JP2018185372A - Information processor, information processing program and building - Google Patents

Abstract

To extract sounds uttered for each user. An information processing apparatus 101 starts buffering information about sounds input to microphones M1 and M2 into buffers B1 and B2 of a storage unit 110. The information processing apparatus 101 identifies the period P when the user of the terminal T speaks and the direction θ from the microphones M1 and M2 to the terminal T based on information transmitted from the terminal T in response to an operation on the terminal T. To do. The information processing apparatus 101 determines the user of the terminal T based on the information about the sound for each of the microphones M1, M2 in the period P among the information about the sound for each of the microphones M1, M2 stored in the storage unit 110 and the direction θ. Extract information about the sound uttered by. The information processing apparatus 101 outputs information related to the extracted sound uttered by the user in association with the identification information of the terminal T. [Selection] Figure 1

Description

  The present invention relates to an information processing apparatus, an information processing program, and a building.

  In recent years, speech input has become familiar and practical due to the development of artificial intelligence technology and the accumulation of a huge conversation database. In addition, far-field speech recognition technology has reached a practical level, and speech recognition at a distance from several meters has become possible.

  As related prior art, for example, the correspondence between the speaker and the spoken language is maintained, the correspondence between the speaker and the spoken language is recorded, and the flow of the dialogue is switched while switching the speaker pair that is currently proceeding with the dialogue. Depending on the language, the language conversion direction is determined. In addition, a portion where the designated section designated by the operator overlaps the utterance section detected from the input voice is detected, and the speaker is determined to be a person other than the operator based on the input voice. In such a case, there is a technique for determining an utterance section including an overlapping portion as a processing section. In addition, each student in the classroom is photographed and the optical flow is used to detect the movement of the student who should be the speaker from the chair and the movement of the mouth. There is a technique for specifying a position and extracting image data of a speaker's face.

JP 2007-322523 A JP 2007-264473 A International Publication No. 2011-013605

  However, in the conventional technology, in an environment where a plurality of people speak, when the direction of the person who speaks from a microphone that collects sound is different, the sound for each person in the section specified by the user can be extracted. It can be difficult. For example, when multiple people utter at the same timing, it is difficult to distinguish and extract the uttered sound for each person.

  In one aspect, the present invention aims to extract sounds uttered for each user.

  In one embodiment, for each microphone included in the plurality of microphones, a storage unit that stores information related to sound input to the microphone, and a signal transmitted from the terminal in response to an operation on the terminal. And a specific unit that specifies a period during which the user of the terminal speaks, a direction from the plurality of microphones to the terminal, and a sound for each microphone of the period stored in the storage unit. An information processing apparatus is provided that includes an extraction unit that extracts information related to the sound uttered by the user based on the information and the direction.

  According to one aspect of the present invention, it is possible to extract a sound uttered for each user.

FIG. 1 is an explanatory diagram of an example of the information processing apparatus 101 according to the embodiment. FIG. 2 is an explanatory diagram illustrating a system configuration example of the information processing system 100. FIG. 3 is an explanatory diagram illustrating a hardware configuration example of the information processing apparatus 101. FIG. 4 is a block diagram illustrating a hardware configuration example of the terminal Ti. FIG. 5 is an explanatory diagram showing an example of the contents stored in the utterance section table 220. FIG. 6 is a block diagram illustrating a functional configuration example of the information processing apparatus 101. FIG. 7 is an explanatory diagram (part 1) illustrating an example of a positional relationship between a plurality of microphones M and a user. FIG. 8 is an explanatory diagram showing a specific example of information related to sound stored in the buffers B1 and B2. FIG. 9 is an explanatory diagram illustrating an example of sound source recognition in the height direction. FIG. 10 is an explanatory diagram illustrating a second hardware configuration example of the information processing apparatus 101. FIG. 11 is an explanatory diagram (part 2) illustrating an example of the positional relationship between the plurality of microphones M and the user. FIG. 12 is a flowchart illustrating an example of an information processing procedure of the information processing apparatus 101.

  Exemplary embodiments of an information processing apparatus, an information processing program, and a building according to the present invention will be described below in detail with reference to the drawings.

(Embodiment)
FIG. 1 is an explanatory diagram of an example of the information processing apparatus 101 according to the embodiment. In FIG. 1, the information processing system 100 includes an information processing apparatus 101 and a terminal T. The information processing apparatus 101 is a computer that includes a plurality of microphones M (for example, microphones M1 and M2) and a storage unit 110, and extracts information related to sounds uttered by the user of the terminal T.

  The microphone M is a device that converts sound into an electrical signal. For example, the plurality of microphones M are installed at substantially the same height. The distance between the microphones M is, for example, about several centimeters to several tens of centimeters. The number of microphones M may be two or more, and may be three or four.

  The storage unit 110 stores information regarding the sound input to the microphone M for each microphone M included in the plurality of microphones M. If it demonstrates in detail, the memory | storage part 110 will have the buffer B (for example, buffer B1, B2) which memorize | stores the information regarding the sound input into the microphone M for every microphone M. FIG. Each buffer B is connected to each microphone M independently of the other buffers B. The information regarding the sound is, for example, time-series data (digital signal) indicating a time-series change in sound pressure or frequency of the sound input to the microphone M.

  The terminal T is a computer operated by a user. For example, the terminal T is attached to the user or installed in a place where the user can easily operate. Specifically, for example, the terminal T is a wearable terminal such as a ring (ring) type, a wristband type, a pendant type, or a batch type. The terminal T may be a switch type device provided on a wall. Furthermore, the terminal T may be a device such as a smartphone or a tablet.

  In the information processing system 100, an environment where a plurality of people speak is assumed, but in FIG. 1, only one user is displayed for convenience of explanation. Further, when a plurality of people speak, the terminal T is provided for each user who speaks.

  Here, in an environment where a plurality of people speak, a plurality of people may speak at the same timing. For example, when talking in groups of several people, people from different groups may speak at the same time. In such a case, in the prior art, it is difficult to distinguish and extract sounds for each person, and as a result, there is a problem in that the voice recognition accuracy is lowered.

  In addition, if voice recognition is continuously performed in a place where conversations are made on a daily basis, conversations that are not intended by the speaker may be recognized, resulting in privacy and security problems. Furthermore, if the speaker does not intend to recognize speech, there is a tendency to use unclear contexts or unrecognizable new words, and it is difficult to ensure speech recognition accuracy.

  Therefore, in the present embodiment, an information processing method that can extract a sound uttered for each user even in an environment where a plurality of people speak is described. Hereinafter, a processing example of the information processing apparatus 101 will be described.

  (1) The information processing apparatus 101 starts buffering information related to sound input to the microphones M1 and M2 in the buffers B1 and B2 of the storage unit 110. It should be noted that it is possible to arbitrarily design how long the information is stored in each of the buffers B1 and B2. For example, the latest information having a time length of about several tens of seconds is stored in each of the buffers B1 and B2.

  (2) The information processing apparatus 101 determines, based on information transmitted from the terminal T in response to an operation on the terminal T, a period P in which the user of the terminal T speaks and a direction θ from the microphones M1 and M2 to the terminal T. And specify. Here, in the information processing system 100, the user operates the terminal T in order to specify his / her speech section.

