JP2020148930A - Voice analyzer, voice analysis system and voice analysis method - Google Patents

Abstract

To be able to reduce a cost of analyzing voice in discussions involving multiple participants.SOLUTION: A voice analyzer 1 according to an embodiment of the present invention is a voice analyzer which analyzes sounds emitted by multiple participants around the sound collecting device, and includes: a receiving unit 111 which receives, from a communication terminal, sound collecting device identification information which can identify a sound collecting device, a participant identification information which can identify the participant, and position designation information which specifies a position with respect to the sound collecting device which are obtained by the communication terminal; a position setting unit 112 which sets the position of each of multiple participants with respect to the sound collecting device, based on the sound collecting device identification information, the participant identification information, and the position designation information which are received by the receiving unit 111; a voice acquisition unit 113 which acquires the voice from the sound collecting device; and a voice analysis unit 114 which analyzes the voice emitted by each of the multiple participants included in the voice acquired by the voice acquisition unit 113, based on each of the positions of the multiple participants set by the position setting unit 112.SELECTED DRAWING: Figure 2

Description

The present invention relates to a voice analyzer, a voice analysis system, and a voice analysis method for analyzing the voice of a discussion by a plurality of participants.

In Patent Document 1, in a discussion such as a conference in which a plurality of participants participate, a device provided with a camera and a microphone is placed on a table, and the orientation of the speaker is specified using a face image captured by the camera. In addition, a system is described in which the utterance content is specified using the voice acquired by the microphone, and the speaker and the utterance content are recorded in association with each other.

Japanese Unexamined Patent Publication No. 2005-274707

Since the system described in Patent Document 1 requires a device in which a camera and a microphone are integrated for each group of discussions, it is costly to introduce the system. In the analysis of voice of active learning performed by students and the analysis of voice of meetings in an organization, it is expected that many groups will have discussions in parallel, so it is possible to analyze voice at low cost. Is required.

The present invention has been made in view of these points, and an object of the present invention is to reduce the cost for analyzing speech in a discussion in which a plurality of participants participate.

The voice analyzer of the first aspect of the present invention is a voice analyzer that analyzes sounds emitted by a plurality of participants around the sound collector, and can identify the sound collector acquired by the communication terminal. A receiving unit that receives the sound collecting device identification information, the participant identification information that can identify the participant, and the position designation information that specifies the position with respect to the sound collecting device from the communication terminal, and the receiving unit. A position setting unit that sets the position of each of the plurality of participants with respect to the sound collector based on the received sound collector identification information, the participant identification information, and the position designation information, and the sound collector from the sound collector. Based on the voice acquisition unit for acquiring the sound and the positions of the plurality of participants set by the position setting unit, each of the plurality of participants included in the sound acquired by the voice acquisition unit emitted the sound. It has a sound analysis unit that analyzes the sound.

The receiving unit uses the communication terminal to obtain the sound collecting device identification information acquired by the communication terminal reading a first identification information presenting unit attached to the sound collecting device that can identify the sound collecting device. May be received from.

The receiving unit communicates the participant identification information acquired by the communication terminal reading a second identification information presenting unit that can identify each of the plurality of participants possessed by the plurality of participants. It may be received from the terminal.

The receiving unit may receive the participant identification information acquired by the communication terminal reading the biometric information of each of the plurality of participants from the communication terminal.

Based on the position designation information, the position setting unit selects the seat positions of the plurality of participants in the seat arrangement around the sound collecting device set in advance, and the selected seat positions. May be set as the position of the participant.

The receiving unit receives the order in which the communication terminal acquires the participant identification information of each of the plurality of participants as the position designation information, and the position setting unit is based on the order received by the receiving unit. Therefore, the position of the seat of each of the plurality of participants may be selected in the seat arrangement.

The receiving unit receives an operation on the communication terminal as the position designation information, and the position setting unit receives each of the plurality of participants in the seat arrangement based on the operation received by the receiving unit. The position of the seat may be selected.

The voice analysis system according to the second aspect of the present invention includes a voice analyzer that analyzes sounds emitted by a plurality of participants around the sound collector, and a communication terminal capable of communicating with the voice analyzer. In the system, the communication terminal has sound collector identification information that can identify the sound collector, participant identification information that can identify the participant, and position designation information that specifies a position with respect to the sound collector. The voice analyzer has the acquisition unit for acquiring the above, the sound collector identification information, the participant identification information, and the transmission unit for transmitting the position designation information, and the communication terminal acquires the voice analyzer. A receiving unit that receives the sound collecting device identification information, the participant identification information, and the position designation information from the communication terminal, the sound collecting device identification information received by the receiving unit, and the participant. A position setting unit that sets the position of each of the plurality of participants with respect to the sound collector based on the identification information and the position designation information, a sound acquisition unit that acquires the sound from the sound collector, and the position setting. It has a voice analysis unit that analyzes the sound emitted by each of the plurality of participants included in the sound acquired by the sound acquisition unit based on the position of each of the plurality of participants set by the unit. ..

The voice analysis method of the third aspect of the present invention is a voice analysis method for analyzing sounds emitted by a plurality of participants around a sound collector, which is executed by a processor and acquired by a communication terminal. A step of receiving from the communication terminal the sound collector identification information that can identify the sound device, the participant identification information that can identify the participant, and the position designation information that specifies the position with respect to the sound collector. A step of setting the position of each of the plurality of participants with respect to the sound collecting device based on the sound collecting device identification information, the participant identification information, and the position designation information received by the receiving step, and the sound collecting. Based on the step of acquiring the sound from the device and the position of each of the plurality of participants set by the step to be set, each of the plurality of participants included in the sound acquired by the acquisition step emits. It has a step of analyzing the said voice.

