JPH10227849A - Sound source position measuring device, camera image pick-up controller, sound source position recording device, and sound source position measurement method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a sound source position by utilizing a time difference at the time when a sound from the same sound source arrives at a plurality of microphones. SOLUTION: Microphones 1-3 are arranged at three vertices of a triangle in a specific shape. A sound signal detection means 4 detects the time when a sound generated by one sound source is captured by each of the microphones 1-3 as a sound detection time. A time difference measurement means 5 measures the arrival time difference of sound to each of three sets of microphone pairs obtained by combining each two of the microphones 1-3 based on the sound detection time for each microphone. A sound source position calculation means 6 calculates a sound source position from the positions of the microphones 1-3 and the arrival time difference of sound for each microphone pair. A camera control means 7 controls the direction and the amount of zoom of the camera so that the camera captures a sound source in a specific size based on the sound source position information that is calculated by the sound source position calculation means 6.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a sound source position measuring device for calculating a sound source position based on sound information from a sound source, a camera photographing control device for controlling a photographing position of a camera based on sound source position information, and an utterance time of a speaker. Sound source position recording device and sound source position measurement method for calculating sound source position based on sound information, in particular, sound source position measurement using sound time difference when sound from the same sound source reaches a plurality of microphones Device, a camera photographing control device that controls the photographing position of the camera based on the calculation result, a sound source position recording device that records speaker information based on the calculation result of the sound source position calculation device, and a plurality of sounds from the same sound source. And a sound source position measuring method for calculating a sound source position using a time difference when the sound reaches a microphone.

[0002]

2. Description of the Related Art In a video recording of a conference or a remote conference system, it is important to take an image of the entire conference or to zoom in on individual speakers. Thereby, the expression of the speaker and the atmosphere of the meeting can be better grasped, and the presence of the meeting can be enhanced. Moreover, it is desired that such shooting be performed automatically.

[0003] Therefore, as a conference system capable of photographing a speaker, Japanese Patent Application Laid-Open Nos. 4-122184 and 4-29.
An invention such as that described in Japanese Patent No. 7196 is made. The former is
A microphone is assigned to each sound source, that is, each speaker, and the speaker is detected by a signal from the microphone. The photographing of the speaker is performed by one camera. On the other hand, the latter invention employs a method in which a camera and a microphone are assigned to each speaker, and the utterance of each speaker is determined based on the voiceprint of each speaker. In addition, Japanese Unexamined Patent Publication No.
In Japanese Patent No. 217304, a transmitter for transmitting a radio signal unique to each speaker is attached to each microphone, and a coordinate position of the transmitter is obtained based on a reception signal measured by a receiver for receiving the radio signal. Has been adopted.

[0004] However, in these systems, microphones must be prepared for all the participants, so that complicated installation work is required. Further, since the camera shoots only at a preset location, It was about flexibility.

On the other hand, in the camera photographing control apparatus disclosed in Japanese Patent Application Laid-Open No. 7-140527, when two microphones capture sound from the same sound source, the sound phase difference between the microphones is used. By taking the method of knowing the direction of the camera, the camera is aimed at many sound source directions with a small number of microphones.

It is widely known that it is possible to know the direction of a sound source, that is, to perform sound source localization, by utilizing the time difference when sounds from the same sound source reach a plurality of microphones. In this case, with the two microphones, the symmetry with the plane perpendicular to the midpoint of the line connecting them remaining as the symmetry plane remains, so that it is impossible to determine the front and rear. For this reason, Japanese Patent Application Laid-Open No.
In the invention disclosed in Japanese Patent Publication No. 27, a microphone is attached to the side of the camera so that only the sound in front of the camera is captured.

In addition, if the information of the utterance time of each speaker can be obtained when recording a video of a conference, it is convenient for editing video later. Therefore, conventionally, the information of the utterance time has been obtained by detecting the voice input to the microphone assigned to each speaker.

[0008]

However, when a speaker is photographed by a camera, there is a problem that the direction obtained by the conventional method is only the direction of the sound source, and the position of the sound source has not yet been obtained. there were. That is, since the direction of the sound source to be obtained is based on the position of the microphone, the microphone and the camera must be integrated. As a result, the microphone and the camera cannot be separated, and there is a restriction that the microphone cannot be arranged at a position where it is easy to catch the utterance from each speaker. In addition, when photographing with a camera, the degree of zoom changes depending on the position from the camera to the sound source. Therefore, the zoom amount cannot be adjusted unless the distance from the sound source is known.

[0009] Further, when acquiring information on the utterance time,
Assigning a microphone to each speaker requires complicated work for installation of the microphone, routing of cables, and the like. Therefore, it is desired that the voice captured by the microphone and the speaker can be associated with a simpler configuration.

The present invention has been made in view of such a point, and a sound source position measuring apparatus capable of obtaining a sound source position by utilizing a time difference when sounds from the same sound source reach a plurality of microphones. The purpose is to provide.

Another object of the present invention is to provide a camera photographing control device which can point a camera toward a sound source even when a camera and a microphone are arranged at different positions.

Another object of the present invention is to provide a sound source position recording apparatus capable of identifying a speaker based on sound source position information and recording the utterance start time of each speaker. It is another object of the present invention to provide a sound source position measuring method for obtaining a sound source position by utilizing a time difference when sounds from the same sound source reach a plurality of microphones.

[0013]

According to the present invention, in order to solve the above-mentioned problems, in a sound source position measuring device for calculating a sound source position based on sound information, a microphone installed at three vertexes of a triangle having a predetermined shape; A sound signal detecting means for detecting a time at which a sound emitted from one sound source is captured by each of the microphones as a sound detection time, and two microphones each based on the sound detection time for each microphone. Time difference measuring means for measuring the arrival time difference of sound for each of the three microphone pairs, sound source position calculating means for calculating the sound source position from the position of each microphone and the sound arrival time difference for each microphone pair, A sound source position measuring device characterized by having:

According to such a sound source position measuring device, when a sound is emitted from a certain sound source, each microphone captures the sound at a time corresponding to a distance from the sound source. Next, the time when the microphone captures the sound is detected by the sound signal detecting means, and is set as the sound detection time. Then, based on the sound detection time, the time difference measuring means measures the arrival time difference of the sound for each microphone pair. Then, the sound source position is calculated by the sound source position calculation means based on the arrival time difference of the sound for each microphone pair.

In a camera photographing apparatus for photographing while controlling the direction of a camera installed on a turntable,
Microphones installed at three vertexes of a triangle having a predetermined shape, sound signal detecting means for detecting a time when a sound emitted from one sound source is captured by each of the microphones as sound detection time, Time difference measuring means for measuring the arrival time difference of the sound for each of the three microphone pairs obtained by combining two of the microphones based on the sound detection time; Sound source position calculating means for calculating a relative position of a sound source with respect to the camera based on the arrival time difference and information on the position of the camera; and a camera moving in the direction of the sound source based on the position information of the sound source calculated by the sound source position calculating means. And a camera control means for controlling the direction of the camera.

