JP5929810B2 - Voice analysis system, voice terminal apparatus and program - Google Patents

Description

  The present invention relates to a voice analysis system, a voice terminal device, and a program.

  Patent Literature 1 includes a plurality of mobile stations and a base station that transmits and receives information to and from the plurality of mobile stations by wireless communication, and detects communication states of a plurality of employees who respectively own the plurality of mobile stations. A detection system that detects the position of each of a plurality of mobile stations based on radio signals transmitted from the antennas of the plurality of mobile stations, records communication elements related to employees of each mobile station based on the detection results, and A communication detection system that calculates the communication activity level between employees based on the recorded contents is disclosed.

JP 2010-21700 A

  An object of this invention is to improve the determination precision at the time of determining the dialog relationship between wearers from the information regarding the sound acquired by the sound acquisition means of a plurality of wearers.

According to the first aspect of the present invention, a plurality of voice acquisition means that are arranged at different distances from the wearer's utterance site and that acquire the voice of the speaker and voices acquired by at least two of the voice acquisition means Identification means for identifying whether the speaker is the wearer or another person other than the wearer based on the sound pressure of the voice, and the voice when the speaker is identified as the wearer by the identification means An utterance signal transmitting means for transmitting an utterance signal related to the utterance of the wearer at a radio wave intensity based on a sound pressure of the wearer's voice acquired by the acquiring means; and the utterance signal transmitted from the utterance signal transmitting means. A speech signal receiving means for receiving the speech signal, and a dialogue relation determining means for judging the dialogue relation of the wearer based on the reception status of the speech signal by the speech signal receiving means and the identification result by the identification means It is an analysis system.
According to a second aspect of the present invention, when the utterance signal is received by the utterance signal receiving means, a reception information transmission means for transmitting reception information based on reception of the utterance signal, and the reception information transmission means Receiving information acquisition means for acquiring the received reception information, wherein the dialogue relationship determination means is based on the identification result by the identification means and the reception information acquired by the reception information acquisition means, The speech analysis system according to claim 1, wherein a dialogue relation of the wearer is determined.

The invention according to claim 3 is arranged at a position where the distance from the utterance part of the wearer is different, and a plurality of voice acquisition means for acquiring the voice of the speaker and the voice acquired by at least two of the voice acquisition means Based on the sound pressure, identification means for identifying whether the speaker is the wearer or another person other than the wearer, and communication with the outside via a wireless communication line based on the identification result by the identification means A plurality of voice terminal devices each provided with a communication means, and a voice analysis device provided with a dialogue relation determination means for judging a dialogue relation between wearers of each of the voice terminal devices, and the voice terminal device The communication means, when the speaker is identified as the wearer by the identification means, the radio wave intensity based on the sound pressure of the wearer's voice acquired by the voice acquisition means, Speech related to utterance An utterance signal transmitting means for transmitting a signal, and an utterance signal receiving means for receiving an utterance signal transmitted from an utterance signal transmitting means in another voice terminal device, wherein the dialogue relation determining means of the voice analysis device is respectively And determining a dialogue relation based on the identification result of the voice terminal device by the identification unit and the reception status of the utterance signal by the utterance signal reception unit of each voice terminal device. It is an analysis system.
According to a fourth aspect of the present invention, in the voice terminal device, the dialogue relation determination unit of the voice analysis device transmits the utterance signal by the utterance signal transmission unit among the wearers of the plurality of voice terminal devices. The speech analysis system according to claim 3, wherein a dialogue relation is determined between the wearer of the wearer and the wearer of the voice terminal device that has received the speech signal by the speech signal receiving means.
According to a fifth aspect of the present invention, the dialogue relation determination unit of the voice analysis device acquires the voice acquired by the voice acquisition unit of the voice terminal device that has transmitted the utterance signal by the utterance signal transmission unit, and The dialogue relation is determined by comparing synchrony with the voice acquired by the voice acquisition unit of the voice terminal device that has received the speech signal by the utterance signal receiving unit. 4. The voice analysis system according to 4.

According to the sixth aspect of the present invention, a plurality of voice acquisition means that are arranged at different distances from the wearer's utterance site and that acquire the voice of the speaker, and voices acquired by at least two of the voice acquisition means Identification means for identifying whether the speaker is the wearer or another person other than the wearer based on the sound pressure of the voice, and the voice when the speaker is identified as the wearer by the identification means An utterance signal transmitting means for transmitting an utterance signal related to the utterance of the wearer at a radio wave intensity based on the sound pressure of the wearer's voice acquired by the acquisition means, and an utterance signal for receiving an utterance signal related to the utterance of the other person And a voice terminal apparatus including a receiving unit.
The invention according to claim 7 further includes speech information transmitting means for transmitting speech information including information related to the voice acquired by the voice acquisition means and the identification result by the identification means. It is the voice terminal device described.
The invention according to claim 8 is characterized in that the utterance signal receiving means receives an utterance signal relating to the utterance of the other person transmitted by the utterance signal transmitting means of another device worn by the other person. Item 8. The voice terminal device according to Item 6 or 7.
The invention according to claim 9 is characterized in that the utterance signal transmitting means transmits the utterance signal with a higher radio field strength as the sound pressure of the wearer's voice acquired by the voice acquisition means is larger. The voice terminal device according to any one of claims 6 to 8.
The invention according to claim 10 is characterized in that the computer has a function of acquiring voice information from a plurality of voice acquisition means that are arranged at different positions from the wearer's utterance site and that acquire the voice of the speaker, A function for identifying whether the speaker is the wearer or another person other than the wearer, and the speaker is identified as the wearer based on the sound pressure difference between the two voices acquired by the two voice acquisition means A function to transmit an utterance signal related to the utterance of the wearer at a radio wave intensity based on the sound pressure of the wearer's voice acquired by the voice acquisition means, and an utterance signal related to the utterance of the other person It is a program that realizes the function to perform.