  For example, the user performs the first operation on the terminal T at the timing of starting the utterance. In this case, the first information indicating that the first operation has been performed is transmitted from the terminal T. In addition, the user performs the second operation on the terminal T at the timing of ending the utterance. In this case, second information indicating that the second operation has been performed is transmitted from the terminal T. Then, the information processing apparatus 101 determines, based on the first and second information transmitted from the terminal T, the period from the time point t1 when the first operation is performed to the time point t2 when the second operation is performed. This is specified as a period P during which the user speaks.

  Note that the information indicating the time points t1 and t2 when the first and second operations are performed may be included in the first and second information, respectively, for example. Further, for example, the information processing apparatus 101 may specify the time points at which the first and second information are received as time points t1 and t2 at which the first and second operations are performed (so-called time stamps). ).

  The direction θ is the direction of the terminal T as viewed from the microphones M1 and M2. The direction θ is, for example, an axis passing through the midpoint between the microphones M1 and M2 and a vector from the midpoint between the microphones M1 and M2 toward the terminal T in the same horizontal plane (a coordinate system viewed from above). Expressed by angle. In other words, the direction θ corresponds to the direction of the user of the terminal T viewed from the microphones M1 and M2.

  For example, the information processing apparatus 101 includes a wireless signal (for example, first information) transmitted from the terminal T by a plurality of receivers (for example, the beacon receiver 201 illustrated in FIG. 2) installed around the terminal T. The position of the terminal T is estimated from the received signal strength when is received. Then, the information processing apparatus 101 specifies the direction θ based on the estimated position of the terminal T and the installation positions of the microphones M1 and M2.

  Note that the information transmitted from the terminal T in response to an operation on the terminal T may include, for example, the position information of the terminal T. In this case, for example, the information processing apparatus 101 estimates the position of the terminal T from the position information included in the first information, and specifies the direction θ from the estimated position of the terminal T and the installation positions of the microphones M1 and M2. be able to.

  (3) The information processing apparatus 101 is a terminal based on the information regarding the sound for each of the microphones M1, M2 in the period P among the information regarding the sound for each of the microphones M1, M2 stored in the storage unit 110 and the direction θ. Information related to the sound uttered by the user T is extracted.

  Specifically, for example, the information processing apparatus 101 reads information related to the sounds of the microphones M1 and M2 in the period P from the buffers B1 and B2 of the storage unit 110, respectively. Then, the information processing apparatus 101 extracts the information related to the sound uttered by the user by performing the beamform process in the direction θ based on the information related to the sound of each of the microphones M1 and M2 in the read period P.

  Here, the beamform process is a process of controlling directivity using a plurality of microphones M, for example, a process of enhancing the sensitivity in a specific direction and enhancing a signal coming from the specific direction. According to the beamform processing, the information about the sound of each microphone M1, M2 stored in each buffer B1, B2 is used to emphasize the sound signal coming from the direction θ, and the sound uttered by the user of the terminal T Information can be extracted.

  A detailed description of the beamform process is omitted because it is a well-known technique. As for the specific processing contents of the beamform processing, for example, “Hozayama Osamu et al.,“ Knowledge Forest ”, IEICE, 2012, Group 2, Chapter 6, Chapter-2 can be referred to.

  (4) The information processing apparatus 101 outputs information on the extracted sound uttered by the user in association with the identification information of the terminal T. The identification information of the terminal T is information that uniquely identifies the terminal T, and is included in information transmitted from the terminal T in response to an operation on the terminal T.

  Specifically, for example, the information processing apparatus 101 transmits information related to the sound uttered by the user of the terminal T in association with the identification information of the terminal T to the external apparatus that executes the voice recognition process. Good. As a result, in the external device, the speech recognition process is performed on the information related to the sound uttered by the user of the terminal T.

  In this way, according to the information processing apparatus 101, the sound uttered by the user of the terminal T by emphasizing the voice signal that has arrived from the direction θ of the terminal T (the direction of the user of the terminal T viewed from the microphones M1 and M2). Can be extracted. Specifically, for example, the information processing apparatus 101 uses the information on the sound for each of the microphones M1 and M2 stored in the buffers B1 and B2, and uses the information about the sound signal that has been emphasized from each direction θ, respectively. Can be generated. Further, the user of the terminal T can be specified from the identification information of the terminal T. For this reason, when the direction of the person who speaks from the microphone that collects the sound is different, the sound for each person in the section specified by the user can be extracted. For example, even when a plurality of people utter at the same timing, it is possible to accurately extract the uttered sound for each user. In addition, since the user can designate the utterance section, it is possible to prevent a conversation unintended by the user from being recognized and leaked. Furthermore, when the user designates the utterance period, the vocabulary that the user wants to recognize can be clarified, and consequently, the voice recognition accuracy can be prevented from being lowered.

(System configuration example of the information processing system 100)
Next, a system configuration example of the information processing system 100 will be described.

  FIG. 2 is an explanatory diagram illustrating a system configuration example of the information processing system 100. In FIG. 2, the information processing system 100 includes an information processing apparatus 101, a plurality of beacon receivers 201, and terminals T1 to Tn. In the information processing system 100, the information processing apparatus 101 and the plurality of beacon receivers 201 are connected via a wired or wireless network 210. The network 210 includes, for example, a LAN, a WAN (Wide Area Network), the Internet, and the like.

  In the following description, an arbitrary terminal among the terminals T1 to Tn may be expressed as “terminal Ti” (i = 1, 2,..., N). The terminal T illustrated in FIG. 1 corresponds to the terminal Ti, for example.

  The information processing apparatus 101 has an utterance section table 220 and extracts information related to the sound uttered by the user of the terminal Ti. The contents stored in the utterance section table 220 will be described later with reference to FIG. The terminal Ti is, for example, a ring (ring) type wearable terminal used by a user of the information processing system 100. The terminal Ti may be installed in a place where the user can easily operate.

  The beacon receiver 201 is a computer that receives a beacon signal transmitted from the terminal Ti. The plurality of beacon receivers 201 are respectively installed at different positions in the space R. The space R is a space where the user of the terminal Ti exists. For example, the space R is a building, a room, a booth, or the like that is used when talking by a plurality of people. The building has a floor and a wall that rises from the floor and surrounds the room (space R). Microphones M1 and M2 are provided on the wall of the building.

  The beacon receiver 201 measures the received signal strength of the beacon signal received from the terminal Ti, adds the measured received signal strength to the beacon signal received from the terminal Ti, and transmits the signal to the information processing apparatus 101. The received signal strength is an index value indicating the strength of the beacon signal received by the beacon receiver 201. As the received signal strength, for example, an RSSI (Received Signal Strength Indicator) value can be used. The unit of the RSSI value is, for example, [dBm].

(Hardware configuration example of information processing apparatus 101)
FIG. 3 is an explanatory diagram illustrating a hardware configuration example of the information processing apparatus 101. In FIG. 3, the information processing apparatus 101 includes a CPU (Central Processing Unit) 301, a memory 302, an I / F (Interface) 303, a beacon receiving unit 304, a disk drive 305, a disk 306, and a sound collecting unit. 307. Each component is connected by a bus 300.