According to such a camera photographing apparatus, when a sound is emitted from a certain sound source, each microphone captures the sound at a time corresponding to a distance from the sound source. The time at which the microphone captures the sound is defined as the sound detection time by the sound signal detection means, and the time difference measurement means measures the arrival time difference of the sound for each microphone pair based on the sound detection time. The sound source position is calculated by the sound source position calculating means based on Then, the camera control means controls the direction of the camera toward the sound source. As a result, the camera can always be pointed at the speaker.

In a sound source position recording apparatus for recording a sound source position based on sound source position information based on sound information,
Microphones installed at three vertexes of a triangle having a predetermined shape, sound signal detecting means for detecting a time when a sound emitted from one sound source is captured by each of the microphones as sound detection time, Time difference measuring means for measuring the arrival time difference of sound for each of three microphone pairs obtained by combining two of the microphones based on the sound detection time, and the position of each microphone and the sound of each microphone pair From the arrival time difference, the sound source position calculating means for calculating the sound source position, and if the sound source position calculated by the sound source position calculating means is within a predetermined range, the sound is emitted from the same sound source, and the sound outside the predetermined range And a utterance information recording means for recording an utterance time for each sound source by using sounds uttered from different sound sources in some cases. That.

According to such utterance information recording means, when a sound is emitted from a certain sound source, each microphone captures the sound at a time corresponding to the distance from the sound source. The time at which the microphone captures the sound is taken as the sound detection time by the sound signal detection means,
The time difference measuring means measures the arrival time difference of the sound for each microphone pair based on the sound detection time, and the sound source position calculating means calculates the sound source position based on the sound arrival time difference for each microphone pair. Then, based on the calculation result of the sound source position calculating means, the utterance information recording means records the utterance time for each sound source position. Thereby, it is possible to sequentially record at what place and at what time a person who was speaking.

In the sound source position measuring method for calculating a sound source position based on sound information, a time at which a sound emitted from one sound source is captured by each of microphones installed at three vertexes of a triangle having a predetermined shape is determined. Detected as sound detection time,
Based on the sound detection time of each microphone, the arrival time difference of the sound of each of the three microphone pairs obtained by combining two of the microphones is measured, and the position of each microphone and the sound of each microphone pair are measured. From the arrival time difference,
A sound source position measurement method is provided, which calculates a sound source position.

According to such a sound source position measuring method, 1
From the sound detection time at which the sounds emitted from the two sound sources are captured by the microphones arranged at the vertices of the triangle, the arrival time difference of the sound for each microphone pair is measured, and the sound source position is calculated based on the arrival time difference.

[0021]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a principle configuration diagram of a camera photographing control device of the present invention. First, microphones 1 to 3 are arranged at three vertices of a triangle having a predetermined shape. The sound signal detecting means 4 detects the time at which the sound emitted from one sound source is captured by each of the microphones 1 to 3 as a sound detection time, and reproduces the audio signal captured by the microphones 1 to 3 for video / audio recording / reproduction. Send to means 12. The time difference measuring means 5 measures the arrival time difference of sound for each of the three microphone pairs obtained by combining two microphones 1 to 3 based on the sound detection time for each microphone. The sound source position calculation means 6 calculates a sound source position from the positions of the microphones 1 to 3 and the arrival time difference of sound for each microphone pair. Information on the calculated sound source position is stored in the camera control unit 7 and the utterance information recording unit 1.
Passed to one.

The camera control means 7 controls a camera driving means 8 for rotating a turntable on which the camera body 9 is mounted, and an electric lens 10 for adjusting the zoom amount of the camera body 9. Then, the camera control means 7 controls the direction and the zoom amount of the camera based on the sound source position information so that the camera captures the sound source at a predetermined size. The video signal captured by the camera body 9 is transmitted to the video / audio recording / reproducing means 1.
Sent to 2.

The utterance information recording means 11 records sound source position information when the microphone captures sound and utterance time. The video / audio recording / reproducing means 12 records and reproduces the video information sent from the camera body 9. The reproduction instructing means 13 gives a reproduction instruction to the video / audio recording / reproducing means 12 using the information stored in the speech information recording means 11.

According to such a camera photographing control device,
When a person speaks in a certain place, the voice is captured by the three microphones 1 to 3. Then, the time when each of the microphones 1 to 3 captures the sound is detected by the sound signal detecting means 4, and is set as the sound detection time. At this time, due to the difference in the distance from the position where the speaker is present to each of the microphones 1 to 3, there is a time difference in the time at which each of the microphones 1 to 3 captures sound. Therefore, three microphone pairs are created by combining two microphones 1 to 3 by the time difference measuring means 5, and the arrival time difference of the sound for each microphone pair is measured. Then, the sound source position is calculated by the sound source position calculating means 6 using the arrival time difference of the sound of each microphone pair. At this time, the sound source position in the coordinate system based on the position of the camera body 9 is calculated.

When the sound source position is determined, the direction and the distance of the sound source viewed from the camera body 9 are calculated by the camera control means 7, and the camera body 9 is directed to the sound source by controlling the camera driving means 8. With the motorized lens 1
By controlling 0, the zoom amount is adjusted.

On the other hand, the audio information detected by the audio signal detecting means 4 and the video information photographed by the camera body 9 are recorded in the video / audio recording / reproducing means 12, and the utterance time for each sound source is recorded in the utterance information recording means. 11 is stored. And
The operator of this apparatus can know the utterance time of an arbitrary speaker based on the information stored in the utterance information recording means 11 and only the utterance of a specific speaker is recorded in the video / audio recording / reproducing means 1.
2 and can be reproduced.

In this way, since the position of the sound source can be measured with only three microphones, the camera can be pointed at the sound source even when the microphone is separated from the camera, and the distance between the camera and the sound source can be adjusted according to the distance from the camera to the sound source. You can also adjust the amount of zoom. Furthermore, since the sound source position and the utterance time can be recorded in association with each other, the utterance time of each speaker is associated with the information recorded in the video / audio recording / reproducing means 12 so that the video index can be obtained. Can be created.

By the way, the camera photographing control device of the present invention can be effectively used as a video conference system and video photographing of important conferences. Therefore, an embodiment will be described by taking as an example a case where the contents of a meeting are photographed using the camera photographing control device of the present invention.

FIG. 2 is a layout diagram of the camera photographing control device according to the first embodiment. In this example, a plurality of participants 22a to 22k arrive at a U-shaped table 23 facing the moderator 21. In the center of the conference room, three microphones 1 to 3 arranged at the vertices of an equilateral triangle are placed. These microphones 1 to 3 are connected to the device main body 100 of the camera photographing control device. In the device body 100,
The camera 200 is connected.

When the microphones 1 to 3 capture the voice of the speaker, the sound source position is calculated by the apparatus main body 100, the camera is turned in the direction of the speaker, and the camera is zoomed in according to the distance from the camera to the speaker. And zoom out.