According to the present invention, there is an effect that the cost for analyzing speech in a discussion in which a plurality of participants participate can be reduced.

It is a schematic diagram of the voice analysis system which concerns on embodiment. It is a block diagram of the voice analysis system which concerns on embodiment. It is a front view of the communication terminal which displays the seat arrangement setting screen. It is a schematic diagram for demonstrating the method of registering the sound collector used for the acquisition of the sound of a discussion. It is a schematic diagram for demonstrating the method of registering a participant who participates in a discussion. It is a front view of the communication terminal which displays the participant registration screen. It is a front view of the communication terminal which displays the analysis result screen. It is a figure which shows the flowchart of the voice analysis method performed by the voice analysis system.

[Overview of voice analysis system SS]
FIG. 1 is a schematic diagram of a voice analysis system SS according to the present embodiment. The voice analysis system SS includes a voice analysis device 1, a communication terminal 2, and a sound collector 3. The number of communication terminals 2 and sound collecting devices 3 included in the voice analysis system SS is not limited. The voice analysis system SS may include other devices such as servers and terminals.

The sound collecting device 3 includes a microphone array including a plurality of sound collecting units (microphones) arranged in different directions. For example, a microphone array includes eight microphones evenly spaced on the same circumference in a horizontal plane with respect to the ground. By using such a microphone array, in the voice analyzer 1, any participant U is a speaker (sound source) based on the voices emitted by a plurality of participants U surrounding the sound collector 3. Can be specified. The sound collecting device 3 transmits the voice acquired by using the microphone array to the voice analyzer 1 as data.

The communication terminal 2 is a computer capable of performing communication. The communication terminal 2 is, for example, a computer terminal such as a personal computer or a mobile terminal such as a smartphone. The communication terminal 2 sets analysis conditions for the voice analyzer 1 and displays information received from the voice analyzer 1.

The voice analysis device 1 is a computer that analyzes voice using the voice acquired by the sound collector 3. The voice analyzer 1 is composed of, for example, a single computer or a cloud, which is a collection of computer resources.

The voice analyzer 1 is connected to the communication terminal 2 and the sound collecting device 3 by wire or wirelessly via a network N such as a local area network or the Internet. The voice analyzer 1 may be directly connected to at least one of the communication terminal 2 and the sound collecting device 3 without going through the network N.

Even if the sound collector 3 communicates with a management terminal arranged for each room or building where discussions are held via a wireless LAN (Local Area Network) and exchanges data with the voice analyzer 1 via the management terminal. Good. Alternatively, the sound collecting device 3 may include a SIM (Subscriber Identity Module) for mobile communication and communicate with the voice analyzer 1 by mobile communication. In this case, since it is not necessary to construct a wireless LAN between the sound collecting device 3 and the management terminal, it is not easily affected by the network configuration of the place where the sound collecting device 3 is arranged and the radio wave congestion situation.

The outline of the processing executed by the voice analysis system SS will be described below. Participant U has the communication terminal 2 read the tag T attached to the sound collecting device 3 before starting the discussion. In addition, the participant U causes the communication terminal 2 to read the card C such as the student ID card and the employee ID card of the participant U.

The communication terminal 2 is a voice analyzer that obtains the identification information of the sound collecting device 3 indicated by the read tag T, the identification information of the participant U indicated by the read card C, and the information specifying the position with respect to the sound collecting device 3. Send to 1. The position with respect to the sound collecting device 3 is indicated by the order in which the communication terminal 2 reads the plurality of cards C of the plurality of participants U, or the operation of the participants U with respect to the communication terminal 2.

The voice analyzer 1 has a plurality of participants for the sound collector 3 based on the identification information of the sound collector 3 received from the communication terminal 2, the identification information of the participant U, and the information for designating the position with respect to the sound collector 3. U Set each position. Then, the voice analysis device 1 analyzes the voice acquired from the sound collector 3 based on the positions of the plurality of participants U set, and outputs the analysis result using the communication terminal 2.

According to the voice analysis system SS according to the present embodiment, the relative positions of the plurality of participants U with respect to the sound collector 3 are specified based on the information acquired by using the communication terminal 2, and the specified positions are specified. The voice acquired by using the sound collecting device 3 is analyzed based on the above, and the utterances of each of the plurality of participants U are analyzed. Therefore, since the voice analysis system SS does not need to provide a camera on the sound collector 3, the cost for analyzing the voice in the discussion in which a plurality of participants U participate can be reduced. Further, since the voice analysis system SS does not need to arrange microphones for each participant U, it is possible to reduce the trouble of arranging a plurality of microphones corresponding to a plurality of participants U.

[Voice analysis system SS configuration]
FIG. 2 is a block diagram of the voice analysis system SS according to the present embodiment. In FIG. 2, the arrows indicate the main data flows, and there may be data flows not shown in FIG. In FIG. 2, each block shows a functional unit configuration, not a hardware (device) unit configuration. Therefore, the block shown in FIG. 2 may be mounted in a single device, or may be mounted separately in a plurality of devices. Data transfer between blocks may be performed via any means such as a data bus, a network, or a portable storage medium.

The voice analyzer 1 has a control unit 11 and a storage unit 12. The control unit 11 includes a reception unit 111, a position setting unit 112, a voice acquisition unit 113, a voice analysis unit 114, and an output unit 115. The storage unit 12 has a setting information storage unit 121 and an analysis result storage unit 122.