  Here, the CPU 301 governs overall control of the information processing apparatus 101. The memory 302 includes, for example, a ROM (Read Only Memory), a RAM (Random Access Memory), and a flash ROM. Specifically, for example, a flash ROM or ROM stores various programs, and a RAM is used as a work area for the CPU 301. The program stored in the memory 302 is loaded into the CPU 301 to cause the CPU 301 to execute the coded process.

  The I / F 303 is connected to the network 210 (see FIG. 2) through a communication line, and is connected to another computer (for example, the beacon receiver 201 shown in FIG. 2) via the network 210. The I / F 303 controls an internal interface with the network 210 and controls input / output of data from other computers. For example, a modem or a NIC (Network Interface Card) can be adopted as the I / F 303.

  The beacon receiving unit 304 receives a beacon signal (wireless signal). Specifically, for example, the beacon receiving unit 304 includes an antenna that receives a beacon signal, and a signal processing unit that converts an analog signal received by the antenna into a digital signal and outputs the digital signal to the bus 300.

  The disk drive 305 controls reading / writing of data with respect to the disk 306 according to the control of the CPU 301. The disk 306 stores data written under the control of the disk drive 305. Examples of the disk 306 include a magnetic disk and an optical disk.

  The sound collection unit 307 includes microphones M1 and M2 and buffers B1 and B2. The microphones M1 and M2 are devices that convert sound into an electrical signal. The microphones M1 and M2 are installed at substantially the same height in the space R. The buffer B1 is connected to the microphone M1 and stores information related to sound input to the microphone M1. The buffer B2 is connected to the microphone M2 and stores information related to the sound input to the microphone M2.

  The information processing apparatus 101 may not include, for example, the disk drive 305 or the disk 306 among the above-described components. The information processing apparatus 101 includes an input device (for example, a keyboard, a mouse, an input pad, etc.), an output device (for example, a display, a speaker, etc.), an SSD (Solid State Drive), and the like in addition to the components described above. You may decide.

(Example of hardware configuration of terminal Ti)
FIG. 4 is a block diagram illustrating a hardware configuration example of the terminal Ti. In FIG. 4, the terminal Ti includes a CPU 401, a memory 402, an operation button 403, an LED (Light Emitting Diode) lamp 404, and a beacon transmission unit 405. Each component is connected by a bus 400.

  The CPU 401 controls the entire terminal Ti. The memory 402 includes, for example, a ROM, a RAM, and the like. Specifically, for example, the ROM stores various programs, and the RAM is used as a work area for the CPU 401. The program stored in the memory 402 is loaded on the CPU 401 to cause the CPU 401 to execute the coded process.

  The operation button 403 is an input device operated to specify the user's utterance section of the terminal Ti. Specifically, for example, the user of the terminal Ti performs an operation of turning on the operation button 403 at the timing of starting the utterance, and performs an operation of turning off the operation button 403 at the timing of ending the utterance.

  More specifically, for example, the user of the terminal Ti presses the operation button 403 with a finger (ON operation) at the timing of starting the utterance, and continues to press the operation button 403 until the utterance ends. Then, the user of the terminal Ti releases his / her finger from the operation button 403 at the timing of ending the speech (OFF operation).

  The LED lamp 404 is a lamp that is turned on in response to the operation of the operation button 403. Specifically, for example, the LED lamp 404 is turned on when the operation button 403 is turned on and turned off when the operation button 403 is turned off. According to the LED lamp 404, the user of the terminal Ti can check the ON / OFF state of the operation button 403.

  The beacon transmission unit 405 transmits a beacon signal (wireless signal) in response to the operation of the operation button 403. Specifically, for example, the beacon transmission unit 405 includes a signal processing unit that converts a digital signal output to the bus 400 into an analog signal and outputs the analog signal, and an antenna that transmits a radio signal output from the signal processing unit. And having. The beacon transmission unit 405 transmits a beacon signal by, for example, BLE (Bluetooth Low Energy) communication. Bluetooth is a registered trademark.

  More specifically, for example, when the operation button 403 is turned on, the beacon transmission unit 405 transmits a beacon signal bs including the terminal ID “Ti” of the terminal Ti and the operation type “ON”. The operation type “ON” indicates an operation for turning on the operation button 403. Further, when the operation button 403 is turned off, the beacon transmission unit 405 transmits a beacon signal bs including the terminal ID “Ti” of the terminal Ti and the operation type “OFF”. The operation type “OFF” indicates an operation for turning off the operation button 403. The beacon signal bs may include information indicating the date and time when the operation button 403 is operated, for example.

  The terminal Ti is driven by, for example, a primary battery or a secondary battery. In addition to the components described above, the terminal Ti includes, for example, a public network I / F connected to the network 210 (see FIG. 2) through a communication line, a GPS (Global Positioning System) unit, and the like. Also good. When the terminal Ti has a GPS unit, the beacon transmission unit 405 may transmit the beacon signal bs including the position information output by the GPS unit.

(Stored contents of the utterance section table 220)
Next, the contents stored in the utterance section table 220 of the information processing apparatus 101 will be described. The utterance section table 220 is realized by a storage device such as the memory 302 and the disk 306 shown in FIG.

  FIG. 5 is an explanatory diagram showing an example of the contents stored in the utterance section table 220. In FIG. 5, the utterance section table 220 includes fields of terminal ID, direction, ON time, and OFF time. By setting information in each field, utterance section information (for example, utterance section information 500-1, 500). -2) is stored as a record.

  The terminal ID is identification information that uniquely identifies the terminal Ti. As the terminal ID, for example, the MAC (Media Access Control) address of the terminal Ti can be used. In the terminal ID field, for example, the terminal ID of the terminal Ti paired with the information processing apparatus 101 may be set in advance.

  The direction is the direction of the terminal Ti viewed from the microphones M1 and M2. The direction is, for example, an axis passing through the midpoint between the microphones M1 and M2 (for example, an axis orthogonal to a line segment connecting the microphones M1 and M2) on the same horizontal plane (coordinate system when the space R is viewed from above), This is expressed by an angle θ formed by a vector from the midpoint between the microphones M1 and M2 toward the terminal Ti.

The ON time indicates the date and time when the operation of turning on the operation button 403 (see FIG. 4) of the terminal Ti is performed. The OFF time indicates the date and time when the operation for turning off the operation button 403 of the terminal Ti is performed. For example, the utterance section information 500-1 indicates the terminal ID “T1”, the direction “θ1”, the ON time “t ₁₁ ”, and the OFF time “t ₁₂ ”.

(Functional configuration example of the information processing apparatus 101)
FIG. 6 is a block diagram illustrating a functional configuration example of the information processing apparatus 101. In FIG. 5, the information processing apparatus 101 includes an acquisition unit 601, a specification unit 602, an extraction unit 603, a voice recognition unit 604, and an output unit 605. The acquisition unit 601 to the output unit 605 are functions serving as a control unit. Specifically, for example, by causing the CPU 301 to execute a program stored in a storage device such as the memory 302 and the disk 306 illustrated in FIG. Alternatively, the function is realized by the I / F 303 and the beacon receiving unit 304. The processing result of each functional unit is stored in the storage device of the memory 302 and the disk 306, for example.