According to the first aspect of the present invention, compared with the case where this configuration is not adopted, the determination when determining the interactive relationship between the wearers from the information regarding the sound acquired by the sound acquisition means of the plurality of wearers. The accuracy can be improved.
According to the invention which concerns on Claim 2, compared with the case where this structure is not employ | adopted, it becomes possible to suppress generation | occurrence | production of a misjudgment at the time of determining the wearer's mutual dialogue relationship.
According to the third aspect of the present invention, compared with the case where the present configuration is not adopted, the determination when determining the interactive relationship between the wearers from the information regarding the sound acquired by the sound acquisition means of the plurality of wearers. The accuracy can be improved.
According to the fourth aspect of the present invention, it is possible to prevent the processing related to the determination of the dialogue relation from becoming complicated as compared with the case where this configuration is not adopted.
According to the invention which concerns on Claim 5, compared with the case where this structure is not employ | adopted, it becomes possible to suppress generation | occurrence | production of a misjudgment at the time of determining the dialog relationship between wearers.

According to the sixth aspect of the present invention, compared with the case where this configuration is not adopted, the determination when determining the interactive relationship between the wearers from the information regarding the sound acquired by the sound acquisition means of the plurality of wearers. The accuracy can be improved.
According to the seventh aspect of the present invention, it is possible to more easily determine the interactive relationship between the wearers than in the case where this configuration is not adopted.
According to the invention which concerns on Claim 8, compared with the case where this structure is not employ | adopted, it becomes possible to suppress generation | occurrence | production of a misjudgment at the time of determining the dialog relationship between wearers.
According to the ninth aspect of the present invention, it is possible to suppress the processing related to the determination of the dialogue relation from becoming complicated as compared with the case where this configuration is not adopted.
According to the tenth aspect of the present invention, as compared with the case where the present configuration is not adopted, the determination when determining the interactive relationship between the wearers from the information regarding the sound acquired by the sound acquisition means of the plurality of wearers. The accuracy can be improved.

It is a figure which shows the structural example of the audio | voice analysis system with which this Embodiment is applied. It is a figure which shows the structural example of a terminal device. It is a figure which shows the condition where the several wearer who respectively mounted | worn with the terminal device of this Embodiment is interacting. It is a figure which shows the example of the speech information of each terminal device in the dialog condition of FIG. It is a flowchart which shows the process performed with the terminal device to which this Embodiment is applied. It is the figure which showed the relationship between the distance between two persons who talk, and the sound pressure of a wearer's utterance voice, and the relationship between the distance between two persons who talk, and the radio field intensity of the speech signal transmitted from a terminal device. It is a flowchart which shows the process performed with the host apparatus to which this Embodiment is applied. It is a figure for demonstrating in detail the process performed with the audio | voice analysis system to which this Embodiment is applied. It is the figure which showed arrangement | positioning of the wearer in Example 1 and Comparative Example 1. FIG. It is the figure which showed the reception condition of the speech signal in Example 1 and Comparative Example 1. FIG.

Embodiments of the present invention will be described below in detail with reference to the accompanying drawings.
<System configuration example>
FIG. 1 is a diagram illustrating a configuration example of a speech analysis system to which the exemplary embodiment is applied.
As shown in FIG. 1, the voice analysis system 1 according to the present embodiment includes a terminal device 10 as an example of a voice terminal device and a host device 20 as an example of a voice analysis device. The terminal device 10 and the host device 20 are connected via a wireless communication line. As a type of the wireless communication line, a line using an existing method such as Wi-Fi (Wireless Fidelity), Bluetooth (registered trademark), ZigBee, or UWB (Ultra Wideband) may be used. In the illustrated example, only one terminal device 10 is described, but as will be described in detail later, the terminal device 10 is worn and used by each user and is actually used. As many terminal devices 10 as the number of persons are prepared. Hereinafter, a user wearing the terminal device 10 is referred to as a wearer.

The terminal device 10 includes a plurality of microphones (first microphone 11 and second microphone 12) and amplifiers (first amplifier 13 and second amplifier 14) as voice acquisition means for acquiring voice. Further, the terminal device 10 includes a voice analysis unit 15 that analyzes the acquired voice and a data transmission unit 16 that transmits the analysis result to the host device 20.
In addition, the terminal device 10 includes a signal receiving unit 17 for receiving an utterance signal (described later) output from another terminal device 10.
Furthermore, the terminal device 10 includes a power supply unit 18 for supplying power to each unit of the terminal device 10.
In the terminal device 10 according to the present embodiment, the data transmitting unit 16 and the signal receiving unit 17 constitute a communication unit.