The storage unit 12 is a storage medium including a ROM (Read Only Memory), a RAM (Random Access Memory), a hard disk drive, and the like. The storage unit 12 stores in advance the program executed by the control unit 11. The storage unit 12 may be provided outside the voice analyzer 1, and in that case, data may be exchanged with the control unit 11 via a network.

The setting information storage unit 121 stores the setting information used for voice analysis. The analysis result storage unit 122 stores the voice analysis result. The setting information storage unit 121 and the analysis result storage unit 122 may be storage areas on the storage unit 12, respectively, or may be a database configured on the storage unit 12.

The control unit 11 is, for example, a processor such as a CPU (Central Processing Unit), and by executing a program stored in the storage unit 12, the reception unit 111, the position setting unit 112, the voice acquisition unit 113, and the voice analysis unit 114 And functions as an output unit 115. At least a part of the function of the control unit 11 may be performed by an electric circuit. Further, at least a part of the functions of the control unit 11 may be executed by a program executed via the network.

The communication terminal 2 has a control unit 21, a storage unit 22, a reading unit 23, and a display unit 24. The control unit 21 has an acquisition unit 211, a transmission unit 212, and a reception unit 213. The display unit 24 includes a display device capable of displaying information such as a liquid crystal display. A touch screen capable of detecting the position of contact by a human may be used as the display unit 24.

The reading unit 23 has a configuration according to a method of acquiring a sound collecting device ID and a participant ID, which will be described later. When the sound collector ID and the participant ID are recorded on the IC (Integrated Circuit) chip mounted on the tag T and the card C, the reading unit 23 provides the information recorded on the IC chip by short-range wireless communication. It is equipped with a short-range wireless communication device capable of reading. The short-range wireless communication is, for example, NFC (Near Field Communication). The reading unit 23 includes an imaging device capable of imaging the sound collecting device ID and the participant ID when they are represented by the codes on the tag T and the card C. The code is, for example, a bar code or a two-dimensional code.

Further, when the participant ID is acquired by recognizing the participant's face, the reading unit 23 includes an imaging device capable of imaging the participant's face. Further, when the participant ID is acquired by recognizing the fingerprint of the participant, the reading unit 23 includes a fingerprint scanner.

The storage unit 22 is a storage medium including a ROM, a RAM, a hard disk drive, and the like. The storage unit 22 stores in advance the program executed by the control unit 21. The storage unit 22 may be provided outside the communication terminal 2, and in that case, data may be exchanged with the control unit 21 via a network.

The control unit 21 is, for example, a processor such as a CPU, and functions as an acquisition unit 211, a transmission unit 212, and a reception unit 213 by executing a program stored in the storage unit 22. At least a part of the function of the control unit 21 may be performed by an electric circuit. Further, at least a part of the functions of the control unit 21 may be executed by a program executed via the network.

The voice analyzer 1 and the communication terminal 2 according to the present embodiment are not limited to the specific configuration shown in FIG. The voice analyzer 1 and the communication terminal 2 are not limited to one device each, and may be configured by connecting two or more physically separated devices by wire or wirelessly.

[Explanation of voice analysis method]
The voice analysis method performed by the voice analysis system SS according to the present embodiment will be described below. Before starting the discussion, the communication terminal 2 causes the display unit 24 to display a seat arrangement setting screen for setting the seat arrangement around the sound collecting device 3 when a predetermined operation is performed. The voice analyst performs an operation of setting the seat arrangement around the sound collecting device 3 on the seat arrangement setting screen on the communication terminal 2. Participants in the discussion, rather than the analyst, may operate the communication terminal 2.

FIG. 3 is a front view of the communication terminal 2 displaying the seat arrangement setting screen. The communication terminal 2 displays the setting area 241. The setting area 241 is a virtual circle centered on the sound collecting device 3, and an angle centered on the sound collecting device 3 is represented around the setting area 241. The seat position is represented by a circle in the setting area 241, and the seat number 242 and the cancel button 243 are represented in the vicinity of the seat position. The seat number 242 is sequentially assigned to the set seat in a predetermined direction (for example, clockwise) with respect to the sound collecting device 3.

When the analyst performs an operation of pressing the position of one point in the setting area 241, the communication terminal 2 sets a new seat at the position, and a circle indicating the position of the seat in the setting area 241. To add. When the analyst performs an operation of pressing any of the cancel buttons 243, the communication terminal 2 deletes or invalidates the seat corresponding to the cancel button 243, and indicates the position of the seat in the setting area 241. Erase or invalidate the circle.

When the analyst presses the completion button 244 on the seat arrangement setting screen, the acquisition unit 211 of the communication terminal 2 determines the set seat arrangement (for example, the angle of each of the plurality of seats centered on the sound collecting device 3). get. The operation shown here is an example, and the communication terminal 2 may set the seat arrangement by another operation by the analyst.

The transmission unit 212 of the communication terminal 2 transmits the information indicating the seat arrangement acquired by the acquisition unit 211 to the voice analyzer 1. The receiving unit 111 of the voice analyzer 1 stores the information indicating the seat arrangement received from the communication terminal 2 in the setting information storage unit 121.

The seat arrangement may be set for all the sound collecting devices 3 at once. In this case, the information indicating the seat arrangement is associated with the sound collecting device IDs of all the sound collecting devices 3 in the setting information storage unit 121. Alternatively, the seat arrangement may be set for each sound collecting device 3. In this case, on the seat arrangement setting screen, the analyst performs an operation of designating one or more sound collecting device IDs to be set. Then, the information indicating the seat arrangement is associated with the designated sound collecting device ID in the setting information storage unit 121.