  The acquisition unit 601 acquires information transmitted from the terminal Ti in response to an operation on the terminal Ti. Here, the information transmitted from the terminal Ti is, for example, the beacon signal bs transmitted from the beacon transmission unit 405 of the terminal Ti in response to the operation of the operation button 403 of the terminal Ti. The beacon signal bs includes, for example, the terminal ID of the terminal Ti and the operation type. The operation type is information indicating the type of operation that triggered the transmission of the beacon signal bs. As the operation type, for example, there are an operation type “ON” and an operation type “OFF”. The operation type “ON” indicates an operation for turning on the operation button 403. The operation type “OFF” indicates an operation for turning off the operation button 403.

  Specifically, for example, the acquisition unit 601 may acquire the beacon signal bs by receiving the beacon signal bs transmitted from the terminal Ti by the beacon reception unit 304 illustrated in FIG. Further, the acquisition unit 601 may acquire the beacon signal bs by receiving the beacon signal bs transmitted from the terminal Ti from the beacon receiver 201 by the I / F 303 illustrated in FIG. 3. The beacon signal bs received from the beacon receiver 201 includes the received signal strength (for example, RSSI value) of the beacon signal bs measured by the beacon receiver 201.

  Based on information transmitted from the terminal Ti in response to an operation on the terminal Ti acquired by the acquiring unit 601, the specifying unit 602 speaks a period P (hereinafter referred to as “speaking section P”) of the user of the terminal Ti. ). Specifically, for example, the identifying unit 602 acquires the beacon signal bs in association with the terminal ID included in the beacon signal bs in response to the acquisition of the beacon signal bs including the operation type “ON” ( The received time is registered in the utterance section table 220 (see FIG. 5) as the ON time.

For example, when the beacon signal bs including the terminal ID “T1” and the operation type “ON” is acquired, the specifying unit 602 associates the terminal ID “T1” with the beacon signal bs and receives the time t _11. Is registered in the utterance section table 220 as the ON time. As a result, the utterance section information 500-1 is registered in the utterance section table 220 as a new record. At this time, the direction and OFF time field of the utterance section information 500-1 are “Null”.

Next, in response to the acquisition of the beacon signal bs including the operation type “OFF”, the identifying unit 602 uses the beacon signal bs as the OFF time of the speech section information corresponding to the terminal ID included in the beacon signal bs. Set the time of acquisition (reception). For example, when the beacon signal bs including the terminal ID “T1” and the operation type “OFF” is acquired, the specifying unit 602 displays the OFF time field of the utterance section information 500-1 corresponding to the terminal ID “T1”. setting the time t ₁₂ to the beacon signal bs is received.

Accordingly, the identifying unit 602 may a period from ON time t ₁₁ speech period information 500-1 to OFF time t _12, identified as speech segment P uttered by a user of the terminal T1.

  When the information processing apparatus 101 receives the beacon signal bs via the beacon receiver 201, the information processing apparatus 101 receives the beacon signal bs including the same terminal ID and operation type almost simultaneously from the plurality of beacon receivers 201. become. In this case, for example, the specifying unit 602 may perform registration in the utterance section table 220 according to the beacon signal bs acquired first.

  Further, the specifying unit 602 specifies the direction θ from the microphones M1 and M2 to the terminal Ti based on information transmitted from the terminal Ti in response to an operation on the terminal Ti acquired by the acquiring unit 601. Specifically, for example, the specifying unit 602 estimates the position of the terminal Ti based on the beacon signal bs including the terminal ID of the terminal Ti received from the plurality of beacon receivers 201.

  More specifically, for example, when the beacon signal bs is received from the beacon receiver 201, the specifying unit 602 receives another beacon signal bs including the same terminal ID and operation type as the beacon signal bs. It waits for a certain time to receive from the machine 201. The certain time is, for example, about 1 second. Next, the specifying unit 602 estimates the position of the terminal Ti in the space R based on the RSSI value of the beacon signal bs received from each beacon receiver 201 and the installation position of each beacon receiver 201. Information indicating the installation position of each beacon receiver 201 is stored in a storage device such as the memory 302 and the disk 306, for example.

  In addition, since the technique which estimates the position of terminal Ti from the received signal strength (RSSI value) when each beacon receiver 201 receives the beacon signal of terminal Ti is a well-known technique, description is abbreviate | omitted here. Also, any existing technique may be used as a technique for estimating the position of the terminal Ti from the received signal strength of the beacon signal. When the terminal Ti has a GPS unit, the information processing apparatus 101 may receive position information output from the terminal Ti by the GPS unit. In this case, the specifying unit 602 may estimate the position of the terminal Ti based on the position information received from the terminal Ti.

  Then, the specifying unit 602 specifies the direction θ from the microphones M1 and M2 to the terminal Ti based on the estimated position of the terminal Ti and the installation positions of the microphones M1 and M2 (sound collection unit 307). The direction θ is, for example, an axis orthogonal to the line segment passing through the midpoint of the line segment connecting the microphones M1 and M2 on the same horizontal plane and a vector from the midpoint between the microphones M1 and M2 toward the terminal Ti. Corresponds to the angle. That is, the direction θ corresponds to the direction of the user of the terminal Ti viewed from the microphones M1 and M2. Information indicating the installation positions of the microphones M1 and M2 (sound collection unit 307) is stored in a storage device such as the memory 302 and the disk 306, for example.

  The identified direction θ is registered in the utterance section table 220 (see FIG. 5) in association with the terminal ID included in the beacon signal bs. For example, when the direction θ1 from the microphones M1 and M2 to the terminal T1 is specified, the specifying unit 602 sets the direction θ1 in the direction field of the speech section information 500-1 corresponding to the terminal ID “T1”.

  Based on the information about the sound for each of the microphones M1, M2 in the utterance section P among the information about the sound for each of the microphones M1, M2 stored in the buffers B1, B2, the extraction unit 603 is based on the direction θ and the terminal Ti. Information on the sound uttered by the user is extracted. Here, the information regarding sound is, for example, time-series data (digital signal) indicating a time-series change in sound pressure.

  Specifically, for example, first, the extraction unit 603 refers to the utterance interval table 220 to identify the utterance interval P of the terminal Ti. Next, the extraction part 603 reads the information regarding the sound for every microphone M1, M2 of the specified speech section P from each buffer B1, B2. Next, the extraction unit 603 refers to the utterance section table 220 and identifies the direction θ of the terminal Ti. And the extraction part 603 performs the beamform process to the specified direction (theta) based on the information regarding the sound of each microphone M1, M2 of the read speech section P, and the information regarding the sound uttered by the user of the terminal Ti Is extracted (generated).

As an example, taking the utterance section information 500-1 in the utterance section table 220 shown in FIG. 5 as an example, the extraction unit 603 first refers to the utterance section information 500-1 and refers to the utterance section P of the terminal T1. identifying the (t ₁₁ ~t _12). Next, the extraction part 603 reads the information regarding the sound for every microphone M1, M2 of the specified speech section P from each buffer B1, B2. Next, the extraction unit 603 refers to the utterance section table 220 and identifies the direction θ1 of the terminal T1. Then, the extraction unit 603 extracts information related to the sound uttered by the user of the terminal T1 by performing beamforming processing in the direction θ1 based on the information related to the sound of each of the microphones M1 and M2 in the utterance section P.