The length of the sides of the equilateral triangle formed by the three microphones 1 to 3, ie, the distance between the microphones, can be arbitrarily set. Here, the microphones are easily installed in a meeting place, and the speed of sound at room temperature is easy. 68cm which is easy to calculate from the value 340m / s
And

FIG. 3 is an internal configuration diagram of the camera photographing control device according to the first embodiment. Three microphones 1 to 3 arranged at the vertices of an equilateral triangle, which are voice input means, are connected to the sound signal detector 110. Sound signal detector 110
Outputs the sound captured by the microphones 1 to 3 as an audio signal and outputs a signal corresponding to the time at which each of the microphones 1 to 3 captures the sound. The signal is input to the time difference measuring device 120. The time difference measuring device 120 calculates a time difference between times when the microphones 1 to 3 capture the sound, and outputs time difference data. The time difference data is input to the sound source position calculator 130. The sound source position calculator 130 calculates and outputs a sound source position from the time difference data.

The reference signal generator 101 generates a reference clock for measuring the time difference between the times when each microphone captures a sound from the same sound source, and uses the reference clock as the time difference measuring device 1
20 and a sound source position calculator 130. Here, a square wave having a period of 0.5 μs is generated. The resolution of the time difference measurement of the sound is determined by this cycle. In this embodiment, since the distance between the microphones is 68 cm, the maximum time difference between the microphones is a value obtained by dividing 68 cm by the speed of sound (340 m / s), that is, 2 ms. Since the value of 0.5 μs is 1/4000 of this value, the resolution is sufficient for speaker position calculation in a conference.

The camera controller 140 receives the sound source position data from the sound source position calculator 130 and
3 and the zoom amount are controlled. Camera body 203
Is fixed on a turntable 201, and the direction of the camera body 203 can be changed by rotating the turntable 201. Further, the camera body 203 includes an electric lens 202 for changing a shooting range, and the zoom amount can be adjusted by adjusting the electric lens 202. The image captured by the camera body 203 is output as a video output 103.

According to the camera photographing control device having such a configuration, the moderator 21 or the participants 22a to 22k in FIG.
, The sound is captured by the three microphones 1 to 3 and input to the sound signal detector 110. Hereinafter, the sound signal detector 110 will be described in detail.

FIG. 4 is a diagram showing the internal configuration of the sound signal detector 110. The sound signal detector 110 basically includes amplifiers 111a to 111c and signal / power converters 112a to 112a.
12c, the level determination units 113a to 113c, and the differentiators 114a to 114c are configured as three sets, each of which is one set.

Signals 103a to 103a from the microphones 1 to 3
3c is input to amplifiers 111a to 111c that amplify the signal. Outputs of the amplifiers 111a to 111c are output from signal / power converters 112a to 112
The signal is input to the mixer 115 as well as to the mixer 115. The mixer 115 converts the audio signal amplified by the amplifier into a stereo signal and outputs the stereo signal as the audio output 102.

Each signal / power converter 112a-112c
Calculates the power of the sound signal from the output signals of the amplifiers 111a to 111c, and inputs the power signal to the level determiners 113a to 113c provided correspondingly. Each of the level determiners 113a to 113c binarizes the power signal with a specific threshold value, and then converts the binarized signal into a TTL (Transistor-T).
ransistor-Logic) level signals are input to differentiators 114a to 114c provided corresponding to the respective signals. Each of the differentiators 114a to 114c detects a rising edge of the output of the input signal, and a trigger signal 104 indicating the rising edge.
a to 104c are output.

Before the outputs of the differentiators 114a to 114c, only positive differential results are output by the action of diodes. In addition, the amplifiers 111a to 11a
The amplification degree of 1c and the threshold values of the level determiners 113a to 113c can be arbitrarily set for each set by the volume.

The change of the signal in the sound signal detector 110 will be described below by taking the signal 103a from the microphone 1 as an example. FIG. 5 is a diagram illustrating a change in a signal in the sound signal detector 110. In this figure, the horizontal axis represents time, and the vertical axis represents voltage.

FIG. 3A is a diagram showing a waveform of a signal input from the microphone 1. A signal 103a having a relatively weak waveform is input from the microphone 1. This signal 103a is amplified by the amplifier 111a.

FIG. 7B is a diagram showing the waveform of the signal amplified by the amplifier 111a. By being amplified by the amplifier 111a, the waveform becomes larger in amplitude than the waveform shown in FIG. This signal is supplied to the signal / voltage converter 112.
The signal is converted into a power signal by a.

(C) is a diagram showing a waveform of the power signal. The power signal is a signal having only a positive value. The level of this signal is determined by the threshold value 116 in the level discriminator 113a.

(D) is a diagram showing the waveform of the signal after the level determination. A binary signal is obtained by the level determination. When this signal passes through the differentiator 114a, it becomes a trigger-like signal.

(E) is a diagram showing a waveform of a signal output from the differentiator 114a. The waveform of the trigger signal output from the differentiator 114a indicates the rising time of the signal input to the differentiator 114a.

In this way, from the sound captured by the microphone 1, a trigger signal-like waveform indicating the time when the sound is captured is obtained. This process is similarly performed on the sounds captured by the other two microphones 2 and 3, and the trigger signals 104 b to 104 c corresponding to the respective microphones 2 and 3 are output signals of the sound signal detector 110. .

The three trigger signals 104a to 104c output from the respective sound signal detectors 110 correspond to the reference signal generator 1
The signal is input to the time difference measuring device 120 together with the clock signal from 01.

FIG. 6 is a diagram showing the internal configuration of the time difference measuring device 120. The time difference measuring device 120 includes three trigger signal converters 121a to 121c and three delay devices 122a.
To 122c. Trigger signal converter 121
a to 121c are input trigger signals 104a to 104c
4c is converted into a square wave having the same time width as the H (high) level of the square wave generated by the reference signal generator 101. Further, each of the delay units 122a to 122c measures a time difference between two square waves.

According to the time difference measuring device 120 having such a configuration, the input trigger signals 104a to 104c are converted by the trigger signal converters 121a to 121c into squares having the same time width as the H level of the square wave as the reference signal. Converted to waves.

FIG. 7 is a diagram showing a square wave obtained from the trigger signal. (A) represents a trigger signal,
(B) represents a square wave. That is, if there is a trigger signal 104a 160.8 μs after a certain time, the trigger signal 104a is set to the next reference clock H
161.0μs to 161.25μ
The signal is converted into a square wave 120a having an H level during s.

The signal quantized in the cycle of the reference clock is then paired with two signals, and the delay unit 122a
To 122c. That is, three pairs of a signal pair of the microphone 1 and the microphone 2, a signal pair of the microphone 2 and the microphone 3, and a signal pair of the microphone 3 and the microphone 1 are input to the delay units 122a to 122c.

FIG. 8 is a diagram showing the internal configuration of the delay unit 122a. The delay unit 122a can be actually constituted by a delay circuit. In this example, the order determination unit 122aa, the counter 122ab, and the data It is constituted by a converter 122ac.

The order determiner 122aa determines which input has come first.
3a, a signal 123b from the microphone 2, and a clear signal 122ad. And the order determiner 1
22aa outputs order determination bits 122ae and 122af, which are outputs for determining the order, to the data converter 122a.
c and outputs a signal 122ag for time difference measurement to the counter 122ab.

