JP2002223422A - Multi-point video conference controller and video packet transmission method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a multi-point video conference system that effectively switches images at a low cost. SOLUTION: Discriminating number of effective voice packets not silent packets per time received from each video conference terminal connected to a multi-point video conference controller detects a talker terminal and the detected talker terminal acting like a sender source video phone terminal delivers a selected image video packet to all video conference terminals.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video conference system, and more particularly, to a method of transmitting video packets to each video conference terminal from a multipoint video conference controller connected to a video conference terminal that does not transmit audio packets when there is no sound. And equipment.

[0002]

2. Description of the Related Art Conventionally, this type of multipoint video conference control device selects an image of a video conference terminal which is a speaker and transmits the selected image to all participating video conference terminals so that a received image of each video conference terminal can be obtained. It is used for displaying the image of the speaker.

For example, Japanese Patent Application Laid-Open No. 2000-134596 describes a technique for detecting a voice level (sound power) received from each video conference terminal and using the video conference terminal having the maximum voice level as a switching image. I have.

FIG. 11 is a block diagram showing an example of a conventional multipoint videophone control device. In the conference room 1, the sound uttered by the conference participants is sent to the video conference terminal 512 via the microphone 510 and the amplifier 511, and is multiplexed with an image signal obtained from an unillustrated imaging device after being compression-encoded and multiplexed. Sent to the line. The multiplexed signal is supplied to the line interface unit 514 of the multipoint video conference controller 513.
-1, the audio signal and the image signal are separated by the MUX / DEMUX 515-1. The separated audio signal is decoded by the audio decoding unit 517-1 and distributed to the control unit 516 and the audio signal attenuating unit 520-1. Similarly, voices emitted in the conference rooms 2 and 3 are also transmitted to the line interface unit 514.
-2, 514-3, MUX / DEMUX 515-2,5
15-3, the control unit 516 and the audio signal attenuating units 520-2, 520 via the audio decoding units 517-2, 517-3.
-3.

FIG. 12 shows a detailed configuration of the control unit 516. Each audio signal input to the control unit 516 is output from an audio level detection unit 530-1, 530-2, 530.
At -3, a level measurement is performed only for a predetermined fixed protection time, and the maximum audio level is measured. The signal representing the measured maximum audio level is sent to the maximum level adjustment unit 531 and the difference between the signal and the existing image switching threshold is calculated. The maximum level adjustment unit 531 is provided for the conference room 1,
The sound level attenuation is calculated so that the differences corresponding to 2 and 3 become equal. The calculated audio level attenuation amount and the estimated difference value after the attenuation adjustment are notified to the control circuit 532, and the control circuit 532 that has received the notification controls the attenuation control of the audio signal attenuation unit 520-1 via the attenuation control unit 533. Do.

In FIG. 11, an audio signal attenuating section 520-
The audio signals smoothed in 1, 520-2, and 520-3 are sent to the audio addition unit 522, where they are added.
At the time of addition, it is measured whether any of the audio signals exceeds the image switching threshold, and the information is sent to the control circuit 532 of the control unit 516. The control circuit 532 determines the source of the distributed image signal based on this information, and
Control. The image switching unit 519 is connected to each video conference terminal 5
12 to select an image signal to be supplied to the MUX / DEMU
X515-1, 515-2, 515-3.

[0007]

The above-mentioned prior art has the following problems.

The problem is that the audio levels from the conference rooms 1, 2, and 3 are smoothed, the audio levels are compared, and the conference room with the higher level is determined to switch the image. .

The reason is that it is necessary to perform a large amount of arithmetic processing for smoothing the audio level and detecting the maximum audio level, which causes problems such as an increase in hardware cost.

It is an object of the present invention to provide a multipoint video conference control apparatus and a video packet transmission method that can effectively and inexpensively switch images by solving the above problems.