At the beginning of the discussion, a plurality of participants are seated around one sound collecting device 3. One of the plurality of participants participating in the discussion performs an operation of registering the sound collecting device 3 and the participants on the communication terminal 2. An analyst, not a participant, may operate the communication terminal 2. Participants may use the same communication terminal 2 as the communication terminal 2 used for setting the seat arrangement described above in order to register the sound collector 3 and the participants, or may use another communication terminal 2. May be good.

First, the voice analysis system SS executes a process of registering the sound collecting device 3 used for acquiring the voice of the discussion. FIG. 4 is a schematic diagram for explaining a method of registering the sound collecting device 3 used for acquiring the sound of the discussion. Before starting the discussion, the communication terminal 2 causes the display unit 24 to display a sound collecting device registration screen for registering the sound collecting device 3 when a predetermined operation is performed. While the communication terminal 2 is displaying the sound collector registration screen, the participant causes the communication terminal 2 to read the tag T attached to the sound collector 3.

The upper part of FIG. 4 shows a process in which the communication terminal 2 reads the tag T attached to the sound collecting device 3. A tag T is attached to the upper part of the sound collecting device 3. The tag T is an identification information presenting unit that presents sound collecting device identification information (sound collecting device ID) capable of identifying the sound collecting device 3. For example, the sound collector ID may be a number or a character string previously assigned to the sound collector 3 by an analyst, or may be a serial number or MAC address (Media Access Control Address) unique to the sound collector 3. Good.

The tag T may be equipped with an IC chip for short-range wireless communication (NFC or the like) that records the sound collecting device ID. In this case, when the participant brings the communication terminal 2 closer to the tag T, the short-range wireless communication device as the reading unit 23 of the communication terminal 2 reads the information recorded on the IC chip of the tag T by the short-range wireless communication. The acquisition unit 211 of the communication terminal 2 acquires the sound collecting device ID indicated by the information read from the IC chip of the tag T by the reading unit 23.

The tag T may represent a code (bar code, two-dimensional code, etc.) generated by encoding the sound collecting device ID according to a predetermined rule on the surface. In this case, when the participant brings the communication terminal 2 closer to the tag T, the image pickup device as the reading unit 23 of the communication terminal 2 images the code of the tag T. The acquisition unit 211 of the communication terminal 2 acquires the sound collecting device ID by decoding the code of the tag T included in the image captured by the reading unit 23 according to a predetermined rule.

The lower figure of FIG. 4 shows the communication terminal 2 displaying the sound collector registration screen. The communication terminal 2 displays the same setting area 241 as in FIG. 3 on the sound collecting device registration screen. When the acquisition unit 211 acquires the sound collecting device ID, the communication terminal 2 displays a label 245 indicating the sound collecting device ID acquired by the acquisition unit 211 in the central portion of the setting area 241. The label 245 may represent the entire sound collector ID, or may represent a part of the sound collector ID.

When the participant presses the completion button 244 on the sound collector registration screen, the voice analysis system SS moves to the process of registering the participants who participate in the discussion. 5 (a) to 5 (c) are schematic diagrams for explaining a method of registering participants who participate in the discussion. When the completion button 244 is pressed on the sound collector registration screen shown in FIG. 4, the communication terminal 2 registers a plurality of participants participating in the discussion regarding the sound collector ID acquired by the acquisition unit 211. The participant registration screen of is displayed on the display unit 24. While the communication terminal 2 is displaying the participant registration screen, the participant can use any of the methods shown in FIGS. 5 (a) to 5 (c) to provide information for identifying the participant. Let 2 read it.

FIG. 5A shows a process in which the communication terminal 2 reads the card C held by the participant. The card C is an identification information presenting unit that presents participant identification information (participant ID) that can identify a participant. For example, the participant ID may be a number or character string previously assigned to the participant by the analyst, or may be an employee number of the company to which the participant belongs or a student number of the school to which the participant belongs. Good. The voice analyzer 1 associates the participant ID with the participant information (for example, name, affiliation, etc.) and stores it in the storage unit 12 in advance.

The card C may be equipped with an IC chip for short-range wireless communication (NFC or the like) that records a participant ID. In this case, when the participant brings the communication terminal 2 close to the card C, the short-range wireless communication device as the reading unit 23 of the communication terminal 2 reads the information recorded on the IC chip of the card C by the short-range wireless communication. The acquisition unit 211 of the communication terminal 2 acquires the participant ID indicated by the information read by the reading unit 23 from the IC chip of the card C.

The card C may display a code (bar code, two-dimensional code, etc.) generated by encoding the participant ID according to a predetermined rule on the surface. In this case, when the participant brings the communication terminal 2 closer to the card C, the image pickup device as the reading unit 23 of the communication terminal 2 images the code of the card C. The acquisition unit 211 of the communication terminal 2 acquires the participant ID by decoding the code of the card C included in the image captured by the reading unit 23 according to a predetermined rule.

FIG. 5B shows a process in which the communication terminal 2 reads the face of the participant. The communication terminal 2 acquires a participant ID by using a known face recognition process. In this case, the individual identified by the face recognition process and the participant ID are associated in advance in the storage unit 22 of the communication terminal 2. The imaging device as the reading unit 23 of the communication terminal 2 images an area including the face of one participant. The acquisition unit 211 of the communication terminal 2 identifies the individual participant included in the image by performing face recognition processing on the image captured by the reading unit 23. Then, the acquisition unit 211 acquires the participant ID associated with the individual identified by the face recognition process in the storage unit 22.

The face recognition process may be performed by the voice analyzer 1 instead of the communication terminal 2. In this case, the communication terminal 2 transmits an image of the area including the face of the participant to the voice analyzer 1, and the participant ID acquired by the voice analyzer 1 based on the image is transmitted from the voice analyzer 1. Receive.