  The first microphone 11 and the second microphone 12 are arranged at different positions from the wearer's mouth (speaking part) (in the following description, the first microphone 11 and the second microphone 12 are not distinguished from each other). May be described as microphones 11 and 12). Here, the first microphone 11 is arranged at a position (for example, about 35 cm) far from the mouth (speaking part) of the wearer, and the second microphone 12 is a position (for example, about 10 cm) near the mouth (speaking part) of the wearer. ). As the types of microphones used as the first microphone 11 and the second microphone 12 of the present embodiment, various existing types such as a dynamic type and a condenser type may be used. In particular, a non-directional MEMS (Micro Electro Mechanical Systems) type microphone is preferable.

  The first amplifier 13 and the second amplifier 14 amplify electrical signals (audio signals) that are output according to the audio acquired by the first microphone 11 and the second microphone 12, respectively. As an amplifier used as the first amplifier 13 and the second amplifier 14 of the present embodiment, an existing operational amplifier or the like may be used.

  The voice analysis unit 15 is an example of an identification unit, and analyzes the voice signals output from the first amplifier 13 and the second amplifier 14. Then, it identifies whether the voice acquired by the first microphone 11 and the second microphone 12 is the voice of the wearer wearing the terminal device 10 or the voice of another person's speech (self-other identification). Details of specific processing for voice identification will be described later.

The data transmission unit 16 transmits the acquired data including the result of speech analysis (result of self-other identification) by the voice analysis unit 15 and the terminal ID of the terminal device 10 to the host device 20 via the wireless communication line. . As information to be transmitted to the host device 20, in addition to the above analysis results, for example, the acquisition time of the sound by the first microphone 11 and the second microphone 12 and the acquired sound according to the contents of the processing performed in the host device 20 Information such as sound pressure may be included. In the present embodiment, these data transmitted from the data transmission unit 16 to the host device 20 are referred to as speech information.
Further, the terminal device 10 may be provided with a data accumulation unit that accumulates utterance information such as analysis results by the voice analysis unit 15, and data (utterance information) stored for a certain period may be transmitted collectively. Note that the transmission of utterance information from the data transmission unit 16 to the host device 20 may be performed using a wired line.

Furthermore, when the voice analysis unit 15 identifies that the voice acquired by the first microphone 11 and the second microphone 12 is the voice of the wearer itself, the data transmission unit 16 of the present embodiment performs the above wireless communication. An utterance signal is transmitted to another terminal device 10 via a communication line.
Furthermore, the data transmission unit 16 according to the present embodiment transmits reception information to the host device 20 when the signal reception unit 17 receives an utterance signal.
In the terminal device 10 of the present embodiment, the data transmission unit 16 constitutes an utterance signal transmission unit, an utterance information transmission unit, and a reception information transmission unit.

Here, the speech signal includes, for example, the ID of the terminal device 10 that transmits the speech signal, the acquisition time of the wearer's speech by the first microphone 11 and the second microphone 12, and the like.
The received information relates to the ID information of the terminal device 10 that transmits the received information, the ID information of the other terminal device 10 that transmitted the utterance signal included in the received utterance signal, and the acquisition time of the utterance voice. Information etc. are included.
Note that the transmission processing of the speech signal and the reception information in the data transmission unit 16 will be described in detail later.

The signal receiving unit 17 is an example of an utterance signal receiving unit, and receives an utterance signal transmitted from another terminal device 10 as described above.
The power supply unit 18 supplies power to the first microphone 11, the second microphone 12, the first amplifier 13, the second amplifier 14, the voice analysis unit 15, the data transmission unit 16, and the signal reception unit 17. As the power source, for example, an existing power source such as a dry battery or a rechargeable battery is used. The power supply unit 18 includes known circuits such as a voltage conversion circuit and a charge control circuit as necessary.

  The host device 20 outputs a data reception unit 21 that receives data transmitted from the terminal device 10, a data storage unit 22 that stores the received data, a data analysis unit 23 that analyzes the stored data, and an analysis result. Output unit 24. The host device 20 is realized by an information processing device such as a personal computer. Further, as described above, a plurality of terminal devices 10 are used in the present embodiment, and the host device 20 receives data transmitted from each of the plurality of terminal devices 10.

The data receiving unit 21 is an example of a reception information acquisition unit and corresponds to the above-described wireless communication line. The data receiving unit 21 receives data from each terminal device 10 and sends the data to the data storage unit 22. Examples of data received by the data receiving unit 21 include speech information and reception information transmitted from the data transmitting unit 16 of each terminal device 10.
The data storage unit 22 is realized by a storage device such as a magnetic disk device of a personal computer, for example, and stores data included in speech information and reception information acquired from the data reception unit 21 for each speaker. Here, the speaker is identified by collating the terminal ID included in the utterance information and reception information transmitted from the terminal device 10 with the utterer name registered in the host device 20 in advance. Further, the wearer state may be transmitted from the terminal device 10 instead of the terminal ID.
Although details will be described later, the data reception unit 21 also receives reception information transmitted from the data transmission unit 16 of each terminal device 10.

The data analysis unit 23 is an example of a dialogue relationship determination unit, and is realized by a program-controlled CPU of a personal computer, for example, and analyzes data stored in the data storage unit 22. Specific analysis contents and analysis methods may take various contents and methods depending on the purpose and use of the system according to the present embodiment. For example, the frequency of dialogue between wearers of the terminal device 10 and the tendency of each wearer's dialogue partner may be analyzed, or the relationship between the dialogues may be inferred from information on individual utterance length and sound pressure in the dialogue. Done.
Although details will be described later, in the present embodiment, the data analysis unit 23 synchronizes the uttered voice based on the utterance information received by the data reception unit 21 and stored in the data storage unit 22 and the reception information. Gender is discriminated, and the relationship of the interlocutor is judged.