[0011]

The video packet transmitting method of the multipoint video conference control device according to the present invention comprises the steps of: transmitting a video packet from a multipoint video conference control device connected to a video conference terminal that does not transmit audio packets during silence; Detecting a speaker terminal by determining the number of valid non-silent voice packets per time received from each video conference terminal, and transmitting the detected speaker terminal. Delivering the selected image video packet to all video conference terminals connected to the multipoint video conference control device as the former video telephone terminal.

According to these processes, in a multipoint video conference apparatus connected to a video conference terminal that does not transmit voice packets when there is no sound, it is possible to effectively identify a speaker terminal and transmit a selected image packet to be delivered to all video conference terminals. Can be switched to the image packet of the user terminal.

The step of detecting a speaker terminal is performed by inputting voice packets received from a plurality of video conference terminals via the Internet and subjected to decompression processing, and the number of valid voice packets that are not silence per unit time. Counting, the terminal that has received the most packets is identified, the identification signal for identifying the terminal as the speaker terminal is output, and the step of delivering the selected image video packet is performed by a plurality of video conference terminals via the Internet. The present invention also includes a method of inputting a received image packet, switching a selected image packet as a speaker terminal according to the identification signal input at the same time, and transmitting the selected image packet to each video conference terminal.

The step of detecting a speaker terminal is performed by inputting voice packets received from a plurality of video conference terminals via the Internet and subjected to expansion processing, and the number of valid voice packets that are not silence per unit time. And a step of inputting and comparing the number of valid voice packets per unit time, and selecting a video conference terminal that has transmitted the maximum number of packets per time as a speaker terminal. Outputting a terminal selection signal.

[0015] The multipoint video conference control device of the present invention is a multipoint video conference control device for holding a video conference in which a plurality of video conference terminals that do not transmit voice packets during silence are connected via the Internet. A speaker discriminator for detecting a speaker terminal by judging the number of valid voice packets that are not silence received from the video conference terminal per time; and a multi-point as the source video conference terminal using the detected speaker terminal as a source video conference terminal An image switching unit that delivers the selected image video packet to all the video conference terminals connected to the video conference control device.

The speaker discriminating unit inputs voice packets received from a plurality of video conference terminals via the Internet and subjected to decompression processing, and counts the number of valid non-silent voice packets per unit time. Discriminating a terminal that has received the most packets, and outputs an identification signal that identifies the terminal as a speaker terminal; the image switching unit inputs image packets received from a plurality of video conference terminals via the Internet;
It includes a method of switching a selected image packet as a speaker terminal according to the identification signal input at the same time and transmitting the selected image packet to each video conference terminal.

The speaker discriminating section is provided corresponding to the plurality of video conference terminals, receives a voice packet transmitted from the corresponding video conference terminal, counts the number of packets per time, and outputs the packet. Performing a comparison by inputting the number of packets per time output by the packet counting unit,
And a packet number comparison unit that outputs a speaker terminal selection signal for selecting a video conference terminal that has transmitted the maximum number of packets per time as a speaker terminal.

[0018]

Next, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 shows a flowchart of an embodiment of a video packet transmitting method of the multipoint video conference controller according to the present invention.

As shown in FIG. 1, the video packet transmission method of the multipoint video conference control apparatus according to the embodiment of the present invention is not a silence received from a plurality of video conference terminals that do not transmit voice packets when there is no voice. The speaker terminal is detected by determining the number of valid voice packets per time (step S1). At this time, audio packets received from each video conference terminal and subjected to decompression processing via the Internet are input, and the number of valid non-silent audio packets per unit time is counted. By comparing the count values, the terminal that has received the most packets is determined, and a speaker terminal identification signal for identifying this terminal as a speaker terminal is output. Subsequently, the selected image video packet is delivered to all the video conference terminals connected to the multipoint video conference control device, using the speaker terminal detected in step S1 as a source video conference terminal (step S2). At this time, image packets received from a plurality of video conference terminals via the Internet are input, and the selected image packets are switched according to the simultaneously input speaker terminal identification signal, and transmitted to each video conference terminal.