  Note that the information processing apparatus 101 may perform a noise canceling process on information regarding the sound uttered by the user of the terminal T1 extracted by the extraction unit 603. The noise canceling process is a process for removing the influence of environmental noise and reflected sound, for example.

  The output unit 605 outputs information on the sound uttered by the user of the terminal Ti extracted by the extraction unit 603 in association with the terminal ID of the terminal Ti. Examples of the output format of the output unit 605 include storage in a storage device such as the memory 302 and the disk 306, transmission to an external device by the I / F 303, display on a display (not shown), and the like.

  Specifically, for example, the output unit 605 may transmit information related to the sound uttered by the user of the terminal Ti in association with the terminal ID of the terminal Ti to the external device that executes the voice recognition process. . As a result, in the external device, a speech recognition process is performed on information related to the sound uttered by the user of the terminal Ti.

  The voice recognition unit 604 performs voice recognition on the information related to the sound uttered by the user of the terminal Ti extracted by the extraction unit 603. Here, the speech recognition is, for example, a process of converting information about the sound uttered by the user into a character string. Note that any existing technology may be used as the speech recognition technology.

  Specifically, for example, the speech recognition unit 604 recognizes information related to sound uttered by the user of the terminal Ti using an acoustic model, a language model, a word dictionary, and the like. The acoustic model represents what frequency characteristic the phoneme to be recognized has. The language model expresses restrictions on how phonemes are arranged.

  The output unit 605 outputs the recognition result recognized by the voice recognition unit 604 in association with the terminal ID of the terminal Ti. Specifically, for example, the output unit 605 may store the recognition result recognized by the voice recognition unit 604 in the disk 306 in association with the terminal ID of the terminal Ti. Further, the output unit 605 may transmit the recognition result recognized by the voice recognition unit 604 to an electronic device capable of voice input.

  Note that the information processing apparatus 101 may not include the voice recognition unit 604. In addition, the terminal Ti transmits event information including the terminal ID of the terminal Ti and the operation type to the information processing apparatus 101 through a public network I / F (not illustrated) in response to the operation of the operation button 403. Good. In this case, the specifying unit 602 may specify the utterance section P uttered by the user of the terminal Ti based on event information transmitted from the terminal Ti in response to an operation on the terminal Ti. Further, the functional unit of the information processing apparatus 101 may be realized by hardware, for example. Specifically, for example, the functional unit includes AND that is a logical product circuit, INVERTER that is a negative logical circuit, OR that is a logical sum circuit, NOR that is a logical sum negation circuit, and FF (Flip Flop) that is a latch circuit. Or the like.

(Extraction example of information about sound uttered by user of terminal Ti)
Next, an example of extracting information related to the sound uttered by the user of the terminal Ti will be described with reference to FIGS.

  FIG. 7 is an explanatory diagram (part 1) illustrating an example of a positional relationship between a plurality of microphones M and a user. In FIG. 7, a user A of the terminal T1 and a user B of the terminal T2 exist in the space R. The space R is, for example, a conference room or a meeting room. Four beacon receivers 201 are installed at different positions in the space R, and microphones M1 and M2 are installed on the right side of the space R.

  Here, assuming that the user A of the terminal T1 and the user B of the terminal T2 speak, information on sounds input to the microphones M1 and M2 will be described. Also, the direction θ from the microphones M1, M2 to the terminal T1 is “direction θ1,” and the direction θ from the microphones M1, M2 to the terminal T2 is “direction θ2.” That is, the direction θ1 corresponds to the direction of the user A of the terminal T1 viewed from the microphones M1 and M2. The direction θ2 corresponds to the direction of the user B of the terminal T2 as viewed from the microphones M1 and M2. However, θ1 indicates a clockwise angle from the reference axis (dotted line in FIG. 7), and θ2 indicates a counterclockwise angle from the reference axis.

  FIG. 8 is an explanatory diagram showing a specific example of information related to sound stored in the buffers B1 and B2. In FIG. 8, sound information 810 is information related to the sound input to the microphone M1 stored in the buffer B1. The sound information 820 is information related to the sound input to the microphone M2 stored in the buffer B2.

Here, the user A, the operation of the operation button 403 of the terminal T1 is turned ON at time t ₁₁ is performed, operation of the operation button 403 of the terminal T1 to OFF at time t ₁₂ is assumed when done. Further, the user B, the operation of the operation button 403 of the terminal T2 is turned ON at time t ₂₁ is performed, operation of the operation button 403 of the terminal T2 is turned OFF at time t ₂₂ is assumed when done.

In this case, the extraction unit 603, from the buffer B1, B2, reads the information about the sound of each microphone M1, M2 from time t ₁₁ to time t ₁₂ (in FIG. 8, a dotted line frame 830). Then, the extraction unit 603 performs beamform processing in the direction θ1 (the direction of the user A) based on the read information regarding the sounds of the microphones M1 and M2. As a result, it is possible to extract information related to the sound uttered by the user A of the terminal T1 while emphasizing the signal arriving from the direction θ1.

The extraction unit 603, from the buffer B1, B2, reads the information about the sound of each microphone M1, M2 from time t ₂₁ to time t ₂₂ (in FIG. 8, a dotted line frame 840). Then, the extraction unit 603 performs beamform processing in the direction θ2 (the direction of the user B) based on the read information regarding the sounds of the microphones M1 and M2. As a result, it is possible to extract information related to the sound uttered by the user B of the terminal T2 while emphasizing the signal arriving from the direction θ2.

Here, the utterance section P of user A (from time t ₁₁ to time t ₁₂ ) and the utterance section P of user B (from time t ₂₁ to time t ₂₂ ) partially overlap. That is, there is a section where the user A and the user B talk at the same time (a portion where the dotted frame 830 and the dotted frame 840 overlap). Even in such a case, the information processing apparatus 101 uses the information related to the sound stored in each of the buffers B1 and B2, and emphasizes the sound signal that has arrived from the directions θ1 and θ2 of the users A and B. Information can be generated to extract sounds uttered by the users A and B.

(Sound source recognition in the height direction)
Next, sound source recognition in the height direction will be described with reference to FIG. As acoustic models used for speech recognition, there are models according to the degree of growth of users as sound sources, such as an acoustic model for adults and an acoustic model for children. Therefore, for example, if it can be determined whether the user of the terminal Ti is an adult or a child, it is possible to determine which acoustic model should be used for speech recognition.

  Therefore, the information processing apparatus 101 has a plurality of microphones M ′ installed at different positions in the space R in addition to the plurality of microphones M (for example, the microphones M1 and M2) described above. Good. Information regarding the sound input to each of the plurality of microphones M ′ is stored in, for example, the memory 302 and the disk 306. However, any one of the plurality of microphones M may be used as one of the microphones M ′.

  FIG. 9 is an explanatory diagram illustrating an example of sound source recognition in the height direction. In FIG. 9, microphones M′a and M′b are installed on a wall 901 in the space R so as to have different heights. The distance between the microphones M′a and M′b is, for example, about several centimeters to several tens of centimeters.

  Here, the midpoint between the microphones M′a and M′b is a height of “130 cm” from the floor 902. However, the height “130 cm” is an example and can be arbitrarily changed. In addition, the terminal Ti is installed at a height (for example, about 100 cm) on the wall 901 that can be operated by a child. Note that another terminal T different from the terminal Ti is also installed at a position substantially the same height as the terminal Ti.