  The output unit 24 outputs the analysis result obtained by the data analysis unit 23 or outputs other data based on the analysis result. The means for outputting the analysis result or the like can take various means such as display display, print output by a printer, voice output, and the like according to the purpose and use of the system and the content and format of the analysis result.

<Configuration example of terminal device>
FIG. 2 is a diagram illustrating a configuration example of the terminal device 10.
As described above, the terminal device 10 is worn and used by each user (wearer). In order for the user to be able to wear the terminal device 10 according to the present embodiment, as shown in FIG. 2, the terminal device 10 includes a device main body 30 and a strap 40 connected to the device main body 30. In the configuration shown in the figure, the user puts the neck through the strap 40 and hangs the apparatus main body 30 from the neck.

  The apparatus main body 30 includes at least a first amplifier 13, a second amplifier 14, a voice analysis unit 15, a data transmission unit 16, a signal reception unit 17, and a power supply unit 18 in a thin rectangular parallelepiped case 31 made of metal, resin, or the like. The circuit to be realized and the power supply (battery) of the power supply unit 18 are accommodated. In the present embodiment, the case 31 is provided with the first microphone 11. Further, the case 31 may be provided with a pocket for inserting an ID card or the like displaying ID information such as the name and affiliation of the wearer. Further, such ID information or the like may be printed on the surface of the case 31 itself, or a sticker describing the ID information or the like may be attached.

  The strap 40 is provided with the second microphone 12. The second microphone 12 is connected to the second amplifier 14 accommodated in the case 31 of the apparatus main body 30 by a cable (electric wire or the like) passing through the inside of the strap 40. As the material of the strap 40, various existing materials such as leather, synthetic leather, cotton and other natural fibers and synthetic fibers such as resin, metal, etc. may be used. Moreover, the coating process using a silicon resin, a fluororesin, etc. may be given.

The strap 40 has a cylindrical structure, and the second microphone 12 is housed inside the strap 40. By providing the second microphone 12 inside the strap 40, the second microphone 12 is prevented from being damaged or soiled, and the conversation person is prevented from being aware of the presence of the second microphone 12.
In the present embodiment, the first microphone 11 disposed at a position far from the wearer's mouth (speaking part) is provided in the apparatus main body 30, but like the second microphone 12, the first microphone 11 is provided. The string 40 may be provided.

In the present embodiment, the voice acquired by the terminal device 10 is identified (self-other identification) whether the speaker is a wearer or another person other than the wearer, and the result of self-other identification is used. Thus, an utterance signal is transmitted from the terminal device 10 to another terminal device 10. Then, the host device 20 determines the dialogue relationship of the wearer wearing each terminal device 10 based on the reception status of the utterance signal in the other terminal devices 10.
Hereinafter, the self-other identification method and the interactive relationship determination method will be described in order.

<Description of how to identify whether a speaker is a wearer or someone else>
Subsequently, in the speech analysis system 1 described above, the speech analysis unit 15 of the terminal device 10 identifies (self-other identification) whether the speaker is a wearer or another person other than the wearer. Will be described.
In the voice analysis system 1 according to the present embodiment, using the voice information acquired by the first microphone 11 and the second microphone 12 provided in the terminal device 10, the acquired voice is transmitted to the wearer himself / herself of the terminal device 10. It is identified whether it is a speech voice or another person's speech voice. In other words, in the voice analysis system 1 of the present embodiment, the other person is identified as to the acquired voice speaker. In the present embodiment, the sound pressure (the input sound volume to the first microphone 11 and the second microphone 12), etc., not the linguistic information obtained by using morphological analysis, dictionary information, or the like among the acquired voice information. The speaker is identified based on the non-linguistic information. In other words, in the present embodiment, the voice speaker is identified not from the utterance content specified by the linguistic information but from the utterance situation specified by the non-linguistic information.

As described with reference to FIGS. 1 and 2, in the terminal device 10 of the present embodiment, the first microphone 11 is disposed at a position far from the wearer's mouth (speaking site), and the second microphone 12 is worn. It is arranged at a position close to the person's mouth (voice part). That is, when the wearer's mouth (speaking part) is a sound source, the distance between the first microphone 11 and the sound source and the distance between the second microphone 12 and the sound source are greatly different. For example, the distance between the first microphone 11 and the sound source can be set to about 1.5 to 4 times the distance between the second microphone 12 and the sound source.
Here, the sound pressure of the sound acquired by the microphones 11 and 12 is attenuated (distance attenuation) as the distance between the microphones 11 and 12 and the sound source increases. Therefore, regarding the voice of the wearer, the sound pressure of the sound acquired by the first microphone 11 and the sound pressure of the sound acquired by the second microphone 12 are greatly different.

