conference speaking voice by a conference control memory 6 corresponding to participants and a voice synthesis section 7. The priority is placed over utterence among participants by making the threshold value of presence of sound volume of a participant from that of other participants, and even the voice of a participant with less sound volume is listened to when the participant is an important participant such as chairman.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a conference call control method for assigning priority to participants' comments in a conference call using a telephone or the like.

Conventionally, as means for realizing conference calls, there are the following two methods from the viewpoint of creating conference call audio.

(2) A method in which the voices of all participants are added together either as is or after a uniform loss is applied.

(2) Monitor the participants' voices for a certain period of time, detect the amount necessary to determine the presence of a sound (for example, the number of times the absolute amplitude value is the maximum), compare it across all participants, and use the resulting ranking as , a method of selecting and adding one or two people from the top.

However, in a meeting with a large number of people, the problem of distortion and noise due to addition becomes noticeable, so normally, the addition is performed after adding a loss, but it is necessary to
For example, it becomes difficult to hear the speaker's speech because the loss is added to it.

In method (2), when the chairperson makes a statement, the content of the statement may not be conveyed because the speaker's statement may not be heard or may be interrupted depending on the number of people who rank higher in the voice recognition ranking than the chairperson. .

As mentioned above, the conventional method (2) cannot prioritize the participants' comments, so it has the disadvantage that it cannot give priority to the comments of participants whose comments are more important. There is also the problem that participants with low voices have fewer opportunities to speak because they are affected by the loudness of their voices.

SUMMARY OF THE INVENTION An object of the present invention is to eliminate the drawbacks of the above-mentioned prior art and provide a conference call control system that has a function of assigning priority to participants' comments and provides higher serviceability.

To achieve the above purpose, the present invention provides a means for detecting the sound volume for a certain period of time in response to participants, and a threshold for presence volume in a certain period of time that serves as a criterion for determining whether a speaker is a speaker or a non-speaker. A function is provided to synthesize only the voice of the participant who is determined to be the speaker based on the comparison result between the detected volume and the sound volume threshold, and to assign superiority or inferiority to the speech sheets based on the difference in the sound volume threshold corresponding to the participants. It is.

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

FIG. 1 is a block diagram of a conference trunk according to the present invention;
FIG. 2 is a circuit diagram of the volume detection section of FIG. 1.

First, in FIG. 1, reference numeral 1 denotes an input highway from a time division network (not shown) to a conference trunk according to the present invention;
2 is the opposite outgoing highway, on which PCM signals (audio samples) are time-division multiplexed and transmitted and received in series. 3 is the conference trunk control unit, 4 is its control bus,
5.6 is a memory, and the memory 5 stores sound volume thresholds 1.6 corresponding to the participants. The memory 6 stores control information (abbreviated as ACT) corresponding to the participants. 7 is a voice synthesis unit, which synthesizes only the participants whose control information ACT from the memory 6 is '1'' as a conference call voice. 8 is a voice synthesis unit that synthesizes voice samples from the voice samples on input highway 1 at fixed time intervals for each channel. This is a volume detection section that detects the sum of squares as volume, and its details are shown in FIG.

In Figure 2, 10 is a serial-to-parallel conversion circuit, and 11 is a linear PCM corresponding to the quantized PCM code as an address.
This is a ROM (read-only memory) that outputs the square value of the sign. 12 is an adder circuit, 13゜14 is a memory, 15,1
6.17 is a selector. 18 counts the frame pulses FP produced by one person, and every time a certain number of FPs are counted,
A frame pulse counting circuit 19 generates a pulse signal at the output at fixed intervals.

A register 2o that repeatedly turns off is an interrupt generation circuit.

The audio samples sent in series on the input highway 1 are converted into variable signal formats for each channel by the serial/parallel conversion circuit 1o, and the ROMII converts the audio samples into a straight line P for the audio samples.
It is converted to the square value of the CM code. This voice sample square value is stored in the memory 13 (or memory 14) for each channel.
Selector 15 - Selector 17 - Addition circuit 12 (adds new audio sample squared of the channel) - Memory 13 (
calculated. Here, the operation of the memo IJs 13 and 14 will be explained. Memo! The outputs of J 13 and 14 are both input to one of selectors 15 and 16, but selector 1
When the output of the memory 13 is selected as the output of the selector 16, the output of the memory 14 appears as the output of the selector 16. On the other hand, when the output of the memory 14 is selected as the output of the selector 15, The relationship is such that the output of the memory 13 appears in the output of the memory 13.