The counter 122ab includes the order judgment unit 12
In addition to the output 122ag of 2aa, the reference signal generator 101
From the clock signal 122ah. The counter 122ab counts the number of H-levels of the input clock signal 122ah only while the signal 122ag output from the sequence determiner 122aa is at H-level, and converts the count output 122ai, which is the counting result, into data conversion. Is output to the detector 122ac. Also, counter 1
22ab also outputs a clear signal 122ad to the order determiner 122aa.

The data converter 122ac includes the order determiner 1
22aa, two order determination bits 122ae and 122a
f and a 2-byte unsigned binary counter 122ab
Is converted into a 2-byte signed binary number. Then, the conversion result is output data 1
05a and the data conversion end signal 106a is set to 1
Output as the H level of the pulse.

The delay unit 122a having such a configuration operates as follows. When the two signals 123a and 123b are input, the order determiner 122aa determines which signal is input earlier. If the input of the signal 123a becomes H level earlier, the order determination bit 122a
e is kept at the H level, and conversely, if the input of the signal 123b goes to the H level earlier, the order determination bit 122af becomes H level.
Kept at the level. The signal 122ag output to the counter 122ab is composed of two signals 123a and 123a.
The H level is maintained from the time when either of the b inputs becomes H level until the other signal becomes H level. The state of these order determiners 122aa is determined by the clear signal 122ad.
Is cleared to H level. Therefore, the order determination bit 122ae and the order determination bit 1
If the output of 22af is (1, 0) in this order, the signal 12
If the input of 3a is fast and (0, 1), the signal 123b
Can be determined to be quick.

When the signal 122ag becomes H level, the counter 122ab
The counting of the number of H levels of ah is started. Signal 122a
When g becomes L (low) level, the counter 122ab stops counting, and counts the counted information as 16-bit unsigned 2 bits.
The count output 122ai is represented as a base number. Accordingly, a value obtained by multiplying the clock cycle of 0.5 μs by the result of the counting is a measured time difference.
Is used as it is. Counter 122ab
Sends out a one-pulse H-level signal to the clear signal 122ad at the same time that the count output 122ai is output. By this clear signal 122ad, the order determining unit 1
The state of 22aa is cleared to prepare for the next input.

Two order determination bits 122ae, 122
af output and 2-byte unsigned binary counter 12
The 2ab count output 122ai is a data converter 122
The data is converted into a 2-byte signed binary number by ac and becomes output data 105a. When the data conversion is completed, the data conversion end signal 106a is output as one pulse H level. Output data 105a and data conversion end signal 10
6a is sent to the next sound source position calculator 130. The data converters of the three delay units 122a to 122c change the order of the microphone pairs to (1, 2), (2,
3) and (3, 1). If the sound of the microphone in the first row is earlier than the sign, the sign is set to plus, and if the opposite, the sign is set to minus.

Before describing the details of the function of the sound source position calculator 130, a method of calculating the sound source position will be described. The basis of the sound source position calculation is to calculate the sound source direction in the two microphones.

FIG. 9 is a diagram showing a microphone coordinate system for calculating the sound source direction. As shown in the figure, a uv coordinate system in which a straight line passing through the two microphones 1 and 2 is a u-axis and a midpoint of the two microphones 1 and 2 is an origin is considered. The microphone 1 is located at a position (-a, 0) away from the origin by -a, and a
Microphone 2 is located at a position (a, 0)
Assuming that the position of the sound source P is at a position (r, Θ) from the origin, the time difference dt when the sound from the sound source P reaches each microphone is:

[0061]

(Equation 1) Becomes However, V ₀ is the value of sound speed (340 m /
s). This is plotted using r as a parameter.

FIG. 10 is a graph showing a relational expression between a time difference between microphones and a sound source direction. Here, a = 1
m. As described above, the effect of r on dt is sufficiently smaller than 、, and the angle error is about 5 degrees even if the value of r is changed, and Θ becomes smaller than about 25 degrees at any r. It can be seen that the angle sensitivity to dt is higher in the region than in the previous regions. From this, knowing the arrival time difference of the sound to the microphone pair, the direction of the sound source of the sound, ie, u
A straight line connecting the origin of the -v coordinate and the sound source can be obtained. If another straight line is obtained by another microphone pair, the sound source position is obtained from the intersection of those straight lines.

In the present embodiment, since the position of the sound source is calculated by the combination of the microphone pairs, ± 60 degrees from the X axis are calculated.
The sound source position calculation is performed using the range of degrees as the measurement range of each microphone pair.

FIG. 11 is a diagram showing a measurement range of one microphone pair. Assuming that the two microphones 1 and 2 are arranged on the u axis with the origin therebetween, the measurement range of this microphone pair is ± 60 in the positive direction about the u axis as viewed from the origin.
°, ± 60 ° in the negative direction. The measurement range is thus determined, and the sound source position is calculated using two microphone pairs whose sound source positions are within the measurement range among the three microphone pairs. Since the three microphones are arranged at the vertices of an equilateral triangle, two microphone pairs always fall within this range for an arbitrary sound source position. Therefore, the data used for the sound source position calculation is obtained by excluding the data with the smallest time difference from the three time difference count data.

Therefore, in calculating the sound source position, first, the sound source direction is obtained by individually using the coordinate system shown in FIG. 11 for each microphone pair. The sound source direction is obtained by comparing the measured time difference data with the angle-time difference information table.

FIG. 12 shows an angle-time difference information table. In this example, since the distance between the microphones is 68 cm, the value of r is set to 3 m in the table in the figure, and in the above equation, the time difference information of the two microphones is input every 5 degrees from 0 to 60 degrees. I have. In the first column of the table, angles from 0 degree to 60 degrees are given in order from the top from 5 degrees, and the second column stores time difference data of the two microphones at that angle. . The time difference data is calculated by using the time difference calculated by the above equation as a reference clock cycle of 0.5 μs.
Is a 2-byte signed integer representation of the value divided by.
Thus, comparison with the time difference data obtained by the time difference measuring device 120 can be directly performed. To find the direction from the time difference data, refer to the second column of this table. If the maximum cell whose absolute value of the time difference data does not exceed the value in the table is found, the angle data is obtained from the first column of the table. Obtainable. At this time, it is symmetrical on both sides of the v-axis in FIG. 11, but if the value of the time difference data is minus, the utterance position is on the microphone 1 side, and if it is plus, the utterance position is on the microphone 2 side. Become.

The angle data obtained here is 2
The axis connecting two microphones (the u-axis in FIG. 11) is a symmetric axis, and two straight lines on both sides from the axis are established as solutions. This symmetry problem can be solved by time difference data and microphone pair combination information. it can.