FIG. 5C shows a process in which the communication terminal 2 reads the fingerprint of the participant. The communication terminal 2 acquires a participant ID by using a known fingerprint authentication process. In this case, the individual identified by the fingerprint authentication process and the participant ID are previously associated with each other in the storage unit 22 of the communication terminal 2. The fingerprint scanner as the reading unit 23 of the communication terminal 2 captures the fingerprint of one participant. The acquisition unit 211 of the communication terminal 2 identifies an individual having a fingerprint included in the image by performing fingerprint authentication processing on the image captured by the reading unit 23. Then, the acquisition unit 211 acquires the participant ID associated with the individual identified by the fingerprint authentication process in the storage unit 22.

The fingerprint authentication process may be performed by the voice analyzer 1 instead of the communication terminal 2. In this case, the communication terminal 2 transmits an image of the area including the fingerprint of the participant to the voice analyzer 1, and the participant ID acquired by the voice analyzer 1 based on the image is transmitted from the voice analyzer 1. Receive.

In the configuration for reading biological information such as the participant's face and fingerprint shown in FIGS. 5 (b) and 5 (c), the voice analysis system SS is in a situation where the participant does not have the card C. Also has the advantage that the participant ID can be specified.

When the participant information associated with the participant ID acquired in FIGS. 5 (a) to 5 (c) is not registered in the voice analyzer 1, the communication terminal 2 registers the participant information. May be accepted. In this case, the communication terminal 2 inquires the voice analyzer 1 whether or not the acquired participant ID is stored in the storage unit 12. When the participant ID is not stored in the storage unit 12 of the voice analyzer 1, the communication terminal 2 displays a participant information input screen (not shown) for inputting participant information (name, affiliation, etc.). It is displayed on the display unit 24.

Participants perform an operation of inputting participant information to the communication terminal 2 displaying the participant information input screen. The transmission unit 212 of the communication terminal 2 transmits the participant information input on the participant information input screen to the voice analyzer 1 together with the participant ID. The receiving unit 111 of the voice analyzer 1 stores the participant ID received from the communication terminal 2 and the participant information in the storage unit 12 in association with each other. As a result, even a participant who has not been registered in the voice analyzer 1 can easily participate in the discussion.

Alternatively, when the participant information associated with the participant ID is not registered in the voice analyzer 1, the communication terminal 2 does not accept the registration of the participant information and registers the participant ID. May be rejected. In this case, the communication terminal 2 inquires the voice analyzer 1 whether or not the acquired participant ID is stored in the storage unit 12. When the participant ID is not stored in the storage unit 12 of the voice analyzer 1, the communication terminal 2 causes the display unit 24 to display information indicating that the participant ID is not registered, and the participant. Reject ID registration.

6 (a) and 6 (b) are front views of the communication terminal 2 displaying the participant registration screen. The voice analyzer 1 sets the positions of a plurality of participants based on the order in which the acquisition unit 211 acquires the participant IDs or the operation of the participants with respect to the communication terminal 2.

FIG. 6A shows a participant registration screen when the voice analyzer 1 sets the positions of a plurality of participants based on the order in which the acquisition unit 211 acquires the participant IDs. The communication terminal 2 displays the same setting area 241 as in FIG. 3 on the participant registration screen.

In the setting area 241, the seat positions set on the seat arrangement setting screen of FIG. 3 are represented by circles. The communication terminal 2 allocates and displays the participant ID 246 to the seats in the order in which the acquisition unit 211 acquires the participant ID in a predetermined direction D (for example, clockwise) with respect to the sound collecting device 3.

When the participant presses the completion button 244 on the participant registration screen, the transmission unit 212 of the communication terminal 2 has the sound collector ID (that is, the sound collector identification information) acquired by the acquisition unit 211 and the acquisition unit 211. The plurality of participant IDs (that is, participant identification information) acquired by the acquisition unit 211 and the order in which the acquisition unit 211 acquired the participant IDs of each of the plurality of participants (that is, position designation information) are associated with each other to the voice analyzer 1. Send.

In the voice analyzer 1, the receiving unit 111 acquires the sound collecting device ID acquired by the communication terminal 2, the participant ID acquired by the communication terminal 2, and the participant ID of each of the plurality of participants in the order in which the communication terminal 2 acquires the participant ID. To receive.

The position setting unit 112 acquires the seat arrangement associated with the sound collecting device ID in the setting information storage unit 121. Then, in the acquired seat arrangement, the position setting unit 112 has a plurality of seat arrangements in the order in which the communication terminal 2 has acquired the participant IDs of the plurality of participants in the predetermined orientation D with respect to the sound collecting device 3. Select a seat for each participant. Then, the position setting unit 112 associates the sound collecting device ID, the plurality of participant IDs, and the positions of the participants of the plurality of participant IDs (that is, the positions of the seats selected for each of the plurality of participants). The positions of the plurality of participants are set by storing the settings in the setting information storage unit 121. The position of the participant is represented by, for example, an angle on a horizontal plane centered on the sound collector 3. Further, when the number of seats is larger than the number of participants, the position setting unit 112 stores the information for setting the seats not set by the participants as vacant seats in the setting information storage unit 121.

FIG. 6B shows a participant registration screen when the voice analyzer 1 sets the positions of a plurality of participants based on the operation of the participants with respect to the communication terminal 2. The communication terminal 2 displays the same setting area 241 as in FIG. 3 and the participant ID 247 acquired by the acquisition unit 211 on the participant registration screen. In the participant ID 247 acquired by the acquisition unit 211, the display mode of the participant ID in which the seat is not set and the participant ID in which the seat is set are shown to be different from each other.