According to these processes, in a multipoint video conference apparatus to which a video conference terminal that does not transmit voice packets during silence is connected, a speaker terminal is effectively identified and a selected image packet to be delivered to all video conference terminals is transmitted. Can be switched to the image packet of the user terminal.

FIG. 2 is a block diagram of a multipoint video conference system including an embodiment of the multipoint video conference device of the present invention, and FIG. 3 is a block diagram of the multipoint video conference device MCU of FIG.

Referring to FIG. 2, the multipoint video conference control unit MCU is connected to video conference terminals EP-1, EP-2, EP-3, and EP-4 via the Internet 10.

Referring to FIG. 3, the multipoint video conference control unit MCU includes a network interface unit 1 and first, second, third and fourth audio compression / decompression units 2-1 and 2-.
2, 2-3, 2-4, a voice addition unit 3, a speaker determination unit 4, an image switching unit 5, and a call control unit 6.

The speaker discriminating unit 4 communicates with the video conference terminals EP-1, EP-2, EP-3, and EP via the Internet 10.
Packet ADo received from -4 and subjected to decompression processing
-1, ADo-2, ADo-3, ADo-4
The number of valid voice packets that are not silence per unit time is counted, the terminal that has received the most packets is determined, and a speaker terminal identification signal SEL is output to the image switching unit 5. The image switching unit 5 transmits the image packets Vi-1, Vi-2, and Vi-3V received from the video conference terminals EP-1, EP-2, EP-3, and EP-4 via the Internet 10.
i-4 is input and the speaker terminal identification signal SE is input at the same time.
The selected image packet is switched according to L. The selected image packets are Vo-1, Vo-2, Vo-3, and Vo-4 as video conference terminals EP-1, EP-2, EP-3, and E, respectively.
Sent to P-4.

According to these processes, in a multipoint video conference apparatus to which a video conference terminal that does not transmit voice packets during silence is connected, a speaker terminal is effectively identified and a selected image packet to be delivered to all video conference terminals is transmitted. Can be switched to the image packet of the user terminal.

The configuration of the multipoint video conference apparatus according to the present embodiment will be described in more detail.

The network interface unit 1 is connected to each of the video conference terminals EP-1 and EP via the Internet 10.
-2, EP-3, EP-4 to network data ND
-1, ND-2, ND-3, ND-4 are transmitted and received. The network interface unit 1 includes first, second, third, and fourth audio compression / decompression units 2-1, 2-2, 2-3,
2-4, each video conference terminal EP-1, EP-2, EP-
3. Compressed voice packet AEi- received from EP-4
1, AEi-2, AEi-3, and AEi-4, respectively, and outputs the first, second, third, and fourth audio compression / decompression units 2-
1, 2-2, 2-3, 2-4 to each video conference terminal EP
-1, EP-2, EP-3, compressed voice packets AEo-1, AEo-2, AEo-3,
Enter AEo-4. Further, the network interface unit 1 provides the image switching unit 5 with each video conference terminal E
Compressed image packets Vi-1, Vi-2, Vi-3, Vi- received from P-1, EP-2, EP-3, EP-4
4 are output, and the compressed image packets Vo-1, Vo-2, Vo-3, and -3 are transmitted from the image switching unit 5 to the video conference terminals EP-1, EP-2, EP-3, and EP-4.
Enter Vo-4. In addition, the network interface unit 1 transmits and receives a call control signal CI to and from the call control unit 6. First, second, third and fourth audio compression / decompression units 2
-1, 2-2, 2-3, and 2-4 are compressed voice packets AEi-
1, AEi-2, AEi-3, and AEi-4 are input and subjected to voice expansion processing, and then expanded voice packets ADi-1, ADi-2, A
Di-3 and ADi-4 are output. The voice adding unit 3 includes first, second, third, and fourth voice compression / decompression units 2-1 and 2-1.
-2, 2-3, and 2-4, decompressed voice packet ADi-
1, ADi-2, ADi-3, and ADi-4 are input and subjected to audio addition processing, and then the first, second, third, and fourth audio compression units 2-1, 2-2, 2- For 3, 2-4,
Decompressed voice packet ADo-1 subjected to voice addition processing,
ADo-2, ADo-3, and ADo-4 are output. The speaker discriminating unit 4 receives the expanded voice packets ADi-1, ADi-2, ADi-2, ADi-3, ADi-4
Is input, the current speaker terminal is determined, and a speaker terminal selection signal SEL for identifying the speaker terminal is output to the image switching unit 5. The image switching unit 5 sends the compressed image packets Vi-1, Vi-2, Vi-
3, Vi-4 is input and the compressed image packets Vo-1, V
o-2, Vo-3, and Vo-4 are output. At the same time, a speaker terminal selection signal SEL is input from the speaker determination unit 4.