  On the other hand, considering the case where the mouth (speaking part) of a person other than the wearer is used as a sound source, since the other person is usually away from the wearer, the distance between the first microphone 11 and the sound source The distance between the second microphone 12 and the sound source does not change greatly. Depending on the position of the other person with respect to the wearer, a difference between the two distances may occur, but the distance between the first microphone 11 and the sound source is the second as in the case where the mouth (speaking part) of the wearer is used as the sound source. It is never several times the distance between the microphone 12 and the sound source. Therefore, the sound pressure of the sound acquired by the first microphone 11 and the sound pressure of the sound acquired by the second microphone 12 with respect to the speech sound of the other person are greatly different as in the case of the wearer's speech sound. There is nothing.

Therefore, in the present embodiment, a sound pressure ratio that is a ratio between the sound pressure of the sound acquired by the first microphone 11 and the sound pressure of the sound acquired by the second microphone 12 is obtained. Then, using this sound pressure ratio, it is identified whether the acquired voice is the wearer's own voice or the other person's voice. More specifically, in the present embodiment, a threshold is set for the ratio of the sound pressure of the sound acquired by the first microphone 11 to the sound pressure of the sound acquired by the second microphone 12 (sound pressure ratio). And when the sound pressure ratio of the acquired sound is larger than the threshold value, it is determined as the speech sound of the wearer itself, and when the sound pressure ratio is smaller than the threshold value, it is determined as the speech sound of the other person.
The identification result obtained by the above-described method as to whether the voice speaker is a wearer or another person is included in the utterance information and transmitted from the data transmission unit 16 of the terminal device 10 to the host device 20. Is done.

In the above-described example, the self / other identification is determined based on the sound pressure of the sound acquired by the first microphone 11 and the second microphone 12, but information on the phase difference of the sound may be added to this. Conceivable. That is, when the wearer's mouth (speaking part) is a sound source, as described above, the distance between the first microphone 11 and the sound source and the distance between the second microphone 12 and the sound source are greatly different. Therefore, the phase difference between the sound acquired by the first microphone 11 and the sound acquired by the second microphone 12 becomes large. On the other hand, when the other person's mouth (speaking part) other than the wearer is used as the sound source, as described above, since the other person is away from the wearer, the distance between the first microphone 11 and the sound source, The distance between the second microphone 12 and the sound source does not change greatly. Therefore, the phase difference between the sound acquired by the first microphone 11 and the sound acquired by the second microphone 12 is small.
Therefore, by considering the phase difference between the sound acquired by the first microphone 11 and the sound acquired by the second microphone 12, the accuracy of the self-other identification determination can be improved.

<Determination of dialogue between wearers>
Next, a method of determining the dialogue relationship between a plurality of wearers based on the utterance information transmitted from each terminal device 10 and received by the data receiving unit 21 in the data analysis unit 23 of the host device 20 will be described. In the present embodiment, the dialogue relationship between the wearers is determined by determining the presence / absence of audio synchronism with respect to the utterance information received from each terminal device 10.
Note that the method described below is an example of a method for determining a dialogue relationship between wearers, and other methods may be adopted.

FIG. 3 is a diagram illustrating a situation in which a plurality of wearers each wearing the terminal device 10 of the present embodiment are interacting with each other. FIG. 4 is a diagram showing an example of utterance information of each terminal device 10A, 10B in the conversation state of FIG.
As shown in FIG. 3, a case is considered in which two wearers A and B who are respectively wearing the terminal device 10A and the terminal device 10B are interacting with each other. At this time, the voice of the wearer A is captured by both the terminal device 10A of the wearer A and the terminal device 10B of the wearer B. Similarly, the voice of the wearer B is captured by both the terminal device 10A of the wearer A and the terminal device 10B of the wearer B.

  The terminal device 10A and the terminal device 10B independently send utterance information to the host device 20. Here, the voice recognized as the utterance of the wearer in the terminal device 10A of the wearer A is recognized as the utterance of the other person in the terminal device 10B of the wearer B. On the other hand, the voice recognized as the utterance of the wearer in the terminal device 10B is recognized as the utterance of the other person in the terminal device 10A. For this reason, the utterance information acquired from the terminal device 10A and the utterance information acquired from the terminal device 10B are information indicating the utterance status such as the length of the utterance time and the timing when the utterer is switched as shown in FIG. Approximate.

Therefore, in this example, the host device 20 compares the information acquired from the terminal device 10A and the information acquired from the terminal device 10B to determine whether or not these information indicate the same utterance situation. Based on the above, the presence or absence of dialogue between the wearer A and the wearer B is recognized.
Here, the information indicating the utterance state includes at least the length of the utterance time in each utterance for each utterer described above, the start time and end time of each utterance, and the time (timing) at which the utterer is switched. As described above, time information related to the utterance is used. In addition, in order to judge the utterance situation concerning a specific conversation, only a part of time information regarding these utterances may be used, or other information may be additionally used.

Here, in this Embodiment, it analyzes from the audio | voice analysis part 15 of the terminal device 10, and uses the self-and-others identification information transmitted via the data transmission part 16 as speech information, from several terminal device 10 The synchronism of the voice signal of the uttered voice received by the microphones 11 and 12 is determined.
That is, since the identification information is added to the utterance information, it can be determined in advance whether the acquired voice is from the wearer itself or from someone other than the wearer. It is possible to clearly grasp the timing when the person switched. If wearers whose speakers are reversed at this timing are found, it can be determined that the utterances are synchronized, and it can be determined that the wearers are interacting with each other.