In the setting area 241, the seat positions set on the seat arrangement setting screen of FIG. 3 are represented by circles. When the participant performs an operation of designating the position of one of the seats with respect to the communication terminal 2, the communication terminal 2 has one participant ID 246 for the designated seat (for example, the participant ID 247 acquired by the acquisition unit 211). In, the highest participant ID among the participant IDs for which seats have not been set) is assigned and displayed. The participant repeats the operation of designating a seat for each of the plurality of participant IDs.

When the participant presses the completion button 244 on the participant registration screen, the transmission unit 212 of the communication terminal 2 has the sound collector ID (that is, the sound collector identification information) acquired by the acquisition unit 211 and the acquisition unit 211. The plurality of participant IDs (that is, participant identification information) acquired by the user and the information indicating the operation of designating a seat for each of the plurality of participant IDs (that is, position designation information) are associated and transmitted to the voice analyzer 1. ..

In the voice analyzer 1, the receiving unit 111 receives information indicating an operation for designating a seat for each of the sound collecting device ID acquired by the communication terminal 2, the participant ID acquired by the communication terminal 2, and the plurality of participant IDs. To do.

The position setting unit 112 acquires the seat arrangement associated with the sound collecting device ID in the setting information storage unit 121. Then, the position setting unit 112 selects the seats of each of the plurality of participants in the acquired seat arrangement, as indicated by the operation of designating the seats for each of the plurality of participant IDs. Then, the position setting unit 112 associates the sound collecting device ID, the plurality of participant IDs, and the positions of the participants of the plurality of participant IDs (that is, the positions of the seats selected for each of the plurality of participants). The positions of the plurality of participants are set by storing the settings in the setting information storage unit 121. The position of the participant is represented by, for example, an angle on a horizontal plane centered on the sound collector 3. Further, when the number of seats is larger than the number of participants, the position setting unit 112 stores the information for setting the seats not set by the participants as vacant seats in the setting information storage unit 121.

In this way, the voice analysis system SS is based on the sound collector identification information, the participant identification information, and the position designation information acquired by using the communication terminal 2, and the positions of a plurality of participants with reference to the sound collector 3 To set. Therefore, the voice analysis system SS can easily set the positions of a plurality of participants without providing a camera on the sound collector 3, and reduces the cost for analyzing the voice in the discussion in which the plurality of participants U participate. it can.

When the participant or the analyst starts the discussion, he / she instructs the start of the discussion by operating the communication terminal 2. In the voice analyzer 1, when the voice acquisition unit 113 receives the signal instructing the start of the discussion from the communication terminal 2, the voice acquisition unit 113 transmits the signal instructing the acquisition of the voice to the sound collector 3. The sound collecting device 3 starts acquiring voice when it receives a signal instructing voice acquisition from the voice analyzer 1.

The sound collecting device 3 acquires sound in each of the plurality of sound collecting units, and internally records the sound as the sound of each channel corresponding to each sound collecting unit. Then, the sound collecting device 3 transmits the acquired voices of the plurality of channels to the voice analyzer 1. The sound collecting device 3 may sequentially transmit the acquired sound, or may transmit the sound of a predetermined amount or a predetermined time. Further, the sound collecting device 3 may collectively transmit the voices from the start to the end of the acquisition. In the voice analysis device 1, the voice acquisition unit 113 receives the voice from the sound collection device 3 and stores it in the analysis result storage unit 122 in association with the identification information (for example, the discussion ID) for identifying the discussion. The discussion ID may be automatically assigned to the discussion or may be entered by a participant or analyst.

At the end of the discussion, the participant or the analyst instructs the end of the discussion by operating the communication terminal 2. In the voice analyzer 1, when the voice acquisition unit 113 receives the signal indicating the end of the discussion from the communication terminal 2, the voice acquisition unit 113 transmits the signal instructing the end of the voice acquisition to the sound collector 3. When the sound collecting device 3 receives a signal instructing the end of voice acquisition from the voice analyzer 1, the sound collecting device 3 ends the voice acquisition.

Subsequent processing is performed when the acquisition of the voice is completed or when the analyst gives a predetermined instruction to the communication terminal 2. In the setting information storage unit 121, the voice analysis unit 114 has a plurality of participant IDs associated with the sound collector ID of the sound collector 3 from which the voice is acquired, and the positions of the participants of the plurality of participant IDs. And get. The voice analysis unit 114 performs sound source localization based on the voices of the plurality of channels received from the sound collector 3. The sound source localization is a process of estimating the direction of the sound source included in the voice acquired by the voice acquisition unit 113 every time (for example, every 10 milliseconds to 100 milliseconds). The voice analysis unit 114 associates the orientation of the sound source estimated for each time with the positions of the plurality of participants acquired from the setting information storage unit 121.

If the direction of the sound source can be specified based on the acquired voice, the voice analysis unit 114 can use a known sound source localization method such as a MUSIC (Multiple Signal Classification) method or a beamforming method.

Next, the voice analysis unit 114 makes a speech (speech) by any participant at predetermined time intervals (for example, every 10 milliseconds to 100 milliseconds) in the discussion based on the acquired voice and the estimated direction of the sound source. Determine if you did. The voice analysis unit 114 specifies a continuous period from the start to the end of the utterance by one participant as the utterance period. When a plurality of participants speak at the same time, at least a part of the utterance period of the plurality of participants overlaps with each other. The voice analysis unit 114 stores the utterance period specified in the discussion in the analysis result storage unit 122 in association with the discussion ID and the participant ID.