FIG. 13 is a diagram for explaining a method for obtaining a sound source position from two pieces of time difference information. In this example,
The sound source position is obtained from the microphone pair of the microphones 3 and 1 and the microphone pair of the microphones 2 and 3. The xy coordinates are defined such that the center of gravity of the equilateral triangle having the microphones 1 to 3 as vertices (the intersection of the line segment connecting each vertex and the midpoint of the opposite side) becomes the origin. Two straight lines 31 from the microphone pair of microphones 3 and 1
a and 31b are obtained, and two straight lines 32a and 32b are also obtained from the microphone pairs of the microphones 2 and 3. Also, since the sound from the sound source P reaches the microphones 1 and 2 faster than the sound reaching the microphone 3, positive angle information from the pair of the microphone 2 and the microphone 3 (indicating that the sound source is on the microphone 2 side). , And negative angle information (indicating that there is a sound source on the microphone 1 side) is obtained from the pair of the microphone 3 and the microphone 1. Moreover, as shown in FIG. 11, the measurement range of each microphone pair is limited. Therefore,
The range of the sound source can be limited by the combination of the time difference information. The sound source position in the example of FIG. 13 is limited to a range in which the angle from the x-axis of the first quadrant is 0 to 60 degrees. That is,
There is no possibility of the straight line 31b and the straight line 32b, and eventually two straight lines 31a and 32a remain.

As described above, the range of the sound source position is specified by the combination of the microphone pairs and the sign of the time difference data in each microphone pair. All combinations are shown below.

FIG. 14 is a diagram showing the range of the sound source determined by the combination of the microphone pair and the sign of the time difference data. In the figure, 1, 2, and 3 indicate the positions of the microphones 1, 2, and 3, respectively. Further, the time difference data of the microphone pair of the microphones 1 and 2 is (1, 2), the time difference data of the microphone pair of the microphones 2 and 3 is (2, 3), and the time difference data of the microphone pair of the microphones 3 and 1 is (3, 1). I have. Therefore, the relationship between the code of the time difference data of the microphone pair and the range of the sound source is as follows. The angle in the following description is represented by an angle (counterclockwise) from the positive x-axis.

If (2,3)> 0 and (3,1) <0, the sound source is limited to an angle from the positive x-axis within the range of 0 to 60 degrees. If (1,2)> 0 and (3,1) <0, the sound source is limited to an angle from the positive x-axis within the range of 60 to 120 degrees.

When (1,2)> 0 and (2,3) <0, the angle of the sound source from the positive x-axis is limited to the range of 120 to 180 degrees. If (2,3) <0, (3,1)> 0, the angle of the sound source from the positive x-axis is limited to a range of 180 to 240 degrees.

If (1,2) <0, (3,1)> 0, the angle of the sound source from the positive x-axis is limited to the range of 240 to 300 degrees. If (1,2) <0, (2,3)> 0, the sound source is limited to an angle from the positive x-axis within a range of 300 to 360 degrees.

The relationship between the code of the time difference data of the microphone pair and the range of the sound source shown in this figure is stored in the sound source position calculator 130 in advance as a sound source position area table. If the two straight lines specified as described above are coordinate-transformed from the uv coordinate system in FIG. 11 to the xy coordinate system as shown in FIG. 13, the sound source position P (x, y) can be obtained from the intersection thereof. Desired. In the example of FIG. 13, in the case of the pair of microphones 3, 1, the point passes through the point (−a / 2, a / 2√3), the x axis and “60 degrees + the uv coordinate system obtained by the above method”. , And in the case of a pair of microphones 2 and 3, through the point (0, −a / 2√3) and the x axis and the uv coordinate system obtained by the above method. It is sufficient to obtain the straight line 32a forming the angle of “the angle of” and obtain the intersection of these straight lines.

The calculation of the position is performed by the sound source position calculator 130 using three pieces of signed binary information obtained by the time difference measuring device 120. The sound source position calculator 130 is constructed on a computer.

FIG. 15 is a diagram showing the internal configuration of the sound source position measuring device. As shown in FIG.
Are configured around a CPU (Central Processing Unit) 133.

The three ports 131a to 131c receive the data measured by the time difference measuring
33. The AND gate 132 outputs the data to the CP
The timing to take in is given to U133. CPU 133
Has registers 133a to 133c for storing data from the ports 131a to 131c. The CPU 133 performs data transfer with various peripheral devices via the bus 134.

The bus 134 includes an input device 135, an output port 136, a memory 137, and a hard disk 13
8 are connected. The input device 135 is for inputting the positional relationship between the camera and the microphone. The output port 136 is a communication port for outputting a calculation result to the camera control device 140. The memory 137 is a CPU 1
33 is for temporarily holding a program to be executed. The hard disk 138 holds a program, an angle-time difference information table, a sound source position area table, and a coordinate conversion table. The sound source position calculation process performed by the sound source position calculator 130 is a process executed by loading a program stored in the hard disk 138 into the memory 137 and executing the program by the CPU 133.

The three data outputs 105 from the time difference measuring device 120 input to the sound source position calculator 130
a to 105c are three ports 131a to 131c
Output from the time difference measuring device 120
106c are all at H level and AND gate 132
At the time when the output of the CPU 133 becomes H level, the data is taken into three registers 133a to 133c in the CPU 133. 3
Of the three time difference data taken into the three registers 133a to 133c, two time difference data excluding the data with the smallest time difference are written to variable areas t1 and t2 previously secured on the memory. Then, the above-described calculation of the sound source position is performed based on the two time difference data. Thereby, the sound source position in the xy coordinate system of FIG. 13 is obtained. Thereafter, coordinate conversion is performed to an XY coordinate system with the camera position as the origin, and a sound source position P (X, Y) in the XY coordinate system is obtained. FIG. 16 is a diagram illustrating the relationship between the coordinate system of the microphone and the coordinate system of the camera. The center of the microphones 1 to 3 (the center of gravity of the triangle) is the origin of the xy coordinates, and the rotation center of the camera 200 (shown in FIG. 2) is XY.
This is the origin of the coordinates.

Actually, the microphones 1 to 3 and the camera 200 are installed such that the x-axis parallel to the line connecting the microphones 2 and 3 and the direction of 0 degrees of the turntable 201 are parallel.
Therefore, the coordinate conversion at this time is only translation. The data of the amount of parallel movement is measured at the time of installation of the device, and is input to the sound source position calculator 130 by the input device 135. The sound source position calculator 130 performs coordinate conversion using this data. The coordinate conversion is not limited to translation, and even when both coordinate systems form a certain angle, it is only necessary to perform coordinate conversion by this rotation. However, in that case, xy
It is necessary to measure the angle between the coordinate system and the XY coordinate system. The sound source position P (X, Y) obtained in this way
Is output port 13 as 2-byte floating point data.
6 and input to the camera control device 140.

FIG. 17 is a diagram showing the internal configuration of the camera control device 140. As shown in this figure, the camera control device 140 includes an angle calculator 141, a zoom amount calculator 14
2, data output unit 143, calibration unit 14
4, a volume 145, and a calibration setting button 146.

