JP4229608B2 - Audio localization method for conference recording system - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an audio localization method in a conference recording system for recording using a plurality of microphones. More specifically, the present invention records conferences, lectures, conference presentations, business presentations, etc. with video and audio, and the video recording is recorded by a speaker. The present invention relates to an audio position specifying method in a conference recording system that makes it easy to record audio with high quality in order to tag and structure with the content of utterances, and to collect location information of conference participants and the like.
[0002]
[Prior art]
Conventionally, a method for detecting the direction of a speaker has been studied in order to record conferences, lectures, conference presentations, business presentations, etc. as video and audio, and tag and structure the video records with the content of the speaker's speech. It has been.
For example, the “camera control system in a video conference system” described in Japanese Patent Application Laid-Open No. Hei 6-276514 is provided with an electromagnetic wave source in a microphone so that the camera is directed in the direction of electromagnetic wave emission.
A “TV conference person camera turning system” described in Japanese Patent Laid-Open No. 5-153582 is a method of specifying a speaker using a plurality of directional microphones arranged radially, and Japanese Patent Laid-Open No. 5-111020. The “video conference screen switching control device” described in the official gazette is a voice / non-voice discrimination method using a plurality of microphones and a priority detection method for a plurality of speakers.
[0003]
[Problems to be solved by the invention]
The method of detecting the direction of the speaker as described above has a problem that a large number of microphones are used and that the devices are difficult to install and can only be used in large-scale conferences. Or hinder use.
Further, a device that devise the sound collecting device to detect the direction of the speaker cannot obtain the distance information of the speaker. Similarly, even if a microphone is provided with an electromagnetic wave (including light) source and a position is specified using a directional antenna or a video camera, distance information between the microphones cannot be obtained. In this case, the distance information between the microphones necessary for regenerating the recorded sound field and removing noise is lost, and it is difficult to realize such a function.
[0004]
Therefore, the present invention facilitates the wiring work between the devices having the most troublesome microphone among the installation operations of the conference recording system, and automatically forms the network between the devices and specifies the relative position of the installed microphone. It is an object of the present invention to provide an audio localization method in a conference recording system that can regenerate a recorded sound field and remove noise more effectively.
[0005]
[Means for Solving the Problems]
The present invention has been made in view of the above circumstances, the invention of claim 1, Oite conference recording to the conference recording system for recording using a plurality of microphones, parents at least one of the plurality of microphones Machine, a plurality of microphones other than the master unit as a slave unit, the master unit acquires distance information between the master unit and all slave units and distance information between each slave unit, based on the distance information, The parent that identifies the position of a child device located at the apex of a triangle composed of the parent device and two child devices, and records the conference record via the network between the parent device and the child device. The position of the machine and the slave unit is automatically identified. According to a second aspect of the present invention, in the audio position specifying method in the conference recording system of the first aspect, the distance between the master unit and the slave unit and the distance between the slave units are based on the arrival time of the distance detecting sound wave. It is calculated.
[0006]
According to a third aspect of the present invention, in the audio localization method in the conference recording system according to the first or second aspect of the present invention, the network between the parent device and the child device is formed using a wireless LAN.
[0007]
According to a fourth aspect of the present invention, in the audio localization method in the conference recording system of the first to third aspects of the present invention, the network is periodically reconfigured in order to cope with the addition or movement of the handset microphone during recording. The position of the master unit and the slave unit is re-specified by sound waves.
[0008]
According to a fifth aspect of the present invention, in the audio position identification method in the conference recording system according to the first to fourth aspects of the present invention, the position of the master unit and the slave unit by the sound wave is performed using a plurality of frequencies. To do.
[0009]
According to a sixth aspect of the present invention, in the audio localization method in the conference recording system of the fifth aspect , all or any one of the plurality of frequencies is outside the audible frequency band.
[0010]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described below based on the examples shown in FIGS.
FIG. 1 is a diagram showing the overall configuration of the voice position specifying method of the present invention.
