JPH0715711A - Automatic speaker image pickup device - Google Patents

Abstract

PURPOSE:To prevent the unnecessary camera controlling or switching operations by inhibiting the decision of a speaker until a clock device decides that the speech decided by plural microphones is not detected for a fixed time or longer and also by inhibiting the operations of other speaker detecting devices. CONSTITUTION:A camera control part 105 reads the control information corresponding to the microphone identification signal received from a speaker deciding part 103 out of a control information memory 106 and sends this information to a camera part 107. The images and voices of the conference attendants are recorded at the part 107 based on the received information. Then a speaker is decided by the part 103, and the absence of a speaker is decided by the detecting result of a speaker detecting part 102 corresponding to a microphone 101 regardless of the results of other parts 102. Under such conditions, a timer 104 counts the time and no speaker is decided until a speaker is not detected any more for a fixed time or longer. Then a speaker is decided again when it is decided that no speech is detected for a fixed time or longer.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a speaker automatic photographing system for automatically photographing a speaker in a conference.

[0002]

2. Description of the Related Art In a conventional speaker automatic image pickup system, the pan / tilt / zoom of a camera is detected so as to detect the highest voice signal level input from a microphone and photograph the speaker corresponding to the microphone.・
The control of focus etc. was performed and the speaker was photographed automatically.

[0003]

However, in the above-mentioned conventional example, when a speaker receives a signal of a level higher than that of the speaker while another speaker is speaking, the camera captures a position corresponding to the microphone. Unfortunately, there was a problem that unnecessary images were taken.

The present invention has been made in order to solve the above problems, and an object thereof is to provide an automatic speaker photographing apparatus capable of preventing unnecessary camera control or switching operation.

[0005]

In order to achieve the above object, the speaker automatic photographing apparatus of the present invention is, in claim 1, characterized in that a plurality of voice inputs are provided for each of a plurality of conference participants. Means, a single or a plurality of image input means controllable to photograph any one or a plurality of the conference participants among the plurality of conference participants, and an audio signal input from the plurality of voice input means. A speaker detecting means for detecting the presence or absence of a speaker from the speaker, a speaker determining means for identifying a voice signal input means having a maximum input signal level among those judged to be present by the speaker detecting means, and the speaker And a prohibiting means for prohibiting the operation of the speaker determining means, and a control means for controlling the image inputting means based on the result of the speaker determining means. The automatic photographing device for persons, in which the determined speech input means prohibits the determination of the next speaker by the prohibiting means until it is judged by the time measuring means that the speech is not made within a certain time. The control means selects one of the plurality of image input means, and further, the control means includes means for notifying a control signal to the image input means. A plurality of voice input means provided for each of the conference participants, and a single or a plurality of input means controllable to photograph any one or a plurality of the conference participants among the plurality of conference participants. And a speaker detection unit that detects the presence or absence of a speaker from the audio signals input from the plurality of audio input units, and a maximum input signal level of those that are judged to be present by the speaker detection unit. Speaker determining means for identifying a certain audio signal inputting means, clocking means for timing the result of the speaker determining means, control means for controlling the image inputting means based on the result of the speaker determining means and the clocking means In the automatic speaker image pickup device provided with, the operation of the other speaker detection means is continued until the detection by the speaker determination means determined by the speaker determination means is eliminated by the time counting means for a predetermined time or more. The control means is one of a plurality of image input means.
Further, the control means includes means for notifying a control signal to the image input means.

[0006]

In the present invention, the determination operation of the speaker is prohibited or the other speaker is detected until the time-determining means determines that the speech from the voice input means determined by the speaker determining means is not present for a certain time or longer. By prohibiting the operation of the means, even if there is a large voice input from another microphone during a speech, the photographing position of the camera does not change until the speech ends.

[0007]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The details of the present invention will be described below based on the embodiments shown in the drawings.

FIG. 1 is a schematic block diagram of a first embodiment in which the present invention is used as a minutes recording system in which a speaker of conference attendees is photographed by a camera and its video and audio are recorded in a VTR. Yes, in FIG. 1, 101-1 to 10-3 are microphones, 102-1 to 10-3 are speaker detection units, 103 is a speaker determination unit, 104 is a timer, 105 is a camera control unit, 10
Reference numeral 6 is a control information memory for storing information for controlling the camera control unit according to the state of each microphone, 107 is a camera unit for photographing the attendees of the conference, and 108 is the addition of the audio signal input from each microphone. A mixer unit 109 is a VTR for recording a video signal from the camera unit 107 and an audio signal from the mixer unit 108.

The operation of FIG. 1 will be described below in order.
The audio signal input from each microphone 101-1 to 10-3
It is added in the mixer section of 8 and sent to the speaker detecting sections of 102-1 to 102-3. Speaker detection units 102-1 to 102-3
Then, the voice signal input from each microphone is converted into a level signal and compared with an arbitrary reference value to detect the presence or absence of a speaker and send the level of the input voice signal to the speaker determination unit 103. In the speaker determination unit 103, the speaker detection unit 1
When it is detected from the results of 02-1 to 3 that there is a speaker, the timer 104 measures the time of this detection result, and when it continues for a certain period of time or more, a microphone identification signal corresponding to it is sent to the camera control unit 105. send. Camera control unit 105
Then, the control information corresponding to the microphone identification signal received from the speaker determination unit 103 is read from the control information memory 106 that has been stored in advance, this information is sent to the camera unit 107, and the camera unit 107 follows the received information. Take pictures of meeting attendees. Then, the video and audio of the photographer are recorded on the VTR 109.