  In this case, the voice recognition unit 604 determines whether or not the height of the user of the terminal Ti is equal to or higher than a predetermined height K based on information about sound input to the microphones M′a and M′b. You may decide. The predetermined height K corresponds to the height at which the microphones M'a and M'b are installed. In the example of FIG. 9, the predetermined height K is, for example, “130 cm”.

  However, only the user of the terminal Ti speaks when recognizing the sound source in the height direction. At this time, the user of the terminal Ti operates the operation button 403 for designating the utterance section. As a result, the beacon signal bs of the terminal Ti is transmitted, and the information processing apparatus 101 can specify the terminal Ti as a sound source from the beacon signal bs.

  Specifically, for example, the speech recognition unit 604 determines whether the microphones M′a and M′b are based on the time difference between sounds input to the microphones M′a and M′b (difference in propagation delay time from the sound source). Determine whether the sound source is in the upward or downward direction. Here, when the sound source is in the upward direction, the voice recognition unit 604 determines that the height of the user of the terminal Ti is equal to or higher than the predetermined height K. On the other hand, when the sound source is in the downward direction, the voice recognition unit 604 determines that the height of the user of the terminal Ti is less than the predetermined height K.

  Based on the determination result, the voice recognition unit 604 recognizes the information related to the sound uttered by the user of the terminal Ti extracted by the extraction unit 603. Specifically, for example, when the height of the user of the terminal Ti is equal to or higher than a predetermined height K, the voice recognition unit 604 performs voice recognition using an adult acoustic model ((9-1 in FIG. 9). )). On the other hand, when the height of the user of the terminal Ti is less than the predetermined height K, the voice recognition unit 604 performs voice recognition using an acoustic model for children ((9-2) in FIG. 9).

  Thereby, it is possible to determine whether the user of the terminal Ti is an adult or a child, and to set an acoustic model used for speech recognition, thereby improving speech recognition accuracy. In addition, by installing the microphones M′a, M′b and the terminal Ti on the wall 901 in the positional relationship as shown in FIG. 9, from the microphones M′a, M′b to the sound source (user of the terminal Ti). The distance is fixed and the estimation accuracy of the sound source direction can be improved.

(Locating the user of the terminal Ti)
Next, the location of the user of the terminal Ti will be described with reference to FIGS. First, a second hardware configuration example of the information processing apparatus 101 will be described with reference to FIG. However, the description of the same components as those described in FIG. 3 is omitted.

  FIG. 10 is an explanatory diagram illustrating a second hardware configuration example of the information processing apparatus 101. 10, the information processing apparatus 101 includes a CPU 301, a memory 302, an I / F 303, a beacon receiving unit 304, a disk drive 305, a disk 306, a sound collecting unit 307, and a second sound collecting unit 1001. And having. Each component is connected by a bus 300.

  The second sound collection unit 1001 includes microphones M3 and M4 and buffers B3 and B4. The microphones M3 and M4 are devices that convert sound into an electrical signal. The microphones M3 and M4 are installed at positions that have substantially the same height in the space R and are different from the microphones M1 and M2.

  The buffer B3 is connected to the microphone M3 and stores information related to sound input to the microphone M3. The buffer B4 is connected to the microphone M4 and stores information related to the sound input to the microphone M4. The second sound collection unit 1001 may include three or more microphones M and a buffer B connected to each of the three or more microphones M.

  FIG. 11 is an explanatory diagram (part 2) illustrating an example of the positional relationship between the plurality of microphones M and the user. In FIG. 11, it is assumed that the user A of the terminal T1 exists in the space R. Further, microphones M1 and M2 are installed on the left side of the space R, and microphones M3 and M4 are installed on the front side of the space R.

  Here, it is assumed that a coordinate system of the origin O composed of the X axis and the Y axis is set. The X axis and the Y axis are parallel to the horizontal plane. That is, the coordinate system composed of the X axis and the Y axis is a coordinate system when the space R is viewed from above. Further, the installation positions of the microphones M1 and M2 are “(X, Y) = (0, H)”, and the installation positions of the microphones M3 and M4 are “(X, Y) = (W, 0)”.

In this case, the specifying unit 602 specifies the direction θ _a from the microphones M1 and M2 to the terminal Ti based on information regarding the sound input to the microphones M1 and M2. However, only the user of the terminal Ti speaks in specifying the position of the user of the terminal Ti. At this time, the user of the terminal Ti operates the operation button 403 for designating the utterance section. As a result, the beacon signal bs of the terminal Ti is transmitted, and the information processing apparatus 101 can specify the terminal Ti as a sound source from the beacon signal bs.

Further, the identifying unit 602, based on the information about the sound input to the microphone M3, M4, respectively, to identify the direction theta _b from the microphone M3, M4 to the terminal Ti. Then, the extracting unit 603 specifies the position of the terminal Ti based on the directions θ _a and θ _b specified by the specifying unit 602, the installation positions of the microphones M1 and M2, and the installation positions of the microphones M3 and M4. .

  Specifically, for example, the extraction unit 603 uses the following formulas (1) and (2) based on the simultaneous input characteristics to the microphones M1 and M2 and the microphones M3 and M4 (the same utterance is input almost simultaneously). Thus, the position (x, y) of the terminal Ti can be specified.

x = (tan θ _b · H + W) / (1−tan θ _a · tan θ _b ) (1)
y = {tan θ _a (tan θ _b · H + W) + H} / (1−tan θ _a · tan θ _b )
... (2)

  Thereby, the position of the user of the terminal Ti in the space R can be specified.

  Then, the extraction unit 603 stores the information stored in either the sound collection unit 307 (buffers B1 and B2) or the second sound collection unit 1001 (buffers B3 and B4) based on the identified position of the terminal Ti. Alternatively, it may be determined whether to extract information related to the sound spoken by the user of the terminal Ti. Specifically, for example, the extraction unit 603 calculates a distance D1 from the installation position of the microphones M1 and M2 to the position of the terminal Ti. Further, the extraction unit 603 calculates a distance D2 from the installation position of the microphones M3 and M4 to the position of the terminal Ti. However, the distances D1 and D2 are distances in a coordinate system composed of the X axis and the Y axis.

Here, when the distance D1 is shorter than the distance D2, the extraction unit 603 determines to extract information on the sound uttered by the user of the terminal Ti based on the information stored in the sound collection unit 307. And the extraction part 603 is based on the information regarding the sound for every microphone M1, M2 of the speech section P among the information regarding the sound for each microphone M1, M2 memorize | stored in buffer B1, B2, and direction (theta) _a . Information on the sound uttered by the user of the terminal Ti is extracted.

On the other hand, when the distance D2 is shorter than the distance D1, the extraction unit 603 determines to extract information about the sound uttered by the user of the terminal Ti based on the information stored in the second sound collection unit 1001. And the extraction part 603 is based on the information regarding the sound for every microphone M3, M4 of the speech section P among the information regarding the sound for each microphone M3, M4 memorize | stored in buffer B3, B4, and direction (theta) _b . Information on the sound uttered by the user of the terminal Ti is extracted.

  Thereby, the information regarding the sound uttered by the user of the terminal Ti can be extracted from the information regarding the sound input to the microphone M physically close to the user of the terminal Ti. As a result, voice recognition can be performed using information with higher sound pressure, and voice recognition accuracy can be improved.