In the audio position identification method in the conference recording system of the embodiment, it is composed of at least one master unit 100 and any plurality of slave units 200a to 200n (in the example shown in FIG. 1, six slave units 200a to 200f). ), Each of which has a microphone, and the master unit and the slave unit have components as shown in FIG.
[0011]
2A and 2B are diagrams showing the configuration of the master unit and the slave unit. FIG. 2A is a block diagram showing the configuration of the master unit 100, and FIG. 2B is a block diagram showing the configuration of each of the slave units 200a to 200n. It is. In addition, since the subunit | mobile_unit 200a-200n is respectively the same structure, the code | symbol of a-n is abbreviate | omitted.
The base unit 100 includes a CPU 101, a clock 102, a RAM 103, a ROM 104, an HDD 105, an expansion bus 106, a wireless network I / F 107, an audio input / output I / F 108, a display card 109, a keyboard I / F 110, a mouse I / F 111, and the like. Are connected to the wireless network I / F 107, the audio input / output I / F 108, the display card 109, the keyboard I / F 110, and the mouse I / F 111, respectively, with a wireless antenna 113, a microphone 114, a speaker 115, a display 116, a keyboard 117, A mouse 118 is connected.
[0012]
Base unit 100 is provided with a microphone 114 and a wireless antenna 113 necessary for automatically configuring a wireless network, and a speaker 115 necessary for generating a position detection sound signal. These are connected to the CPU 101 via the wireless network I / F 107, the audio input / output I / F 108, and the bus 112.
The CPU 101 has a function of performing control operations necessary for the present invention using the clock 102 as a reference signal for operation control and a reference signal for time control, and exchanging start / end of recording, system configuration, etc. with an operator. .
[0013]
Components other than those described above constituting the parent device 100 are devices that are generally necessary for operating the CPU 101.
That is, the RAM 103, the ROM 104, and the HDD 105 for storing programs, the display 116, the display card 109, the keyboard 117, the keyboard I / F 110, the mouse 118, and the mouse necessary for the operator to instruct and monitor the operation of the apparatus. I / F 111 or the like. An expansion bus 106 is provided for future expansion of the device.
[0014]
In the slave unit 200 (200a to 200n), a CPU 201, a clock 202, a RAM 203, a ROM 204, a wireless network I / F 205, an audio input / output I / F 206, and the like are connected to a bus 207, and a wireless network I / F 205, an audio input / output I / F are connected. A wireless antenna 208, a microphone 209, and a speaker 210 are connected to F206, respectively.
Since the handset 200 does not require an operator other than turning on the power, this portion is omitted from the base unit 100.
[0015]
The operation of the audio position specifying method in the conference recording system of the embodiment will be described.
FIG. 3 and FIG. 4 are diagrams showing the process of the audio position specifying method in the conference recording system of the embodiment, FIG. 3 is a flowchart showing the process of the master unit, and FIG. 4 is a flowchart showing the process of the slave unit. .
The general flow of processing performed by the parent device and the child device is as follows.
A1. A network is formed between the parent device and the child device using a wireless LAN. (Initialization process 300 in FIG. 3 and initialization process 400 in FIG. 4)
A2. Distance calculation between the master unit and the slave unit (master unit-slave unit distance calculation 310 in FIG. 3, master unit-slave unit distance calculation 410 in FIG. 4), distance calculation between the slave unit and slave unit (slave unit in FIG. 3) The relative position between the master unit and the slave unit and between the slave units is determined by the slave unit distance calculation 320 and the slave unit-slave unit distance calculation 420 in FIG.
A3. Start recording. (Recording process 340 in FIG. 3, recording process 440 in FIG. 4)
[0016]
First, the operation of A1 will be described based on FIG.
The power is turned on by the operation of the operator (S301). After performing initialization that is generally required as a CPU (S302), a broadcast (Broadcast: all devices connected to a network) that releases a master device ID that can be recognized by a slave device as a minimum device (S303) and waits for a response from the slave unit (S304).