FIG. 4 is a block diagram of the speaker discriminating section 4 shown in FIG.

The decompressed voice packets ADi-1, ADi-2, inputted from the first, second, third, and fourth voice compression / decompression units 2-1, 2-2, 2-3, 2-4, respectively. ADi-
3. ADi-4 is input to the first, second, third and fourth packet counting units 40-1, 40-2, 40-3 and 40-4. The first, second, third, and fourth packet counting units 40-1, 40-2, 40-3, and 40-4 respectively input the decompressed voice packets ADi-1 and ADi-
2, the number of packets per time of ADi-3 and ADi-4 is counted, and the number of packets NUM-
1, NUM-2, NUM-3 and NUM-4 are output.
The number-of-packets comparing unit 41 calculates the number of input packets NUM−
1, NUM-2, NUM-3, and NUM-4 are compared, and a speaker terminal selection signal SEL is output.

FIG. 5 is a block diagram of the voice adding unit 3 shown in FIG.

The adder ADD-1 performs an addition process on the input decompressed voice packets ADi-2, ADi-3, and ADi-4, and outputs a decompressed voice packet ADo-1. The adder ADD-2 performs an addition process on the input decompressed voice packets ADi-1, ADi-3, and ADi-4, and outputs a decompressed voice packet ADo-2. Similarly, the adder ADD-3 outputs the input decompressed voice packet ADi.
-1, ADi-2 and ADi-4 are added, and an expanded voice packet ADo-3 is output. Further, adder A
DD-4 is an input decompressed voice packet ADi-1,
ADi-2 and ADi-3 are added and an expanded voice packet ADo-4 is output.

FIG. 6 is a block diagram of the image switching unit 5 of FIG.

In the image switching section 5, the compressed image packets Vi-1, Vi-2, Vi-3, Vi- are inputted according to the speaker terminal selection signal SEL inputted by the switch SW.
4 is selected, and the selected compressed image packet is output as compressed image packets Vo-1, Vo-2, Vo-3, and Vo-4.

FIG. 7 shows the video conference terminals EP-1, E of FIG.
The block diagram of P-2, EP-3, EP-4 is shown.

The microphone 100 outputs the collected audio signal to the audio compression unit 101. The audio compression unit 101 performs audio compression processing on the input audio signal, and outputs a compressed audio packet to the network interface unit 108. The video camera 102 outputs a captured image signal to the image compression unit 103. The image compression unit 103 performs an image compression process on the input image signal, and outputs a compressed audio packet to the network interface unit 108. The network interface unit 108 transmits the input compressed voice packet and the compressed image packet to the MCU connected via the Internet 10 to the network data ND-n.
Send as At the same time, the network data ND-n received from the MCU is input to the network interface unit 108, and outputs a compressed image packet and a compressed audio packet to the image expansion unit 104 and the audio expansion unit 106, respectively. The image decompression unit 104 performs image decompression processing on a compressed image packet received from the multipoint video conference device MCU via the Internet 10 and outputs an image signal. The display 105 draws an image signal input from the image expansion unit 104. The audio decompression unit 106 performs an audio decompression process on a compressed audio packet received from the multipoint video conference device MCU via the Internet 10 and outputs an audio signal. The speaker 107 reproduces an audio signal input from the audio expansion unit 106.