(Information processing procedure of the information processing apparatus 101)
Next, an information processing procedure of the information processing apparatus 101 will be described with reference to FIG.

  FIG. 12 is a flowchart illustrating an example of an information processing procedure of the information processing apparatus 101. In the flowchart of FIG. 12, first, the information processing apparatus 101 starts buffering information related to sound input to the microphones M1 and M2 to the buffers B1 and B2 of the sound collection unit 307 (step S1201).

  Next, the information processing apparatus 101 determines whether or not the beacon signal bs transmitted from the terminal Ti has been received (step S1202). Here, the information processing apparatus 101 waits for reception of the beacon signal bs (step S1202: No). When the information processing apparatus 101 receives the beacon signal bs (step S1202: Yes), the information processing apparatus 101 determines whether the operation type included in the beacon signal bs is “ON” (step S1203).

  When the operation type is “ON” (step S1203: Yes), the information processing apparatus 101 associates the terminal ID included in the beacon signal bs with the time when the beacon signal bs is received as the ON time. It registers in the utterance section table 220 (step S1204). Thereby, new utterance section information is registered in the utterance section table 220 as a record.

  However, when the information processing apparatus 101 receives the beacon signal bs via the beacon receiver 201, the information processing apparatus 101 receives the beacon signal bs including the same terminal ID and operation type from the plurality of beacon receivers 201 almost simultaneously. become. In this case, for example, the information processing apparatus 101 performs registration in the utterance section table 220 according to the beacon signal bs received first.

  Next, the information processing apparatus 101 specifies the direction θ from the microphones M1 and M2 to the terminal Ti based on the received beacon signal bs (step S1205). Specifically, for example, the information processing apparatus 101 determines the RSSI value of the beacon signal bs (including the terminal ID of the terminal Ti) received from the plurality of beacon receivers 201 and the installation position of each beacon receiver 201. Based on this, the position of the terminal Ti in the space R is estimated. Then, the information processing apparatus 101 specifies the direction θ from the microphones M1 and M2 to the terminal Ti based on the estimated position of the terminal Ti and the installation positions of the microphones M1 and M2.

  Next, the information processing apparatus 101 registers the specified direction θ in the utterance section table 220 in association with the terminal ID included in the beacon signal bs (step S1206), and returns to step S1202. That is, the information processing apparatus 101 sets the specified direction θ in the direction field of the utterance period information corresponding to the terminal ID included in the beacon signal bs.

  If the operation type is “OFF” in step S1203 (step S1203: No), the information processing apparatus 101 turns off the time when the beacon signal bs is received in association with the terminal ID included in the beacon signal bs. The time is registered in the utterance section table 220 (step S1207). That is, the information processing apparatus 101 sets the time when the beacon signal bs is received in the OFF time field of the utterance period information corresponding to the terminal ID included in the beacon signal bs.

  Next, the information processing apparatus 101 refers to the utterance interval table 220 and specifies the utterance interval P (ON time to OFF time) and the direction θ corresponding to the terminal ID included in the beacon signal bs (step S1208). Then, the information processing apparatus 101 reads information related to the sound for each of the microphones M1 and M2 in the specified speech section P from each of the buffers B1 and B2 (step S1209).

  Next, the information processing apparatus 101 performs beamform processing in the specified direction θ based on the information related to the sounds of the microphones M1 and M2 in the read utterance period P (step S1210). Then, the information processing apparatus 101 performs voice recognition processing on information regarding the sound uttered by the user of the terminal Ti (terminal Ti of the terminal ID included in the beacon signal bs) extracted by the beamform process (step S1211).

  Next, the information processing apparatus 101 outputs a voice recognition result in association with the terminal ID included in the beacon signal bs (step S1212). Then, the information processing apparatus 101 determines whether or not the information processing system 100 has ended (step S1213). If the information processing system 100 has not ended (step S1213: No), the information processing apparatus 101 returns to step S1202.

  On the other hand, when the information processing system 100 ends (step S1213: Yes), the information processing apparatus 101 ends a series of processes according to this flowchart. Thereby, the information regarding the sound uttered for each user of the terminal Ti can be extracted.

  As described above, according to the information processing apparatus 101 according to the embodiment, the utterance period P uttered by the user of the terminal Ti based on the beacon signal bs transmitted from the terminal Ti in response to an operation to the terminal Ti. Then, the direction θ from the microphones M1, M2 to the terminal Ti can be specified. Then, according to the information processing apparatus 101, the information related to the sound uttered by the user of the terminal Ti based on the information about the sound for each of the microphones M1 and M2 in the utterance section P stored in the buffers B1 and B2 and the direction θ. Can be extracted. Specifically, for example, according to the information processing apparatus 101, information on sound uttered by the user is performed by performing beamform processing for the direction θ based on information on sound for each of the microphones M1 and M2 in the utterance section P. Can be extracted.

  As a result, it is possible to extract the sound uttered by the user of the terminal Ti by emphasizing the voice signal that has arrived from the direction θ of the terminal Ti (the direction of the user of the terminal Ti viewed from the microphones M1 and M2). In addition, sound information in which the sound signal arriving from each direction θ is emphasized can be generated using the information regarding the sound for each of the microphones M1 and M2 stored in the buffers B1 and B2. For this reason, even when a plurality of people speak at the same timing, it is possible to accurately extract the sound spoken for each user.

  Further, according to the information processing apparatus 101, it is possible to output information relating to the extracted sound uttered by the user of the terminal Ti in association with the terminal ID of the terminal Ti. Thereby, the information regarding the sound uttered by the user can be identified and provided to an external device or the like (for example, a voice recognition device).

  Further, according to the information processing apparatus 101, it is possible to recognize the information related to the sound uttered by the user of the extracted terminal Ti and output the recognition result of the recognized voice in association with the terminal ID of the terminal Ti. As a result, the recognition result obtained by voice recognition of the information related to the sound uttered by the user can be provided to an external device or the like (for example, an electronic device capable of voice input) while making the user discriminable.

  Further, according to the information processing apparatus 101, the height of the user of the terminal Ti is set to a predetermined height based on information about sound input to the microphones M′a and M′b installed at different heights. It can be determined whether or not it is K or more. And according to the information processing apparatus 101, based on the determined result, the information regarding the sound uttered by the user of the extracted terminal Ti can be recognized. Thereby, for example, it is possible to determine whether the user of the terminal Ti is an adult or a child, and to set an acoustic model used for speech recognition, thereby improving speech recognition accuracy.

Further, according to the information processing apparatus 101 can be based on information about the sound input to the microphone M1, M2 respectively, to identify the direction theta _a from the microphone M1, M2 to the terminal Ti. Further, according to the information processing apparatus 101, the direction θ _b from the microphones M3 and M4 to the terminal Ti is based on information about sound input to the microphones M3 and M4 installed at positions different from the microphones M1 and M2. Can be specified. Then, according to the information processing apparatus 101, the position of the terminal Ti can be specified based on the specified directions θ _a and θ _b , the installation positions of the microphones M1 and M2, and the installation positions of the microphones M3 and M4. it can.