By the way, conventionally, in the voice analysis system 1, when there are a plurality of wearers wearing the terminal device 10, all of the wearers existing in a predetermined space are determined in order to determine the dialogue relationship between the wearers. As described above, the above-mentioned dialogue relation is determined. Specifically, a reference wearer is determined in advance from a plurality of wearers existing in a predetermined space, and the utterance information from the terminal device 10 worn by the reference wearer, By sequentially comparing the utterance information from all the terminal devices 10 worn by other wearers present in the space, the presence or absence of synchronism of the uttered voice is determined, and the reference wearer and other wearers The dialogue relationship with is determined.
Therefore, when the number of wearers existing in a predetermined space is large, the number of other wearers that determine the presence / absence of synchronism of speech with the wearer serving as a reference increases. This complicates the process for determining the interaction relationship between a plurality of wearers, and makes it easy to generate erroneous determinations such as determining that the user is interacting without actually performing the interaction. There is a concern that the accuracy of the determination of dialogue relations will be reduced.

In order to solve such a problem, among other wearers existing in a predetermined space, other wearers whose distance from the reference wearer is within a range smaller than the predetermined distance It is conceivable that a person is a target for determining a dialogue relationship.
However, when the distance from the reference wearer is set large, the number of other wearers existing within the range for determining the dialogue relation increases, and the same problem as the above problem may occur. .
In addition, when the distance from the reference wearer is set small, other wearers who are actually interacting with the reference wearer may not be within the range for determining the dialogue relationship. is there. In other words, depending on the environment in which the conversation is performed, it may be possible to perform the conversation with a distance from the other party, and the distance between the wearer serving as a reference for the conversation and the other wearer is determined in advance. May be greater than distance. In such a case, it is difficult to accurately grasp the dialogue relationship between the wearers because the dialogue relationship between the wearer serving as the reference and other wearers is not determined.

Therefore, when the terminal device 10 according to the present embodiment identifies that the voice acquired by the terminal device 10 is the utterance voice of the wearer, the terminal apparatus 10 uses the radio wave intensity based on the sound pressure of the utterance voice of the wearer. The transmitted speech signal is transmitted to another terminal device 10. The terminal device 10 that has received the speech signal among the other terminal devices 10 transmits the reception information to the host device 20.
Furthermore, in the host device 20 of the present embodiment, based on the utterance information and reception information received from the terminal device 10, the combination of the terminal devices 10 that perform dialogue-related determination among the plurality of terminal devices 10 is limited, The above-described determination of the synchronism of the voice and the determination of the dialogue relationship between the wearers are performed.
Hereinafter, processing executed by the terminal device 10 and processing executed by the host device 20 will be described in order.

<Processing executed by terminal device 10>
FIG. 5 is a flowchart showing processing executed by the terminal device 10 to which the exemplary embodiment is applied. Next, with reference to FIG. 5 and FIG. 1 described above, processing executed in the terminal device 10 of the present embodiment will be described.
In the terminal device 10 according to the present embodiment, first, when sound is acquired by the first microphone 11 and the second microphone 12 (step 501), sound signals based on the acquired sound are converted into the first amplifier 13 and the second amplifier 14 respectively. Sent to.
When the first amplifier 13 and the second amplifier 14 acquire the audio signals from the first microphone 11 and the second microphone 12, respectively, the acquired audio signals are amplified and sent to the audio analysis unit 15 (step 502).

The voice analysis unit 15 calculates the sound pressure of the voice signal amplified by the first amplifier 13 and the second amplifier 14 (step 503).
Specifically, for example, for each of the audio signal first acquired by the first microphone 11 and amplified by the first amplifier 13 and the audio signal acquired by the second microphone 12 and amplified by the second amplifier 14, respectively. Filtering processing is performed to remove noise components such as environmental sounds from the audio signal. Then, the sound analysis unit 15 calculates an average sound pressure for each sound signal from which the noise component has been removed, every certain time unit (for example, tens of seconds to hundreds of seconds).

  Subsequently, the voice analysis unit 15 uses the above-described self-other identification method and the voices acquired by the first microphone 11 and the second microphone 12 based on the sound pressure of the voice signal calculated in Step 503. It is determined whether the voice is from the wearer or from other person than the wearer (step 504).

If the voice analysis unit 15 determines that the voice acquired by the first microphone 11 and the second microphone 12 is from the wearer (YES in step 504), the calculation is performed in step 503. Based on the sound pressure of the voice, the radio field intensity of wireless communication for transmitting the speech signal is calculated (step 505).
Subsequently, an utterance signal is transmitted to another terminal apparatus 10 through the wireless communication line with the radio wave intensity calculated in step 505 (step 506).
Note that the method of calculating the radio field intensity in step 505 and the transmission of the speech signal in step 506 will be described in detail later.

  On the other hand, if the voice analysis unit 15 determines that the voice is from another person other than the wearer (NO in step 504), the process proceeds to the next step without executing steps 505 and 506. .

Subsequently, the terminal device 10 determines whether or not the utterance signal transmitted from the other terminal device 10 is received by the signal receiving unit 17 (step 507).
When the signal reception unit 17 receives an utterance signal transmitted from another terminal device (YES in step 507), the reception information is transmitted to the host device 20 via the data transmission unit 16 (step 507). 508).
On the other hand, when the signal receiving unit 17 does not receive an utterance signal from another terminal device 10 (NO in step 507), the process proceeds to the next step without executing step 508.