Next, the operation after the speaker is once determined will be described. When the speaker is determined by the speaker determination unit 103, the speaker detection unit 102 corresponding to the microphone 101 is irrespective of the result of the speaker detection unit 102 other than the determined speaker.
If there is no speaker from the detection result from the timer 10
In step 4, the time is counted, and the speaker is not determined until the speaker is no longer detected for a certain period of time. When it is determined that no speaker has been spoken for a certain period of time, the work of determining the speaker is restarted.

[0011]

Other Embodiments A second embodiment of the present invention will be described in detail with reference to FIG. Also in this embodiment, the minutes recording system for recording the video and audio of the speaker of the conference attendees on the VTR is taken up as the minutes. FIG. 2 is a schematic configuration diagram of the present invention. In FIG. 2, 201-1 to 201-3 are microphones and 202-
1 to 3 are speaker detectors, 203-1 to 20 3 are timers, 20
4 is a speaker determination unit, 205-1 to 205 are camera units, 206
Is a selector unit for selecting video signals from the camera units 205-1 to 205, 207 is a mixer unit for adding the audio signals from the microphones 201-1 to 201, and 208 is a selector unit and audio signals mixed by the mixer unit 207. A VTR for recording a video signal from the camera unit selected in 206.

Hereinafter, the operation of FIG. 2 will be described step by step.
The voice signal input from each microphone 201-1 to 20 is 20
It is added in the mixer section 7 and is sent to the speaker detection sections 202-1 to 202-3. Speaker detection units 202-1 to 202-3
Then, the audio signal input from each microphone is converted into a level signal and compared with an arbitrary reference value, and the timers 203-1 to 203-3 are used.
When the continuity of the values equal to or greater than the reference value is checked and it is determined that there is a message, the level of the input voice signal is sent to the speaker determination unit 204 together with the speaker detection signal. In the speaker determination unit 204,
Of the speakers detected from the results of the speaker detection units 202-1 to 202-3, the one having the highest level is set as the final speaker, and the image of the camera unit 205 capturing the microphone 201 corresponding to the speaker is detected. Selector unit 2 to select
A selection signal is sent to 06. The result determined by the speaker determination unit 204 is sent to the speaker detection unit 202, the undetermined speaker detection unit 202 prohibits the speaker detection operation, and the determined speaker detection unit 202 uses the timer. Using 203, the time when the input signal does not reach the reference value is measured, and when it does not reach the reference value continuously for a certain period of time, it is determined that the utterance has ended, and the speaker deciding unit 204 is notified of that fact. The speaker determination unit 204 that has received the notification of the end of the notification notifies the other speaker detection units of the cancellation of the determination, and each speaker detection unit 202 restarts the speech detection operation.

Although the embodiment relating to the conference minutes creating apparatus has been described above, it can be applied to a video conference system or the like.

[0014]

As described above, according to the present invention, the determining operation of the speaker is prohibited until the time measuring means determines that the speech from the voice input means determined by the speaker determining means is not present for a predetermined time or longer. , Or by prohibiting the operation of other speaker detection means, the shooting position of the camera does not change until the end of the statement even if there is a large voice input from another microphone during a statement. It is possible to prevent unnecessary camera control or switching operation.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of a main part of a first embodiment of the present invention.

FIG. 2 is a block diagram showing a schematic configuration of a main part of a second embodiment of the present invention.

[Explanation of symbols]

101-1, 101-2, 101-3, 201-1, 2
01-2, 202-3 microphone 102-1, 102-2, 102-3, 202-1, 2
02-2, 202-3 speaker detection unit 104, 203-1, 203-2, 203-3 timer 103, 204 speaker determination unit 105 camera control unit 106 control information memory 107 camera unit 108, 207 mixer 109, 208 VTR 205-1, 205-2, 205-3 Camera 206 Selector

Claims (6)

[Claims] 1. A plurality of audio input means provided for each of a plurality of conference participants, and controllable to photograph any one or a plurality of conference participants among the plurality of conference participants. A single or a plurality of image input means, a speaker detection means for detecting the presence / absence of a speaker from the audio signals input from the plurality of voice input means, and a speaker detection means for judging the presence of a speaker Speaker determination means for identifying a voice signal input means having a maximum input signal level, and clocking means for timing the result of the speaker determination means,
A speaker automatic image capturing apparatus comprising: a control unit for controlling the image input unit based on a result of the speaker determining unit; and a prohibiting unit for prohibiting an operation of the speaker determining unit, wherein a determined voice A speaker automatic image pickup device characterized in that the prohibition means prohibits the determination of the next speaker until the time counting means determines that the speech from the input means is less than a certain time. 2. The control means is one of a plurality of image input means.
2. The speaker automatic photographing device according to claim 1, wherein one of the two speakers is selected. 3. The automatic speaker imaging apparatus according to claim 1, wherein the control means includes means for notifying a control signal to the image input means. 4. A plurality of audio input means provided for each of a plurality of conference participants, and controllable to photograph any one or a plurality of conference participants among the plurality of conference participants. A single or a plurality of image input means, a speaker detection means for detecting the presence / absence of a speaker from the audio signals input from the plurality of voice input means, and a speaker detection means for judging the presence of a speaker Speaker determination means for identifying a voice signal input means having the maximum input signal level, timing means for timing the result of the speaker determination means, and image input based on the results of the speaker determination means and the timing means. A speaker automatic photographing apparatus provided with a control means for controlling the means, wherein it is determined that the detection by the speaker detection means determined by the speaker determination means has been lost for a predetermined time or more by the time counting means. The automatic speaker image pickup device is characterized in that the operation of other speaker detection means is prohibited until the time. 5. The control means is one of a plurality of image input means.
5. The speaker automatic photographing device according to claim 4, wherein one of the two is selected. 6. The automatic speaker imaging apparatus according to claim 4, wherein the control means includes means for notifying a control signal to the image input means.