The angle calculator 141 stores the sound source position data P
(X, Y) and signal 204 indicating operation of turntable 201
Is entered. The angle calculator 141 calculates the signal 204
Is at the L level, the sound source position data P (X, Y) is converted into polar coordinate system data (R, Θ). R is the distance from the camera to the sound source position, and Θ is the angle formed by the line connecting the sound source position P and the origin of the XY coordinate system with the X axis. In addition, Θ
Is converted into the data format of the rotary encoder and sent to the data output unit 143. Here, an integer value obtained by dividing the angle で by 1.4 is converted into an 8-bit unsigned binary number. The value of 1.4 is obtained by dividing 360 degrees by the maximum value 256 of 8-bit data.
The distance R to the sound source calculated by the angle calculation unit 141 is sent to the zoom amount calculation unit 142 and the calibration unit 144. The calibration unit 144 sets two pieces of reference information, that is, a reference distance and a value of the reference zoom amount, in the zoom amount calculation unit 142 because the zoom amount of the camera also changes when the distance between the camera and the person changes. It is provided for. The calibration section 144 is connected with a volume 145 for manually setting the zoom amount and a calibration setting button 146. The calibration unit 144 uses the distance data output by the angle calculation unit 141 and sets the zoom amount at that time using the volume 145. When the volume 145 is turned, the motorized lens of the camera 200 is driven, and the zoom amount is changed. By operating the volume 145 and pressing the calibration setting button 146 at a time when a proper zoom amount is obtained, the distance data Ro at that time is pressed.
And the zoom amount data Zo are output to the zoom amount calculating unit 142 as reference information. Upon receiving these data, the zoom amount calculation unit 142 stores the values in a memory and uses them in subsequent calculations of the zoom amount. However, in the case where the photographing target is a person as in the present apparatus, the size of the photographing target can be considered to be fixed, so even if the default values of the above two reference information are set in advance. Good.

The distance R to the sound source calculated by the angle calculator 141 is sent to the zoom calculator 142, and the zoom is calculated. The calculation of the zoom amount is performed by dividing the distance R by the reference distance Ro and multiplying the value by the reference zoom amount Zo. The calculated angle data and zoom amount are sent to the turntable 201 and the motorized lens 202 via the data output unit 143, respectively, and the camera photographing device is driven.
At this time, a signal 204 indicating the operation of the turntable 201 is sent from the turntable 201 to the camera control device 140 so that the next imaging position information is not output before the operation of the turntable 201 ends. If this signal is at the H level, it indicates that the turntable 201 is in operation, and during that time, the calculation by the camera control device 140 is not performed. In particular,
This signal is sent to the angle calculator 141, and while this signal is at the H level, the data sent from the sound source position calculator 130 is not taken in.

The conference in which the camera shooting is performed can be recorded through the video output 103 and the audio output 102 or transmitted to a remote place. At this time, it is not necessary to prepare microphones for all the participants in each conference hall, and it is not necessary that the three microphones and the camera are in the same place. Therefore, the microphone can be placed at a position where it is easy to catch the conversation of all, and the camera can be placed at a position where it is easy to catch the expression of all.

In the above description, the sound source position is calculated by utilizing the fact that the relationship between the arrival time difference between the sounds captured by the two microphones and the sound source direction does not depend on the distance to the sound source. Third, the sound source position can be calculated geometrically from the time difference between the sounds captured by the three microphones. Hereinafter, the calculation method will be described.

First, for simplicity, three microphones 1 to 3 are installed at three locations (0, a), (0, 0), and (a, 0) in an xy coordinate system. And Sound source P
The sounds emitted from (x, y) are the respective microphones 1 to 3
Are reached as t1, t2, and t3, respectively. Assuming that the distance between the sound source P and each microphone is l1, l2, l3,

[0087]

(Equation 2) It is. The time difference between each microphone, i.e., t1-t2 is dt1, t2
Assuming that −t3 is dt2, t3−t1 is dt3, and the value of the sound speed is vs, the arrival time difference of the sound between the microphones is given as follows according to the distance from the sound source to each microphone.

[0088]

(Equation 3) From these equations (2) and (3),

[0089]

[Number 4] ^{dt1 · vs = x 2 + (} a-y) 2 1/2 - (x 2 + y 2) 1/2 dt2 · vs = (x 2 + y 2) 1/2 - (a-x) 2 + Y ^{2 1/2} dt3 · vs = (a−x) ² + y ^{2 1/2} −x ² + (a−y) ^{2 1/2} ... (4) (X, y)
Can be requested. Such a method of obtaining the sound source position is similar to the principle of triangulation and is widely known. This calculation method is used, or the first
Whether the calculation is performed by the method used in the embodiment is determined based on the performance of the sound source position calculator and the required resolution of the time difference measurement of the sound. For example, when a high resolution is required, the method described in the first embodiment, which can be operated at high speed with a simple calculation, is used.
It is possible to use a geometrical sound source position calculation method such as an angle-time difference information table, which does not require advance preparation.

Further, in the first embodiment, the system is such that the height of the microphone from the floor is the same as the height of the camera from the floor. Is also possible. In this case, a value obtained by subtracting the height of the camera from the floor from the height of the microphone from the floor is set as an offset amount d in the Z-axis direction, and the vertical camera direction is calculated in the calculation in the camera control device. Good. However, in the upper coordinate system, the vertically downward direction is the positive direction of the Z axis. That is, the sound source position P (X, Y) obtained by adding the camera position offset obtained by the sound source position calculator 130 is represented by P
(X, Y, d), and the angle Φ in the direction perpendicular to the floor is

[0091]

(Equation 5) Can be obtained by If a rotation mechanism that can rotate in the Z-axis direction is added to the turntable of the first embodiment, it is possible to correctly catch a speaker even when there is an offset in the Z-axis direction. This rotation mechanism can be realized with the same configuration as that of the first embodiment.

Further, in the first embodiment, the position of the sound source in the three-dimensional space is calculated by arranging the microphones at the four vertices of the regular triangular pyramid. It is possible to know. There are four equilateral triangles in the equilateral triangular pyramid, and a calculation perpendicular to the triangular surface indicating the sound source direction on the plane of each triangle can be calculated by performing the calculation as in the first embodiment for each triangle. If a plane indicating the sound source direction is found for each equilateral triangle, the position of the sound source in the three-dimensional space can be known by finding the intersection of each plane.

In the first embodiment, only the case where the calculated sound source position information is used for controlling the camera has been described. However, the sound source position information can be used as index information of a video signal. A sound source position recording device having such an index function will be described below as a second embodiment.

FIG. 18 is an internal configuration diagram of the sound source position recording device. This sound source position recording device has three microphones 1 to
3. The configurations of the sound signal detector 110, the time difference measuring device 120, the sound source position calculator 130, and the reference signal generator 101 are the same as those of the first embodiment shown in FIG.
The same numbers are assigned and the description is omitted. Although not shown in this figure, a camera control device 140 similar to that shown in FIG. 3 is connected to the sound source position calculator 130, and the camera control device 140 is provided with a camera body 203. The turntable 201 and the electric lens 202 are controlled.

In this embodiment, the sound source position calculator 13
The output of 0 is also input to the computer 151 together with the camera control device 140. The computer 151 includes an input device 152, a display device 153, and a storage device 154.
And the control device 155 are connected. Control device 1
A recording / reproducing device 156 is further connected to 55.

The computer 151 has a sound source position table and a time table.
Is registered in the sound source position table, and the time at which the information was received is registered in the time table. The details of the processing executed by the computer 151 and the details of the sound source position table and the time table will be described later.