(Description of Operation) Next, the operation of this embodiment will be described.

In FIG. 2, a multipoint video conference control unit MCU is connected to each of the video conference terminals EP-1, EP-2, EP-3, and EP-4 via the Internet 10, and receives compressed audio packets and compressed images. Network data ND-1, ND-2, ND-3, ND including packet
-4 is transmitted and received to realize a multipoint video conference.

Referring to FIG. 3, each of the video conference terminals EP-1, EP-2, and EP-
3. Received from EP-4, voice compression / decompression units 2-1 and 2-
2, 2-3, and 2-4, decompressed voice packets ADi-1, ADi-2, and AD subjected to voice decompression processing, respectively.
i-3 and ADi-4 are input to the speaker determination unit 4. The speaker identification unit 4 receives the input expanded voice packet ADi-1,
From the number of decompressed voice packets per unit time of ADi-2, ADi-3, and ADi-4, the speaker of the current video conference can determine the video conference terminals EP-1, EP-2, EP-3, E
One of P-4 is determined, and a speaker terminal selection signal SEL is output. At the same time, each of the video conference terminals EP-1, EP-2, EP-3, and EP-4 via the Internet 10.
Compressed video packets Vi-1, Vi-2, V received from
i-3 and Vi-4 are input to the image switching unit 5. The image switching unit 5 again transmits the video conference terminals EP-1, EP-
2, a compressed image packet V output to EP-3 and EP-4
Compressed image packets Vi-1, Vi-2, Vi-3, Vi according to the speaker terminal selection signal SEL output by the speaker discriminating unit 4 as o-1, Vo-2, Vo-3, and Vo-4. -4
Is output. That is, the image of the speaker terminal of the current video conference is transmitted to all the video conference terminals as the selected image.

Referring to FIG. 4, each video conference terminal EP-1, EP-2, EP-
3. Received from EP-4, voice compression / decompression units 2-1 and 2-
2, 2-3, and 2-4, decompressed voice packets ADi-1, ADi-2, and AD subjected to voice decompression processing, respectively.
i-3 and ADi-4 are respectively a packet counting unit 40-
1, 40-2, 40-3, and 40-4. The packet counting units 40-1, 40-2, 40-3, and 40-4 each count the number of input expanded voice packets per unit time, and count the number of packets NUM-
1, NUM-2, NUM-3 and NUM-4 are output.
The packet number comparing unit 41 includes packet counting units 40-1 and 40
-2, number of packets N input from 40-3, 40-4
UM-1, NUM-2, NUM-3, and NUM-4 are input, the input having the largest number of packets is determined to be the speaker of the video conference, and the speaker terminal selection signal SEL is output.

FIG. 8 shows the packet counting units 40-1 and 4 in FIG.
The figure explaining operation | movement of 0-2, 40-3, and 40-4 is shown.

Now, each of the video conference terminals EP-1, EP-
2. Decompressed voice packets ADi-1, ADi-2, ADi received from EP-3 and EP-4 and subjected to voice decompression processing
-3, ADi-4 are at time timings T1, T
2,. . . , T5. At this time, the number of decompressed voice packets received from each video conference terminal is counted, and the number of packets NUM-1, NUM-2, NUM-3, NU
M-4 is obtained. At this time, if no decompressed voice packet is received from the video conference terminal, counting is not performed. In the example of FIG.
NUM-1 is 5, NUM-2 is 3, NUM-3 is 4, N
For UM-4, 2 and the number of packets NUM are counted.

FIG. 9 is a diagram for explaining the operation of the packet number comparison unit 41 of FIG.