Subsequently, the data transmission unit 16 transmits the utterance information to the host device 20 (step 509).
With the above steps, the process executed by the terminal device 10 of the present embodiment is completed.
In this example, the transmission of the reception information and the utterance information is performed after the transmission of the utterance signal. However, the order of the transmission of these signals and information is not limited to this. For example, the utterance information is transmitted. After that, an utterance signal may be transmitted.

Next, the calculation of the radio wave intensity in step 505 and the transmission of the speech signal in step 506 will be described in detail.
In step 505, as described above, the radio field intensity for transmitting the speech signal is calculated based on the sound pressure of the wearer's speech sound calculated by the speech analysis unit 15 in step 503.

FIG. 6 shows the relationship between the distance between the two parties having a conversation (the distance between the wearer and the conversation partner) and the sound pressure of the voice of the wearer, and the distance between the two parties having the conversation and the terminal device 10. It is the figure which showed the relationship with the radio field intensity of the speech signal transmitted from.
In FIG. 6, the solid line indicates the relationship between the sound pressure of the voice generated by the wearer and the distance between the two persons, and the broken line indicates the relationship between the radio wave intensity of the speech signal and the distance between the two persons. ing. In FIG. 6, the sound pressure of the generated speech and the radio wave intensity of the speech signal are expressed as relative values when the value when the distance between the two is 1 m is 1.

Here, generally, when a plurality of persons have a conversation, the loudness of the person who speaks (speaker) (the sound pressure of the spoken voice) varies depending on the distance to the conversation partner. That is, the speaker usually changes the sound pressure of the uttered voice according to the distance from the conversation partner so that the spoken voice reaches the conversation partner and the conversation is established.
Specifically, as shown in FIG. 6, when the distance between the two parties having a conversation is small, the sound pressure of the uttered voice of the speaker is decreased, and the utterance is increased as the distance between the two parties having the conversation is increased. There is a tendency for the sound pressure of the uttered voice of the person to increase.

  Further, when a signal such as an utterance signal is transmitted from the terminal device 10 via a wireless communication line, the terminal device 10 (the wearer of the terminal device 10) that transmits the signal and another terminal device 10 (other terminal device 10). Depending on the distance between the wearer and the conversation partner, the signal strength of the signal required for the speech signal transmitted from the terminal device 10 to reach the other terminal device 10 differs. That is, as shown in FIG. 6, when the distance between the two persons (the distance between the wearer and the conversation partner) is small, the signal is transmitted from the terminal device 10 even when the radio field intensity of the speech signal is small. The utterance signal reaches the other terminal device 10, but in order to make the utterance signal transmitted from the terminal device 10 reach the other terminal device 10 as the distance between the two increases, the radio field intensity of the utterance signal is changed. It needs to be bigger.

Based on the above relationship, the voice analysis unit 15 according to the present embodiment calculates the radio field intensity when the speech signal is transmitted from the terminal device 10 from the sound pressure of the speech sound of the wearer wearing the terminal device 10. ing.
That is, in step 505, when it is determined that the voice acquired by the terminal device 10 in step 504 is due to the wearer, based on the sound pressure of the uttered voice of the wearer calculated in step 503, FIG. The distance between the wearer and the conversation partner is calculated using the relationship shown. Based on the calculated distance between the two persons, the radio field intensity for transmitting the speech signal is calculated using the relationship shown in FIG.
Note that the sound pressure of the utterance voice used for the calculation of the radio field intensity in step 505 is acquired by the second microphone 12 even if it is the sound pressure of the utterance voice acquired by the first microphone 11 in the terminal device 10. The sound pressure of the uttered voice may be used, or the average value or the sum of these may be used.
The voice analysis unit 15 may store the sound pressure of the utterance voice of the wearer and the radio wave intensity of the utterance signal in association with each other, and directly based on the sound pressure of the wearer's utterance voice. The radio field intensity of the speech signal may be calculated.

In step 506, the speech signal is transmitted by wireless communication with the radio wave intensity calculated in step 505.
As a result, in the voice analysis system 1 of the present embodiment, the terminal device 10 transmits the message to the wearer (dialog partner) of the other terminal device 10 that is considered to have a dialogue with the wearer of the terminal device 10. The other utterance signal is sent to the wearer of the other terminal device 10 that is considered to have not reached the dialogue with the wearer of the terminal device 10 because the utterance signal has arrived and is far away from the wearer of the terminal device 10. Will not reach. As a result, the other terminal device 10 worn by the conversation partner of the wearer of the terminal device 10 can receive the speech signal transmitted from the terminal device 10.

<Processing executed by host device 20>
FIG. 7 is a flowchart showing processing executed by the host device 20 to which this exemplary embodiment is applied. Next, with reference to FIG. 7 and FIG. 1 described above, processing executed by the host device 20 of the present embodiment will be described.
In the host device 20 of the present embodiment, first, the data receiving unit 21 receives utterance information related to speech signals of uttered speech from a plurality of terminal devices 10 (step 701). Note that the utterance information includes information such as the ID information of the terminal device 10, the acquisition time of the utterance voice acquired by the terminal device 10, and the analysis result of the self-other identification obtained at the above-described step 504 in the terminal device 10. It is.
Then, the utterance information received by the data receiving unit 21 is temporarily sent to the data storage unit 22 and stored (step 702).