The input device 152 is used to input a signal indicating completion of sound source position calculation, comment data, and the like. The display device 153 is for displaying sound source position data, time data, and comment data. The storage device 154 stores sound source position data, time data, and comment data, and uses a hard disk device or the like.

The control device 155 is also for controlling the recording / reproducing device 156. The recording / reproducing device 156 stores the video recording recorded according to the first embodiment simultaneously with the recording of the sound source position. In the second embodiment, video recording digitally recorded on a hard disk is used. Unlike a normal tape recording, a hard disk can perform random access quickly, so that a video search operation can be completed in a short time. The user displays the sound source position data, the time data, and the comment data already stored in the storage device 154 on the display device 153, and creates a reproduction instruction here. For example, by playing back only through the moderator's remarks, a task such as creating a meeting agenda list can be efficiently performed. Control device 1
Reference numeral 55 designates fast-forward or playback of the playback device based on the playback time list created by the computer. When the reproduction is started, the reproduction is continued until a fast-forward instruction is given from the input device.

FIG. 19 is a diagram showing an example of the sound source position table. The sound source position table 41 is a matrix of 2 columns and 36 rows, and an x coordinate value and ay coordinate value are stored in a first column and a second column, respectively. "Null" is input to this matrix as an initial value. Note that 36
The number of rows is that the angular resolution is about 5 degrees,
This is because the number of participants at the venue is assumed to be at most this level.

When new sound source position data not stored in the matrix is sent from the sound source position calculator 130, the coordinate values are added to the next row and stored. At this time, if the angle error is within 5 degrees compared with the already registered data and the distance error is within ± 25 cm in both the x and y directions, it is regarded as the same position and no additional processing is performed. The reason for this is that the speaker does not always keep a fixed posture even if he or she keeps the sitting posture, and may move the chair or shake his head.

FIG. 20 is a diagram showing an example of the time table. The time table 42 is a matrix of 36 columns and 1024 rows in an initial state, and the index of each row corresponds to the index of each row of the sound source position table 41. This matrix has "nu" as an initial value except for the first column.
l ”has been entered, but the first column has an initial value of 1
Is stored. This first column is used to store pointer information. That is, in this example, since three pieces of time data of the sound source P1 have already been recorded, 4 is set in the first line giving the pointer of P1. With this information, if there is next time data from the sound source P1,
By referring to this pointer, it is easy to know at which address the time data should be written. When the entry of the time data is completed, the value of the pointer is incremented.

In the expression format of time, the elapsed time from the start of measurement is measured in units of 0.1 second, and the value is expressed by a 2-byte unsigned binary number. Therefore,
The time expression after 1.3 seconds is “13”. Of course, another expression may be used as long as the elapsed time from the start of the measurement is known. At this time, if the received sound source position is already registered in the sound source position table 41, a row index indicating the sound source position is obtained,
New time information is added after the already recorded time information of the corresponding column index of the time table 42. If the received sound source position is not registered in the sound source position table 41, this sound source position is newly added to the sound source position table 4.
1 is registered with the time.

The sound source position table 41 is a two-dimensional matrix having a size of 2 × 36, and is secured on the memory of the computer. Also, the time table 42 is 3 at the start of recording.
A two-dimensional matrix of 6.times.1024 is secured in the memory of the computer, and when the time information at the same position becomes full, 1024 new rows are automatically added by the memory allocation function.

FIG. 21 is a flowchart showing the procedure for recording the sound source position. This is a process performed by the computer 151. Hereinafter, the processing procedure will be described along the step numbers. [S1] the region of the sound source position table T ₁ and the time table T ₂ to secure the memory. [S2] A position storage variable P and a time storage variable x are secured. [S3] the sound source position table T ₁ and the time table T ₂ is initialized. [S4] Start counting at time t [S5] It is determined whether there is an output from the sound source position calculator. If there is an output, the process proceeds to step S6, and if there is no output, this process is repeated. [S6] The current time t is input to the time storage variable x. [S7] The sound source position (x, y) is input to the position storage variable P. [S8] the position storage variable P is already registered in the sound source position table T ₁ to determine. If registered, the process proceeds to step S10, and if not registered, the process proceeds to step S9. [S9] storing new values of the position storage variable P to the sound source position table T _1, the process proceeds to step S10. [S10] Request row index k position storage variable P at the sound source position table T _1. [S11] time seek pointer value a stored in the first row of column k of the table T _2, stores the value of the time table T ₂ of the a row k columns x. [S12] increments the pointer value a which is the time stored in the first row of column k of the table T _2. [S13] It is determined whether the calculation of the sound source position is completed.
If it is completed, the process proceeds to step S14. If not, the process proceeds to step 5. [S14] to store the data of the sound source position table T ₁ and the time table T ₂ on the storage device. [S15] A comment is added if necessary, and the process ends.

In this way, the information on the sound source position and the utterance time can be used as the index of the video.

[0106]

As described above, in the sound source position measuring apparatus of the present invention, three microphones are arranged at the vertices of a predetermined triangle, and the time difference between the sounds from the same sound source reaching the three microphones is measured. Thus, not only the direction of the sound source but also the sound source position can be calculated from the position of each microphone and the arrival time difference of the sound for each microphone pair.

In the camera photographing control device of the present invention,
Since the relative position of the sound source with respect to the camera is calculated from the position of each microphone, the arrival time difference of sound for each microphone pair, and the position of the camera, even if the camera and microphone are far apart, the camera can It can be controlled to face the direction.

In the sound source position recording device of the present invention, the position information of the sound source obtained by measuring the time difference between the sound from the same sound source reaching the three microphones is stored together with the time at which the sound was captured. Therefore, it is easy to obtain information on the utterance time of the speaker.

Further, according to the sound source position measuring method of the present invention, 3
The microphones were placed at the vertices of a predetermined triangle, and the time difference between the sound from the same sound source reaching the three microphones was measured. In addition to the direction, the sound source position can be calculated.

[Brief description of the drawings]

FIG. 1 is a principle configuration diagram of a camera photographing control device of the present invention.

FIG. 2 is a layout diagram of the camera photographing control device according to the first embodiment.

FIG. 3 is an internal configuration diagram of the camera photographing control device according to the first embodiment.

FIG. 4 is a diagram showing an internal configuration of a sound signal detection device.

FIG. 5 is a diagram illustrating a change in a signal in a sound signal detector. (A) is a diagram showing a waveform of a signal input from a microphone, (B) is a diagram showing a waveform of a signal amplified by an amplifier, and (C) is a diagram showing a waveform of a power signal. ,
(D) is a diagram showing the waveform of the signal after the level determination,
(E) is a diagram showing a waveform of a signal output from the differentiator.

FIG. 6 is a diagram showing an internal configuration of a time difference measuring device.

FIG. 7 is a diagram illustrating a square wave obtained from a trigger signal. (A) represents a trigger signal, and (B) represents a square wave.

FIG. 8 is a diagram showing an internal configuration of a delay unit.

FIG. 9 is a diagram showing a coordinate system of a microphone for calculating a sound source direction.