Now, the packet counting units 40-1, 40-2, 40-3, and 40-4 corresponding to the respective video conference terminals output the packet numbers NUM-1, NUM-2, and NUM-, respectively.
3. When NUM-4 is input, the packet number comparing unit 41
Compares the number of packets, determines that the video conference terminal corresponding to the maximum number of packets is the current speaker video conference terminal, and outputs a speaker terminal selection signal SEL. In the example of FIG. 9, NUM-1 = 5, NUM-2 = 3,
NUM-3 = 4, NUM-4 = 2 are compared, and NUM-1
Has the largest number of packets, so the speaker terminal selection signal SE
L = 1 is output.

In FIG. 6, according to the speaker terminal selection signal SEL inputted by the switch SW, the video conference terminal EP
-1, EP-2, EP-3, and compressed image packets Vi-1, Vi-2, Vi-3 received and input from EP-4,
Vi-4 is selected, and the selected compressed image packets are converted into compressed image packets Vo-1, Vo-2, Vo-3, and Vo-4.
Output as Therefore, each video conference terminal EP-1,
The compressed image packet from the speaker terminal is output to each of EP-2, EP-3, and EP-4, and the image of the speaker terminal is displayed in the same manner on all the video conference terminals. . In FIG. 6, the speaker terminal selection signal SEL = 1
Therefore, the compressed image packets Vo-1, Vo-2, V
o-3 and Vo-4 are all compressed image packets Vi-1, and the result is that all the video conference terminals display an image output from the multipoint conference control unit MCU by the video conference terminal EP-1.

The operation of the voice adding unit 3 will be described with reference to FIG.

The audio adder 3 is connected to each video conference terminal EP-
1, EP-2, EP-3, EP-4, and decompressed voice packets ADi- which have been subjected to voice decompression processing in the voice compression / decompression units 2-1, 2-2, 2-3, 2-4, respectively.
1, ADi-2, ADi-3, and ADi-4 are input. On the other hand, the output is an extended voice packet ADo-1 to be finally transmitted to the video conference terminal EP-1, an extended voice packet ADo-2 to be transmitted to the video conference terminal EP-2, and a video conference. It outputs an expanded voice packet ADo-3 to be transmitted to the terminal EP-3 and an expanded voice packet ADo-4 to be transmitted to the video conference terminal EP-4. Here, when attention is paid to the expanded voice packet ADo-1, the expanded voice packet A
Do-1 is a teleconference terminal EP-2, EP-3, EP
, The expanded voice packets ADi-2 and ADi received from
-3 and ADi-4 are added by an adder ADD-1. Similarly, the decompressed voice packet ADo-2 is obtained by adding the decompressed voice packets ADi-1, ADi-3, and ADi-4 received from the video conference terminals EP-1, EP-3, and EP-4 by the adder ADD-2. , The extended voice packet ADo-3 is a teleconference terminal EP-1, EP-2, E
Decompressed voice packets ADi-1, AD received from P-4
i-2 and ADi-4 are added by an adder ADD-3, and the decompressed voice packet ADo-4 is a video conference terminal E
The decompressed voice packets ADi-1, ADi-2 and ADi-3 received from P-1, EP-2 and EP-3 are added to the adder AD
This is the result of addition processing at D-4. By these operations, the voice adding unit 3 performs an addition process on voices other than the voice input from the video conference terminal of the transmission destination and outputs the processed voice.

Next, referring to FIG. 7, a video conference terminal EP-
1, the operations of EP-2, EP-3, and EP-4 will be described.