If the response from the slave unit is within a preset timeout period, the ID of the slave unit is registered (S305, S306). After the time-out has elapsed (S306-YES), it is considered that there has been a response from all the currently available slave units, the processing of A1 is completed, and the processing proceeds to A2.
[0017]
In the slave unit, when the initialization process of the initialization process 400 in FIG. 4 has progressed to Listen at the time of the master unit operation (S303) (S403), after receiving the broadcast (Broadcast) from the master unit (S404-YES), ID transmission of a subunit | mobile_unit is performed (S405). Thereafter, confirmation from the parent device that is returned as confirmation of the slave device registration operation (S305) by the parent device is received (S406), the processing of A1 is completed, and the processing proceeds to A2.
[0018]
The operation of A2 will be described.
A broadcast (Broadcast) for notifying that the distance detection process between the parent device and the child device is started flows from the parent device (S311). Immediately after the end of this process, a timer count process is performed (S312), and the timer is operated from the count value 0.
[0019]
The slave unit receives the broadcast from the master unit (S411), similarly performs timer count processing (S412), and operates the timer from the count value 0. As a result, the count of the timer is started from the count value 0 at almost the same time (preferably about several microseconds) in all the slave units.
[0020]
Here, until the notification (broadcast) is issued by the wireless antenna 113 after the CPU 101 of the parent device actually issues the notification command, the CPU 201 of the child device starts the timer count from the reception of the wireless antenna 208 of the child device (S412). ) Is theoretically calculated by the number of instructions, etc., the timer count values of the parent device and the child device are corrected appropriately.
In addition, the delay between the wireless antenna 113 of the parent device and the wireless antenna 208 of the child device is much higher than the speed of the sound wave to be used for distance calculation, and is ignored.
[0021]
When the master unit confirms that all the slave units have received a response indicating that the notification has been accepted (YES in S313), the master unit outputs a distance detection sound signal (S314). Next, the time when the sound signal output is detected from the slave unit (timer count value in the slave unit) is received (S315), and the distance between the master unit and all the slave units is calculated based on this data (S316). In this distance calculation, the distance D is obtained by multiplying the value T obtained by subtracting the timer count value T2 of the parent device from the timer count value T1 at which the child device has received the distance detection sound wave by the sound speed S and dividing by two (D = S * (T1-T2) / 2).
[0022]
Next, the parent device transmits a distance detection command to each child device ID registered in (S305) (S321).
On the other hand, the slave unit confirms the commanded slave unit ID and its own ID (S421), and if they match (S422-YES), emits a distance detection sound wave (S423). If they do not match (S422-NO), a distance detection sound wave from another child device is detected (S424), and a timer count at the time of detection is transmitted to the parent device (S425).
[0023]
The master unit receives it and calculates the distance between the slave units (S324). This is performed for all the slave units (S325). When this is completed, distance information between the parent device and each child device is collected in the parent device. From these pieces of information, a positional relationship and distance without contradiction are obtained based on the triangular construction theorem (S326). Then, the end of the position detection process is notified from the master unit to all the slave units (S327), and the process of A2 is completed (S328).
[0024]
Finally, the recording start of A3 will be described.
When the master receives a recording command from the operator (S341), it issues a recording start command to all the slaves (S342).
The slave unit receives this (S441), transmits the recorded data to the master unit (S442), and the master unit receives and records the recorded data (S344).
[0025]
At this time, every time the set time t elapses (S346), the set time counter is re-executed every time the set time t elapses so as to correspond to the extension of the slave units during recording and the movement of the slave units. Is set (S343).
When it is detected that the set time t has elapsed (S346-YES), update processing is performed (S347). For this, the slave unit always checks whether there is a notification in the master unit (S346) (S443), and if there is a notification (YES in S443), a corresponding process is performed (S444).