FIG. 10 is a graph showing a relational expression between a time difference between microphones and a sound source direction.

FIG. 11 is a diagram showing a measurement range of one microphone pair.

FIG. 12 is a diagram showing an angle-time difference information table.

FIG. 13 is a diagram illustrating a method for obtaining a sound source position from two pieces of time difference information.

FIG. 14 is a diagram illustrating a range of a sound source determined by a combination of a microphone pair and a sign of time difference data.

FIG. 15 is a diagram showing an internal configuration of a sound source position measuring device.

FIG. 16 is a diagram showing a relationship between a coordinate system of a microphone and a coordinate system of a camera.

FIG. 17 is a diagram showing an internal configuration of a camera control device.

FIG. 18 is an internal configuration diagram of a sound source position recording device.

FIG. 19 is a diagram illustrating an example of a sound source position table.

FIG. 20 is a diagram illustrating an example of a time table.

FIG. 21 is a flowchart showing a processing procedure of sound source position recording.

[Explanation of symbols]

 1-3 microphone 4 sound signal detecting means 5 time difference measuring means 6 sound source position calculating means 7 camera controlling means 8 camera driving means 9 camera body 10 motorized lens 11 utterance information recording means 12 video / audio recording / reproducing means 13 reproduction instructing means

Continued on the front page (51) Int.Cl. ⁶ Identification symbol FI G03B 15/00 G02B 7/11 F H04N 7/15 G03B 3/00 A

Claims (16)

[Claims] 1. A sound source position measuring apparatus for calculating a sound source position based on sound information, wherein microphones installed at three vertices of a triangle having a predetermined shape, and sound emitted from one sound source is captured by each of the microphones. Signal detection means for detecting the detected time as a sound detection time; and, based on the sound detection time for each microphone, the arrival time difference between sounds for each of three microphone pairs obtained by combining two microphones each. A sound source position measuring device comprising: a time difference measuring means for measuring; and a sound source position calculating means for calculating a sound source position from a position of each of the microphones and a difference in arrival time of sound for each microphone pair. 2. The sound source position calculating means, wherein the correspondence between the sound arrival time difference and the direction of the sound source is defined in advance, and each microphone pair is determined from the defined correspondence and the arrival time difference measured for each microphone pair. 2. The sound source position measuring device according to claim 1, wherein a sound source position is calculated by obtaining a straight line connecting the intermediate point and the sound source, and obtaining an intersection of the straight line obtained for each microphone pair. 3. The sound source position measuring device according to claim 2, wherein said sound source position calculating means holds a correspondence relationship between a sound arrival time difference and a sound source direction in a table format. 4. The microphone is disposed at a vertex of an equilateral triangle, and the sound source position calculation means includes a microphone connected to a line connecting each microphone pair.
2. The sound source position measuring device according to claim 1, wherein a range of 60 degrees is defined as a sound source measurement range of each microphone pair, and a possible range of a sound source is limited by a sign of a time difference between sounds captured by each microphone pair. 5. The sound source position measuring device according to claim 1, wherein said sound source position calculating means calculates a sound source position based on a principle of triangulation. 6. The microphone is arranged at four vertices of a triangular pyramid to form a plurality of triangles, and the time difference measuring means is formed based on a sound detection time for each of the microphones. The sound source position calculating means measures the arrival time difference of sound for each of three microphone pairs obtained for each triangle, and calculates the sound arrival position difference in the three-dimensional space from the position of each microphone and the arrival time difference of sound for each microphone pair. The sound source position measuring device according to claim 1, wherein the sound source position is calculated. 7. A camera photographing apparatus for photographing while controlling the direction of a camera installed on a turntable, comprising: microphones installed at three vertices of a triangle having a predetermined shape; Sound signal detection means for detecting a time at which sound is captured by each of the microphones as a sound detection time; three sets obtained by combining two of the microphones based on the sound detection time for each of the microphones Time difference measuring means for measuring a sound arrival time difference for each microphone pair, based on information on the position of each microphone, sound arrival time difference for each microphone pair, and the position of the camera, relative position information of the sound source with respect to the camera Sound source position calculating means for calculating the position of the sound source based on the position information of the sound source calculated by the sound source position calculating means. The camera imaging control apparatus characterized by comprising: a stage, a. 8. The camera photographing control device according to claim 7, wherein said camera control means also controls a zoom amount based on the sound source position information calculated by said sound source position calculation means. 9. The sound source position calculating means, wherein the correspondence between the sound arrival time difference and the direction of the sound source is defined in advance, and based on the defined correspondence and the arrival time difference measured for each microphone pair, 8. The sound source position is calculated by obtaining a straight line connecting the intermediate point and the sound source, and obtaining an intersection of the straight line obtained for each microphone pair.
A camera photographing control device as described in the above. 10. The camera photographing control apparatus according to claim 7, wherein said sound source position calculating means holds a correspondence between a sound arrival time difference and a sound source direction in a table format. 11. The microphone is disposed at an apex of an equilateral triangle, and the sound source position calculating means includes a microphone connected to a line connecting each microphone pair.
8. The camera photographing control device according to claim 7, wherein a range of 60 degrees is defined as a sound source measurement range of each microphone pair, and a possible range of a sound source is limited by a sign of a time difference between sounds captured by each microphone pair. 12. The sound source position calculating means three-dimensionally calculates relative position information of a sound source with respect to the camera based on information on a height from the camera to a plane on which the three microphones are placed. The camera photographing control device according to claim 7, wherein: 13. The camera photographing control apparatus according to claim 7, further comprising: utterance information recording means for recording an utterance time for each sound source position based on a calculation result of said sound source position calculation means. 14. A video / audio recording / reproducing unit for recording and reproducing a sound captured by the microphone and a video taken by the camera, wherein the utterance information recording unit includes a video / audio recording / reproducing unit. 14. The camera photographing control device according to claim 13, wherein an utterance time for each sound source position is recorded as an index of the recorded video. 15. A sound source position recording apparatus for recording a sound source position based on sound information, wherein: microphones installed at three vertices of a triangle having a predetermined shape; Sound signal detecting means for detecting a captured time as a sound detection time; and arrival of sound for each of three microphone pairs obtained by combining two microphones based on the sound detection time for each microphone. A time difference measuring means for measuring a time difference; a sound source position calculating means for calculating a sound source position from a position of each of the microphones and a sound arrival time difference for each microphone pair; and a sound source position calculated by the sound source position calculating means. When the sound is within the range, the sound is emitted from the same sound source, and when the sound is out of the predetermined range, the sound is emitted from a different sound source. Sound source position recording apparatus characterized by having a speech information recording means for recording. 16. A sound source position measuring method for calculating a sound source position based on sound information, wherein a time at which a sound emitted from one sound source is captured by each of microphones installed at three vertexes of a triangle having a predetermined shape is obtained. Detected as a sound detection time, based on the sound detection time for each microphone, measure the arrival time difference of sound for each of the three microphone pairs obtained by combining two of the microphones, A sound source position measurement method, comprising calculating a sound source position from a position and a difference in arrival time of sound for each microphone pair.