Further, the voice analysis unit 114 acquires the time-series utterance amount (also referred to as the utterance amount) of each of the plurality of participants in the discussion based on the specified utterance period. Specifically, the voice analysis unit 114 divides the discussion into frames (time range) having a predetermined window width (for example, 30 seconds). The frames are shifted by a predetermined shift width (for example, 10 seconds) shorter than the window width, and some of the adjacent frames overlap each other in chronological order.

Then, the voice analysis unit 114 calculates the value obtained by dividing the length of the participant's utterance period (total utterance time) in the frame by the window width as the utterance amount for each frame. The voice analysis unit 114 calculates the amount of utterance for each frame from the start time to the end time of the discussion for each of the plurality of participants. The voice analysis unit 114 stores information indicating the amount of utterances for each frame of the plurality of participants in the discussion in the analysis result storage unit 122 in association with the discussion ID and the participant ID.

The output unit 115 outputs the voice analysis result. For example, the output unit 115 outputs information indicating the utterance amount of each of the plurality of participants as an analysis result. The output unit 115 may output other information that can be analyzed using the voice acquired by the sound collecting device 3 as an analysis result.

The output unit 115 may display the analysis result on the same communication terminal 2 as the communication terminal 2 used for the registration of the sound collecting device 3 and the registration of the participants, and may display the analysis result on another communication terminal 2. It may be displayed. The output unit 115 outputs the analysis result not only by displaying the screen but also by printing on paper using a printer, storing it as data in a storage medium, or transmitting it to the outside via a communication line. May be good.

The output unit 115 stores in the analysis result storage unit 122 when the processing of the voice analysis unit 114 is completed or when the analyst gives an instruction to output the analysis result to the communication terminal 2. Display information for displaying the analysis result is generated based on the information, and is transmitted to the communication terminal 2. The receiving unit 213 of the communication terminal 2 receives the display information transmitted by the voice analyzer 1 and causes the display unit 24 to display the analysis result screen shown in FIG. 7.

FIG. 7 is a front view of the communication terminal 2 displaying the analysis result screen. The analysis result screen is a screen for displaying information related to one discussion. The analysis result screen includes a graph 248 of the time-series utterance volume of a plurality of participants. Graph 248 represents the amount of utterances of a plurality of participants as a stacked graph. The horizontal axis of the graph 248 is time, and the vertical axis is the amount of utterance. In the area of the graph 248, different patterns are shown according to each of the plurality of participants. As a result, the analyst can know the result of voice analysis by the voice analyzer 1.

[Flow of voice analysis method]
FIG. 8 is a diagram showing a flowchart of a voice analysis method performed by the voice analysis system SS. Before starting the discussion, the acquisition unit 211 of the communication terminal 2 acquires the seat arrangement set on the seat arrangement setting screen for setting the seat arrangement. The transmission unit 212 of the communication terminal 2 transmits the information indicating the seat arrangement acquired by the acquisition unit 211 to the voice analyzer 1. The receiving unit 111 of the voice analyzer 1 stores the information indicating the seat arrangement around the sound collecting device 3 received from the communication terminal 2 in the setting information storage unit 121 (S11).

At the beginning of the discussion, a plurality of participants are seated around one sound collecting device 3. One of the plurality of participants participating in the discussion performs an operation of registering the sound collecting device 3 and the participants on the communication terminal 2.

The acquisition unit 211 of the communication terminal 2 acquires the sound collecting device ID indicated by the information read from the IC chip of the tag T by the reading unit 23, or determines the code of the tag T included in the image captured by the reading unit 23. The sound collector ID is acquired by decoding according to the rules.

Further, the acquisition unit 211 of the communication terminal 2 acquires the participant ID indicated by the information read from the IC chip of the card C by the reading unit 23, or determines the code of the card C included in the image captured by the reading unit 23. The participant ID is obtained by decrypting according to the rules of. Alternatively, the acquisition unit 211 of the communication terminal 2 may acquire the participant ID by reading the biometric information of the participant by using the face recognition process or the fingerprint authentication process.

The transmission unit 212 of the communication terminal 2 associates the sound collector identification information (sound collector ID), the participant identification information (participant ID), and the position designation information acquired by the acquisition unit 211 with each other to the voice analyzer 1. Send. The position designation information is information indicating the order in which the acquisition unit 211 has acquired the participant IDs of each of the plurality of participants, or the operation of designating a seat for each of the plurality of participant IDs.

In the voice analyzer 1, the receiving unit 111 receives the sound collecting device identification information, the participant identification information, and the position designation information acquired by the communication terminal 2 (S12). The position setting unit 112 selects a seat for each of the plurality of participants based on the sound collector identification information, the participant identification information, and the position designation information received by the reception unit 111, and uses the positions of the selected seats to obtain a plurality of seats. The position of each participant is set (S13).

After the discussion is started, the voice acquisition unit 113 receives the voice of the discussion from the sound collecting device 3 (S14). In the setting information storage unit 121, the voice analysis unit 114 has a plurality of participant IDs associated with the sound collector ID of the sound collector 3 from which the voice is acquired, and the positions of the participants of the plurality of participant IDs. And get.

The voice analysis unit 114 analyzes the received voice of the discussion by performing sound source localization using the received voice and the acquired position of the participant and calculating the utterance period and the utterance amount of each of the plurality of participants (the voice analysis unit 114). S15). The output unit 115 outputs the result of voice analysis by the voice analysis unit 114 (S16).