[0026]
The recording process continues until recording stop is input to the master unit by the operator (S345). When recording stop is input to the master unit (S345-YES), a recording end notification is notified to the slave unit (S332), the slave unit receives it (S445), and the master unit and the slave unit terminate the processing. (S331) (S431).
[0027]
【The invention's effect】
As is clear from the above description, according to the present invention, the most troublesome wiring and adjustment processing in the installation operation of the conference recording system is automatically performed, and the relative position of the installed microphone is known, so that the recording can be performed. The sound field generated can be regenerated and noise can be removed more effectively, and dictation, speech recognition, quality, and efficiency can be improved.
[0028]
In particular, the invention of claim 1, 2, the master unit and the slave unit having a microphone, respectively, between before Symbol host and terminals, forming a network between handset another and by ultrasound using the speaker Since the positions of the parent device and the child device are automatically identified, setting processing for a plurality of microphones can be reduced. The invention of claim 3 can further reduce the installation processing of a plurality of microphones by using a wireless LAN. The invention according to claim 4 can cope with the addition and movement of the handset microphone during recording as needed. The invention of claim 5 can provide information necessary for more accurate sound field regeneration and noise removal by having distance information by sound waves of a plurality of frequencies. According to the sixth aspect of the present invention, it is possible to perform the processing without letting a person hear the sound generated by the position detection processing during recording.
[Brief description of the drawings]
FIG. 1 is a diagram showing an overall configuration of the present invention.
FIG. 2 is a block diagram showing a device configuration of a parent device and a child device shown in FIG.
FIG. 3 is a flowchart showing initialization processing (network generation with a child device), direction detection processing with a child device, and recording operation after power-on of the parent device.
FIG. 4 is a flowchart showing initialization processing (network generation with the parent device) from the power-on of the child device, direction detection processing between the child devices, and recording operation.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 100 ... Master unit, 101 ... CPU, 102 ... Clock, 103 ... RAM, 104 ... ROM, 105 ... HDD, 106 ... Expansion bus, 107 ... Wireless network I / F, 108 ... Audio input / output I / F, 109 ... Display Card 110, keyboard I / F, 111 mouse I / F, 112 bus, 113 wireless antenna 114 microphone, 115 speaker 116 display 117 keyboard 118 mouse 200 (200a to 200f) ) ... Slave unit 201 ... CPU, 202 ... Clock, 203 ... RAM, 204 ... ROM, 205 ... Wireless network I / F, 206 ... Audio input / output I / F, 207 ... Bus, 208 ... Wireless antenna, 209 ... Microphone 210 ... Speakers.

Claims (6)

Oite conference recording to the conference recording system for recording using a plurality of microphones, the at least one of the plurality of microphones to the master unit, and slave unit a plurality of microphones other than the parent machine, the master machine, The distance information between the parent device and all the child devices and the distance information between each child device are obtained, and based on the distance information, is located at the apex of a triangle composed of the parent device and the two child devices. A conference record that identifies the location of a slave unit and automatically identifies the location of the master unit and the slave unit that records the conference record via a network between the master unit and the slave unit Voice localization method in the system. 2. The audio position specifying method in the conference recording system according to claim 1, wherein the distance between the master unit and the slave unit and the distance between the slave units are calculated based on the arrival time of the distance detecting sound wave. Audio location method in conference recording system. 3. The audio position specifying method in the conference recording system according to claim 1, wherein the network between the parent device and the child device is formed using a wireless LAN. 4. The audio position specifying method in the conference recording system according to claim 1, wherein the network is periodically reconfigured to cope with the addition or movement of a slave microphone during recording, and the master by sound waves And an audio position specifying method in the conference recording system, wherein the position of the slave unit is specified again. In the speech localization method in conference recording system according to any one of claims 1 to 4, wherein the specific position of the master unit and the slave unit by sound waves, in a conference recording system and performs using a plurality of frequencies Voice location method. 6. The audio position specifying method in the conference recording system according to claim 5 , wherein all or any one of the plurality of frequencies is outside an audible frequency band.

Images