JPH0685934A - Picture changeover system - Google Patents

Abstract

PURPOSE:To prevent the careless changeover of a picture during utterance and the changeover of a picture caused frequently by inhibiting the changeover to other picture till the end of utterance in the picture changeover resulting from detecting a voice signal. CONSTITUTION:The utterance in plural conference rooms 1a-1n connecting by communication lines 2a-2n is discriminated at a voice detection section 4 depending on a voice level, and an utterance party discrimination section 5 selects a specific conference room among conference rooms making utterance. A changeover control section 7 gives a command to a picture exchange section 6 to distribute a picture of a conference room to be selected to the conference rooms 1a-1n. When the conference room is once selected, an utterance discrimination section 5 keeps selecting a specific conference room till the end of utterance. Thus, it is prevented that the undesired picture changeover of the conference room while making utterance is made and the frequent picture changeover takes place.

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 for conducting a conference by voice and image between respective conference rooms provided at remote multiple points.

[0002]

2. Description of the Related Art Generally, in a multipoint video conference system,
The following are the main methods of selecting image data to be sent to each conference room.

1) Operator switching method 2) Automatic switching method In the operator switching method of 1), the operator station collectively determines transmission images during a conference and controls switching. In the automatic switching method of 2), the audio levels of the conference rooms are compared, the person with the highest level is determined as the speaker, and the image of the conference room is transmitted to each conference room.

Incidentally, as a related one, Japanese Patent Laid-Open No. 62-62
-49755 and 60-43954.

[0005]

In the conventional automatic switching method, the speaker level is determined by comparing the voice levels of the speakers, so that the voice level of another conference room is detected even if a speaker is already speaking. If you do, it will switch to another image. Therefore, each conference participant needs to pay attention to his / her own audio level so that the images are not inadvertently switched during the speech. In addition, when speaking at a plurality of points at the same time, the selected conference room is not specified, and the images frequently switch.

According to the present invention, when a speaker is detected in the determination of the speaker, the switching of the image is prevented until the utterance ends, and the conference participants automatically switch the image without being aware of their voice level. The purpose of the present invention is to provide a multipoint video conferencing system.

[0007]

The present invention comprises the following means.

A line control unit for connecting a communication line between the conference rooms, and a voice detection unit for determining that a voice signal transmitted from each conference room has reached a certain voice level or higher as a speaker. , A speaker determination unit that selects the conference room that was first determined as the speaker by the voice detection unit and determines that it is the only speaker until the speech in the conference room ends, and an image exchange unit according to the speaker determination content On the other hand, a switching control unit for instructing distribution of the speaker's image to each conference room is provided.

[0009]

As for the image selected by the speech from a certain conference room, the image of the speaker is ensured as long as the speech is continued, and the speaker can be spoken regardless of the voice level from another conference room during the speech. Since the determination unit does not determine the speaker, the image being spoken is not switched.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS In FIG. 1, a plurality of conference rooms 1a to 1n provided at remote multiple points are accommodated in a line control unit 3 via communication lines 2a to 2n, and the conferences extracted by the line control unit 3 are accommodated. The voice signals of the rooms 1a to 1n are compared with the reference voice level in the voice detection unit 4 to determine which of the conference rooms 1a to 1n is speaking. The conference room determined by the voice detection unit 4 to be the speaker is the speaker determination unit 5
In step 1, the first speaker is detected and the only speaker in the conference is determined. This determination continues as long as the voice detection unit 4 determines that the user is speaking. Further, when the speaker finishes his / her utterance and the voice detection unit 4 determines that the voice level has not reached a certain level, the speaker determination unit 5 is not notified as a speaker and the utterance is determined to be finished. After that, the speaker determination unit 5 determines a specific speaker according to the same procedure as described above by the notification from the voice detection unit 4.

The image signals extracted by the line control unit 3 are re-connected by the image exchange unit 6 and distributed to the conference rooms 1a to 1n via the line control unit 3.

Based on the information judged by the speaker judging section 5, the switching control section 7 tells the image exchanging section 6 about the conference rooms 1a ...
Instruct 1n to distribute the image of the conference room determined to be the only speaker.

As described above, according to the above-described embodiment, the image of the selected speaker is not switched unless the statement is completed.

[0014]

According to the present invention, the image of the speaker is not inadvertently switched by another utterance, and therefore the conference participants can speak without paying attention to their voice level. Further, even if a plurality of simultaneous speeches occur, the images of the selected conference room do not frequently switch.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

[Explanation of symbols]

 1a to 1n ... Conference room, 2a to 2n ... Communication line, 3 ... Line control unit, 4 ... Voice detection unit, 5 ... Speaker determination unit, 6 ... Image exchange unit, 7 ... Switching control unit.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Masato Josho 216 Totsuka-cho, Totsuka-ku, Yokohama-shi Hitachi Ltd. Information & Communication Division

Claims (1)

[Claims] 1. A video conferencing system in which conference rooms provided at multiple remote points are connected to each other by a communication line and a conference between the conference rooms is made by voice and video. The control unit, the image exchange unit that distributes images to each conference room, the voice detection unit that detects a certain voice level for each conference room to determine the conference room that is speaking, and the voice detection unit A speaker determination unit that determines a specific speaker from the conference room that is speaking based on the notification, and a switching control unit that controls the image exchanging unit according to the speaker determination content are provided. Image switching method.