[0049] The audio compression unit 101 includes the microphone 10
A compressed voice packet obtained by subjecting a voice signal inputted from 0 to voice compression processing is transmitted to the network interface unit 108,
The data is transmitted to the multipoint video conference control unit MCU via the Internet 10. Similarly, the image compression unit 103 converts a compressed image packet obtained by performing image compression processing on an image signal captured by the video camera 102 into a network interface unit 10.
8. Transmission to the multipoint video conference control unit MCU via the Internet 10. Further, the image decompression unit 104 transmits the Internet 1 from the multipoint video conference control unit MCU.
0, receive the compressed image packet via the network interface unit 108. Here, the image decompression unit 104 performs an image decompression process, and outputs the obtained video signal to a display, whereby an image is displayed on the display. Here, since the compressed image packet output by the multipoint video conference control unit MCU is a compressed image packet output by the video conference terminal that is the current speaker, the display uses the camera of the speaker's video conference terminal. The displayed video is displayed.

FIG. 10 is a diagram for explaining the operation of the audio compression unit 101 in FIG.

Now, when the audio compression section 101 performs the audio compression processing, a silence threshold TH for judging silence is provided. Next, as illustrated in FIG. 10B, when the audio compression unit 101 performs the audio compression process and finds that the audio signal satisfies the silence threshold TH illustrated in FIG. Indicates that the audio signal exceeds the silence threshold TH, and the compression process is performed as a normal audio packet. FIG. 10C shows that among the compressed voice packets, the packets output by the voice compression unit 101 are only valid voice packets excluding silence packets. This indicates that the audio packets transmitted from the video conference terminal EP-n to the multipoint video conference control unit MCU are only valid audio compression packets.

With these operations, the multipoint video conference control device according to the present invention provides the present speaker video conference terminal with the number of valid (non-silent) compressed voice packets input from each video conference terminal per unit time. Is determined, and the image of the speaker video conference terminal is transmitted to all the video conference terminals.

[0053]

As described above, according to the present invention, the speaker terminal is effectively detected by judging the number of valid non-silent voice packets per time received from each video conference terminal. A multipoint videoconference control device that effectively and inexpensively switches images is realized by using the videophone terminal that transmits video packets to all videoconference terminals connected to the local videoconference control device as the speaker terminal. There is an effect that can be.

[Brief description of the drawings]

FIG. 1 is a flowchart of an embodiment of a video packet transmission method of a multipoint video conference controller according to the present invention.

FIG. 2 is a block diagram of a multipoint video conference system including an embodiment of the multipoint video conference device of the present invention.

FIG. 3 is a block diagram of the multipoint video conference device MCU of FIG. 2;

FIG. 4 is a block diagram of a speaker determination unit 4 of FIG. 3;

FIG. 5 is a block diagram of a voice adding unit 3 of FIG. 3;

FIG. 6 is a block diagram of the image switching unit 5 of FIG. 3;

FIG. 7 is a block diagram of the video conference terminal EP-n in FIG. 2;

FIG. 8 is a diagram illustrating the operation of the packet counting unit 40-n in FIG.

FIG. 9 is a diagram illustrating the operation of the packet number comparison unit 41 of FIG.

10A and 10B are diagrams illustrating the operation of the audio compression unit 101 in FIG. 7, wherein FIG. 10A illustrates an audio signal from the microphone 100 and a silence threshold, and FIG. 10B illustrates an audio compression process performed by the audio compression unit 101. (C) represents the compressed data after the audio compression processing.

FIG. 11 is a block diagram showing a conventional example of a multipoint videophone control device.

12 is a block diagram illustrating a detailed configuration of a control unit 516 in FIG.

[Explanation of symbols]

1, 108 Network interface unit 2-1, 2-2, 2-3, 2-4 Voice compression / expansion unit 3 Voice addition unit 4 Speaker discrimination unit 5 Image switching unit 6 Call control unit 10 Internet 40-1, 40 -2, 40-3, 40-4 Packet counting unit 41 Packet number comparison unit 100 Microphone 101 Audio compression unit 102 Video camera 103 Image compression unit 104 Image expansion unit 105 Display 106 Audio expansion unit 107 Speakers ADi-1, ADi-2 , ADi-3, ADi-4
Expanded voice packets ADo-1, ADo-2, ADo-3, ADo-4
Decompressed voice packets subjected to voice addition processing AEi-1, AEi-2, AEi-3, AEi-4
Compressed voice packet AEo-1, AEo-2, AEo-3, AEo-4
Compressed voice packet EP-1, EP-2, EP-3, EP-4 Video conference terminal MCU Multipoint video conference device ND-1, ND-2, ND-3, ND-4, ND-n
Network data NUM-1, NUM-2, NUM-3, NUM-4
Number of packets SEL Speaker terminal identification signal SW switch T1, T2,. . . T5 time timing Vi-1, Vi-2, Vi-3, Vi-4 image packet, compressed image packet Vo-1, Vo-2, Vo-3, Vo-4 selected image packet, compressed image packet