  Thereby, the position of the user of the terminal Ti in the space R can be specified. However, only the user of the terminal Ti speaks in specifying the position of the user of the terminal Ti. For this reason, when there are a plurality of users, for example, each of the plurality of users speaks in order and specifies the position of each user.

  Further, according to the information processing apparatus 101, the information is stored in either the sound collection unit 307 (buffers B1 and B2) or the second sound collection unit 1001 (buffers B3 and B4) based on the specified position of the terminal Ti. Based on the information, it can be determined whether or not to extract information related to the sound spoken by the user of the terminal Ti.

  Thereby, the information regarding the sound uttered by the user of the terminal Ti can be extracted from the information regarding the sound input to the microphone M physically close to the user of the terminal Ti. As a result, voice recognition can be performed using information with higher sound pressure, and voice recognition accuracy can be improved.

  The following additional notes are disclosed with respect to the embodiment described above.

(Supplementary note 1) a plurality of microphones;
For each microphone included in the plurality of microphones, a storage unit that stores information about sound input to the microphones;
Based on information transmitted from the terminal in response to an operation to the terminal, a specifying unit that specifies a period during which the user of the terminal spoke and a direction from the plurality of microphones to the terminal;
An extraction unit for extracting information on the sound uttered by the user based on the information on the sound for each microphone in the period stored in the storage unit and the direction;
An information processing apparatus comprising:

(Supplementary note 2) The information processing apparatus according to supplementary note 1, further comprising: an output unit that outputs information relating to the sound uttered by the user extracted by the extraction unit in association with identification information of the terminal.

(Supplementary Note 3) A voice recognition unit that recognizes information related to the sound uttered by the user extracted by the extraction unit;
An output unit that outputs a recognition result recognized by the voice recognition unit in association with identification information of the terminal;
The information processing apparatus according to appendix 1 or 2, characterized by comprising:

(Supplementary Note 4) The extraction unit
Any one of appendices 1 to 3, wherein information related to the sound spoken by the user is extracted by performing beamform processing for the direction based on information related to the sound for each microphone in the period. Information processing apparatus described in one.

(Supplementary Note 5) The voice recognition unit
Determining whether the height of the user is greater than or equal to a predetermined height based on information about sound input to each of the plurality of second microphones installed at different heights;
4. The information processing apparatus according to appendix 3, wherein information related to the sound uttered by the user extracted by the extraction unit is recognized based on the determined determination result.

(Appendix 6) A plurality of third microphones installed at positions different from the plurality of microphones;
A second storage unit that stores information about sound input to the third microphone for each third microphone included in the plurality of third microphones;
The specific part is:
Identifying a first direction from the plurality of microphones to the terminal based on information about sound input to each of the plurality of microphones;
Identifying a second direction from the plurality of third microphones to the terminal based on information about sound input to each of the plurality of third microphones;
The extraction unit includes:
Identifying the position of the terminal based on the first and second directions, the installation positions of the plurality of microphones, and the installation positions of the plurality of third microphones;
Based on the identified position of the terminal, it is determined whether to extract information related to the sound spoken by the user based on information stored in either the storage unit or the second storage unit.
The information processing apparatus according to any one of supplementary notes 1 to 5, wherein:

(Appendix 7) Based on information transmitted from the terminal in response to an operation on the terminal, a period during which the user of the terminal speaks and directions from a plurality of microphones to the terminal are identified,
For each microphone included in the plurality of microphones, the user based on the information about the sound for each microphone in the period stored in the storage unit that stores information about the sound input to the microphone, and the direction Extract information about the sound uttered by
An information processing program for causing a computer to execute processing.

(Appendix 8) A building having a plurality of microphones on a wall,
For each microphone included in the plurality of microphones, based on information stored in the storage unit that stores information related to sound input to the microphone and information transmitted from the terminal in response to an operation on the terminal, the user of the terminal Based on the period of utterance, the identification unit that identifies the direction from the plurality of microphones to the terminal, the information on the sound for each microphone of the period stored in the storage unit, and the direction, A building comprising: an information processing apparatus including an extraction unit that extracts information related to sound uttered by a user.

DESCRIPTION OF SYMBOLS 100 Information processing system 101 Information processing apparatus 110 Memory | storage part 201 Beacon receiver 210 Network 220 Speech area table 300,400 Bus 301,401 CPU
302, 402 Memory 303 I / F
304 beacon reception unit 305 disk drive 306 disk 307,1001 sound collection unit 403 operation button 404 LED lamp 405 beacon transmission unit 601 acquisition unit 602 identification unit 603 extraction unit 604 voice recognition unit 605 output unit

Claims (7)

Multiple microphones,
For each microphone included in the plurality of microphones, a storage unit that stores information about sound input to the microphones;
Based on information transmitted from the terminal in response to an operation to the terminal, a specifying unit that specifies a period during which the user of the terminal spoke and a direction from the plurality of microphones to the terminal;
An extraction unit for extracting information on the sound uttered by the user based on the information on the sound for each microphone in the period stored in the storage unit and the direction;
An information processing apparatus comprising:   The information processing apparatus according to claim 1, further comprising: an output unit that outputs information related to the sound uttered by the user extracted by the extraction unit in association with identification information of the terminal. A voice recognition unit for recognizing information about the sound uttered by the user extracted by the extraction unit;
An output unit that outputs a recognition result recognized by the voice recognition unit in association with identification information of the terminal;
The information processing apparatus according to claim 1, further comprising: The voice recognition unit
Determining whether the height of the user is greater than or equal to a predetermined height based on information about sound input to each of the plurality of second microphones installed at different heights;
The information processing apparatus according to claim 3, wherein information on the sound uttered by the user extracted by the extraction unit is recognized based on the determined determination result. A plurality of third microphones installed at positions different from the plurality of microphones;
A second storage unit that stores information about sound input to the third microphone for each third microphone included in the plurality of third microphones;
The specific part is:
Identifying a first direction from the plurality of microphones to the terminal based on information about sound input to each of the plurality of microphones;
Identifying a second direction from the plurality of third microphones to the terminal based on information about sound input to each of the plurality of third microphones;
The extraction unit includes:
Identifying the position of the terminal based on the first and second directions, the installation positions of the plurality of microphones, and the installation positions of the plurality of third microphones;
Based on the identified position of the terminal, it is determined whether to extract information related to the sound spoken by the user based on information stored in either the storage unit or the second storage unit.
The information processing apparatus according to claim 1, wherein the information processing apparatus is an information processing apparatus. Based on information transmitted from the terminal in response to an operation to the terminal, specify a period during which the user of the terminal speaks and directions from a plurality of microphones to the terminal,
For each microphone included in the plurality of microphones, the user based on the information about the sound for each microphone in the period stored in the storage unit that stores information about the sound input to the microphone, and the direction Extract information about the sound uttered by
An information processing program for causing a computer to execute processing. A building having a plurality of microphones on the wall,
For each microphone included in the plurality of microphones, based on information stored in the storage unit that stores information related to sound input to the microphone and information transmitted from the terminal in response to an operation on the terminal, the user of the terminal Based on the period of utterance, the identification unit that identifies the direction from the plurality of microphones to the terminal, the information on the sound for each microphone of the period stored in the storage unit, and the direction, A building comprising: an information processing apparatus including an extraction unit that extracts information related to sound uttered by a user.

Images