During the period in which the memory is selected as the output of the selector 15, the sum of squares of the audio samples for the period can be held by stopping writing to the memory once the input highway is selected in the above-mentioned circular loop. . Therefore, by switching the selection of memo 1) 13.14 at regular time intervals, one side of the memory will always hold the audio samples 2 to 1 within the immediately preceding fixed time period. Here, the switching by the selector 15.16 at fixed time intervals is achieved by turning on/off the register 19 using the output of the frame pulse counting circuit 18, and controlling the selector 15.16 by the on/off output. The output of the frame pulse counting circuit 18 is transmitted to the selector 17 in the first frame of the fixed time period.
By selecting "silent 1θ" as the output,
The square value of the audio sample of the frame is directly written into the memory 13 (or memory 14) for initialization. The output of the frame pulse counting circuit 18 is simultaneously transmitted to the interrupt generating circuit 2.
0 is activated, and this circuit issues an interrupt to the control section 3 via the control bus 4. When the control unit 3 accepts the interrupt at fixed time intervals, the control unit 3 reads the sum of squares (volume) of the audio samples of the conference participants over a fixed time period via the selector 16 and the control bus 4, and stores it in the memory 5. and compares it with the voiced in value at the position assigned to the participant. According to the result, '1Ii11 supra3 is the control 1 of the participant in memory 6.
ACT on or off. That is, if the detected volume is a dog than the mF4 value, the tUIJ control information ACT is turned on, and the voice of the participant is synthesized as the conference call voice by the voice synthesis unit 7.

On the other hand, if it is small, the control information "AC'r is turned off to cut off the voice of the participant.

Here, if the sound sensitivity threshold of a specific participant is set to the minimum value in the memory 5, the voice of that participant will always be unconditionally synthesized as the conference call voice, and the voice will be given preferential speaking rights. become equivalent.

On the other hand, if the loudness threshold for a specific participant is set to the maximum value, all speech by that participant is blocked, making it possible for the participant to temporarily leave the room or sit in the room.

In addition, by preparing several sets of different calms as thresholds for the presence of sound volume, and using these one shift as a
It is possible to prioritize calls between participants during a call.

There are several methods for setting the above-mentioned active volume fl-j It for each participant, and it is possible to set it in advance or to set it at any time during the conference.

For example, before the conference convenor calls a participant, the conference convenor may access the memory 5 by dialing from a telephone or the like to set and store the threshold for responding to the participants, or the conference convener may be regarded as the chairperson using a predetermined program. method to automatically set the sub-value corresponding to the convenor to be the minimum value, and also to rewrite the threshold value in memory via the exchange by hooking from a participant or PB double signal during a conference call. This can be achieved by methods etc. These methods can be implemented using techniques similar to those used in general telephone exchange services, so detailed explanations will be omitted.

In this way, it is possible to prioritize the voices of participants depending on their positions and situations, and it is possible to make the meeting proceed smoothly.

As shown above, according to the present invention, by assigning superiority to participants in a conference call depending on their positions and situations,
It allows for flexible conference calls. Furthermore, as can be easily inferred from the above, since it is possible to realize one-way calls only in a specific direction, new services such as repeat calls are also possible.

[Brief explanation of drawings]

FIG. 1 is a block diagram of a conference trunk according to the present invention, and FIG. 2 is a circuit diagram of the volume detection section of FIG. 1. Construction...Inbound highway 2...Outbound highway 3...
-Control unit 4...Control bus 5.6. i3,1
4...Memory 7...Speech synthesis section 8...Volume detection section 10...Serial to parallel conversion circuit 11...ROM
12...Addition circuit 15.16.17...
Selector 18...Frame pulse counting circuit 19...Register 20...Interrupt generation circuit ACT...Control information FP...Frame pulse Akio Takahashi, patent attorney

Claims (1)

[Claims] 1. In a conference call device having a multi-party call function, means for constantly monitoring the conference call status and detecting the volume at a certain time in response to conference participants; A storage unit is provided in which a threshold for the presence of sound can be set, and a speaker or a non-speaker at the conference is determined by comparing the volume detected by the means at regular intervals with the threshold for the presence of sound of the participant stored in the storage. By synthesizing only the voices of the person considered to be the speaker as the conference call voice, and making the loudness threshold of one participant different from the loudness threshold of the other participants, A conference call control method characterized by assigning priority to speech. 2. Conference call control according to claim 1, characterized in that by setting the volume threshold of a specific participant to the minimum threshold, the speech of the specific participant is given priority over other participants. method. 3. The conference call control system according to claim 1, wherein the voice volume threshold of a certain number of participants is set to the maximum threshold to suppress the speech of the specific participant.