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification code FI Theme coat ゛ (Reference) H04M 11/00 302 G10L 9/00 NF Term (Reference) 5C064 AA02 AC04 AC06 AC12 AC17 AD01 AD08 AD13 AD14 5K015 AA10 AB01 AB02 AF05 JA10 JA11 5K030 HA08 HB01 HB02 HC01 JT01 JT04 LD08 5K101 KK04 LL02 MM04 NN06 NN07 NN18 NN25 SS07 TT03 UU19 UU20

Claims (6)

[Claims] 1. A method of transmitting a video packet to each video conference terminal from a multipoint video conference control device to which a video conference terminal that does not transmit audio packets during silence is connected. Detecting the speaker terminal by determining the number of audio packets per time, and transmitting the detected speaker terminal to all the video conference terminals connected to the multipoint video conference control device as the source video phone terminal. Delivering the selected image video packet to the multipoint video conference controller. 2. The step of detecting a speaker terminal includes the steps of: inputting a voice packet which has been received from a plurality of video conference terminals via the Internet and subjected to decompression processing; Counting the number of packets, outputting the identification signal for identifying the terminal that has received the most packets as the speaker terminal, and distributing the selected image video packet. 2. The multipoint video conference control device according to claim 1, wherein an image packet received from a terminal is input, and a selected image packet as a speaker terminal is switched according to the identification signal input at the same time and transmitted to each video conference terminal. Video packet transmission method. 3. The step of detecting a speaker terminal includes inputting voice packets that have been received from a plurality of video conference terminals via the Internet and subjected to decompression processing, and the unit time of each valid non-silent voice packet is And outputting the number of valid voice packets per unit time, performing comparison, and selecting the video conference terminal that has transmitted the maximum number of packets per time as the speaker terminal. Outputting a speaker terminal selection signal. 4. The method according to claim 2, further comprising: outputting a speaker terminal selection signal. 4. A multi-point video conference control device for holding a video conference connected via the Internet to a plurality of video conference terminals that do not transmit voice packets when there is no sound, wherein there is no silence received from each video conference terminal. A speaker discriminating unit for detecting a speaker terminal by judging the number of valid voice packets per time, and a speaker connection unit connected to the multipoint video conference control device using the detected speaker terminal as a source video conference terminal. An image switching unit that delivers a selected image video packet to a video conference terminal. 5. A speaker discriminating unit which receives voice packets received from a plurality of video conference terminals via the Internet and subjected to decompression processing, and calculates the number of valid voice packets which are not silence per unit time. Count, determine the terminal that has received the most packets, and output an identification signal that identifies the terminal as the speaker terminal. The image switching unit inputs image packets received from a plurality of video conference terminals via the Internet. 5. The multipoint video conference control device according to claim 4, wherein a selected image packet as a speaker terminal is switched and transmitted to each video conference terminal in accordance with the identification signal input at the same time. 6. A speaker counting unit provided corresponding to a plurality of video conference terminals, receiving voice packets transmitted from the corresponding video conference terminals, counting the number of packets per time and outputting the packets. And a packet for outputting a speaker terminal selection signal for selecting, as a speaker terminal, a video conference terminal that has transmitted and input the number of packets per time output by the packet counting unit, 6. The multipoint video conference control device according to claim 5, further comprising a number comparison unit.