[Effect of this embodiment]
The voice analysis system SS according to the present embodiment has a plurality of participants based on the sound collector 3 based on the sound collector identification information, the participant identification information, and the position designation information acquired by using the communication terminal 2. Set the position. With such a configuration, the voice analysis system SS can set the positions of a plurality of participants without the need to provide a camera on the sound collecting device 3 as in the system described in Patent Document 1. The cost of analyzing audio in discussions in which participants participate can be reduced.

Further, although it takes a lot of time and effort for one analyst to set the positions of a plurality of participants with respect to a plurality of discussion groups, in the voice analysis system SS according to the present embodiment, the participants themselves communicate with each other. Since the positions of the participants participating in the discussion can be set using the terminal 2, the time and effort for setting the positions of the participants can be reduced.

Although the present invention has been described above using the embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments, and various modifications and changes can be made within the scope of the gist. is there. For example, all or a part of the device can be functionally or physically distributed / integrated in any unit. Also included in the embodiments of the present invention are new embodiments resulting from any combination of the plurality of embodiments. The effect of the new embodiment produced by the combination has the effect of the original embodiment.

The processor of the voice analyzer 1 and the communication terminal 2 is the main body of each step included in the voice analysis method shown in FIG. That is, the processors of the voice analyzer 1 and the communication terminal 2 read a program for executing the voice analysis method shown in FIG. 8 from the storage unit, and execute the program to execute each part of the voice analyzer 1 and the communication terminal 2. By controlling, the voice analysis method shown in FIG. 8 is executed. Some of the steps included in the voice analysis method shown in FIG. 8 may be omitted, the order between the steps may be changed, and a plurality of steps may be performed in parallel.

SS Voice analysis system 1 Voice analysis device 11 Control unit 111 Reception unit 112 Position setting unit 113 Voice acquisition unit 114 Voice analysis unit 2 Communication terminal 21 Control unit 211 Acquisition unit 212 Transmission unit 3 Sound collector

Claims (9)

A voice analyzer that analyzes the voices emitted by multiple participants around the sound collector.
The sound collector identification information that can identify the sound collector, the participant identification information that can identify the participant, and the position designation information that specifies the position with respect to the sound collector, which are acquired by the communication terminal, are described above. The receiver that receives from the communication terminal and
A position setting unit that sets the position of each of the plurality of participants with respect to the sound collector based on the sound collector identification information, the participant identification information, and the position designation information received by the reception unit.
A sound acquisition unit that acquires the sound from the sound collector, and
A voice analysis unit that analyzes the voice emitted by each of the plurality of participants included in the voice acquired by the voice acquisition unit based on the positions of the plurality of participants set by the position setting unit. ,
A voice analyzer. The receiving unit uses the communication terminal to obtain the sound collecting device identification information acquired by the communication terminal reading a first identification information presenting unit attached to the sound collecting device that can identify the sound collecting device. The voice analyzer according to claim 1, which receives from. The receiving unit communicates the participant identification information acquired by the communication terminal reading a second identification information presenting unit that can identify each of the plurality of participants possessed by the plurality of participants. The voice analyzer according to claim 1 or 2, which is received from a terminal. The voice analyzer according to claim 1 or 2, wherein the receiving unit receives the participant identification information acquired by the communication terminal reading the biometric information of each of the plurality of participants from the communication terminal. .. Based on the position designation information, the position setting unit selects the seat positions of the plurality of participants in the seat arrangement around the sound collecting device set in advance, and the selected seat positions. The voice analyzer according to any one of claims 1 to 4, wherein is set as the position of the participant. The receiving unit receives the order in which the communication terminal acquires the participant identification information of each of the plurality of participants as the position designation information.
The voice analyzer according to claim 5, wherein the position setting unit selects the position of the seat of each of the plurality of participants in the seat arrangement based on the order received by the receiving unit. The receiving unit receives an operation on the communication terminal as the position designation information, and receives the operation.
The voice analyzer according to claim 5, wherein the position setting unit selects the position of the seat of each of the plurality of participants in the seat arrangement based on the operation received by the receiving unit. A voice analysis system including a voice analyzer that analyzes voices emitted by a plurality of participants around the sound collector and a communication terminal capable of communicating with the voice analyzer.
The communication terminal is
An acquisition unit that acquires sound collector identification information that can identify the sound collector, participant identification information that can identify the participant, and position designation information that specifies a position with respect to the sound collector.
A transmission unit that transmits the sound collector identification information, the participant identification information, and the position designation information.
Have,
The voice analyzer is
A receiving unit that receives the sound collecting device identification information, the participant identification information, and the position designation information acquired by the communication terminal from the communication terminal.
A position setting unit that sets the position of each of the plurality of participants with respect to the sound collector based on the sound collector identification information, the participant identification information, and the position designation information received by the reception unit.
A sound acquisition unit that acquires the sound from the sound collector, and
A voice analysis unit that analyzes the voice emitted by each of the plurality of participants included in the voice acquired by the voice acquisition unit based on the positions of the plurality of participants set by the position setting unit. ,
Has a voice analysis system. It is a voice analysis method that analyzes the voices emitted by multiple participants around the sound collector.
The processor runs,
The sound collector identification information that can identify the sound collector, the participant identification information that can identify the participant, and the position designation information that specifies the position with respect to the sound collector, which are acquired by the communication terminal, are described above. Steps to receive from the communication terminal and
A step of setting the position of each of the plurality of participants with respect to the sound collector based on the sound collector identification information, the participant identification information, and the position designation information received by the receiving step.
The step of acquiring the sound from the sound collector, and
A step of analyzing the voice emitted by each of the plurality of participants included in the voice acquired by the acquisition step based on the position of each of the plurality of participants set by the setting step.
A voice analysis method.

Images