Subsequently, when the reception information transmitted from the terminal device 10 is received by the data reception unit 21 (YES in step 703), the received reception information is temporarily sent to the data storage unit 22 and stored (step 704). ).
If the reception information is not received by the data receiving unit 21 (NO in step 703), the subsequent steps are not executed and the process returns to step 701.

Next, the data analysis unit 23 acquires the utterance information acquired in step 701 and stored in the data storage unit 22 in step 702, and the reception information acquired in step 703 and stored in the data storage unit 22 in step 704. Based on the above, the combinations of the terminal devices 10 that execute the above-described determination of the audio synchrony among the plurality of terminal devices 10 are limited (step 705).
Note that the process of limiting the combination of the terminal devices 10 that execute the determination of the audio synchronism from the plurality of terminal devices 10 in step 705 will be described in detail later.

Subsequently, for the set of terminal devices 10 limited in step 705, the above-described sound synchrony is determined (step 706).
If it is determined that there is audio synchronism between the terminal devices 10 (YES in step 707), there is an interactive relationship between the wearers of the terminal device 10 that have determined the audio synchronism. It is determined that there is (step 708).
On the other hand, when it is determined that there is no audio synchronism between the terminal devices 10 (NO in step 707), the process returns to step 701.

<Description of the program>
Note that the processing performed by the terminal device 10 in the present embodiment described in FIG. 5 and the processing performed by the host device 20 in the present embodiment described in FIG. 7 are performed by software and hardware resources. Realized by working. That is, the CPUs in the control computers provided in the terminal device 10 and the host device 20 execute programs that realize the functions of the terminal device 10 and the host device 20, and these functions are realized.

<Specific example of processing executed in speech analysis system>
Next, a specific example of processing executed in the voice analysis system 1 will be described.
FIG. 8 is a diagram for explaining in detail processing executed in the speech analysis system 1 to which this exemplary embodiment is applied.
Here, in the example illustrated in FIG. 8, there are wearers X, Y, and Z wearing the terminal devices 10X, 10Y, and 10Z, respectively. In this example, it is assumed that the wearer X and the wearer Y have a dialogue, and there is no dialogue between the wearer X and the wearer Z and between the wearer Y and the wearer Z. . In this example, the distance m2 between the wearer X and the wearer Z and the distance between the wearer Y and the wearer Z are compared with the distance m1 between the wearer X and the wearer Y. m3 is increased (m1 <m2, m1 <m3). In the example shown in FIG. 8, although not shown, there is a host device 20 (see FIG. 1) connected to the terminal devices 10X, 10Y, and 10Z via a wireless communication line.
Here, in the example shown in FIG. 8, although there is no dialogue between the wearer X and the wearer Z and between the wearer Y and the wearer Z, the wearer Z (terminal device 10Z) The voices of the wearer X and the wearer Y arrive, and the microphones 11 and 12 of the terminal device 10Z may acquire the voices of the wearer X and the wearer Y.

(Specific example of processing performed in terminal device 10X)
First, in the example shown in FIG. 8, the processing performed in the terminal device 10 </ b> X worn by the wearer X when the wearer X speaks to the wearer Y who is the conversation partner has been described with reference to FIG. 5. This will be explained based on the procedure.
When the wearer X utters in the example shown in FIG. 8, in the terminal device 10X, the first microphone 11 and the second microphone 12 (see FIG. 1 respectively) acquire the uttered voice (step 501).
Subsequently, the speech analysis unit 15 (see FIG. 1) performs processing such as amplification of the speech signal on the acquired speech speech (step 502), and then calculates the sound pressure of the speech speech based on the amplified speech signal. (Step 503).
Next, the utterance voice is identified by the method as described above (step 504). In this example, since the wearer X is speaking, it is determined that the uttered voice acquired by the terminal device 10X is due to the wearer X itself (YES in step 504).

Subsequently, in the voice analysis unit 15 of the terminal device 10X, based on the sound pressure of the uttered voice of the wearer X calculated in Step 503 and the relationship shown in FIG. Calculate (step 505).
Here, in the example shown in FIG. 8, the wearer X is speaking toward the wearer Y, and the sound pressure of the uttered voice of the wearer X is a position where the distance from the wearer X is m1. The size is suitable for a conversation with the wearer Y in
Thereby, the voice analysis unit 15 can calculate the distance between the wearer X and the wearer Y as m1 based on the sound pressure of the voice of the wearer X based on the relationship shown in FIG. Then, the voice analysis unit 15 calculates the radio field intensity for transmitting the speech signal from the calculated distance m1 between the wearer X and the wearer Y. That is, from the relationship shown in FIG. 6, the utterance signal reaches the range where the distance from the terminal device 10X is less than or equal to m1, and the utterance signal does not reach the range where the distance from the terminal device 10X is more than m1. Radio field strength is calculated.

Subsequently, the data transmission unit 16 (see FIG. 1) of the terminal device 10X transmits an utterance signal using wireless communication with the radio wave intensity calculated in Step 505 (Step 506). Note that the speech signal includes information such as the ID information of the terminal device 10X and the acquisition time of the speech sound of the wearer X acquired by the terminal device 10X.
FIG. 8 shows a range in which the speech signal transmitted from the terminal device 10X reaches with the radio wave intensity calculated in step 505 (that is, a range in which the distance from the wearer X (terminal device 10X) is within m1). This is indicated by a broken line.
Here, since the distance between the wearer X (terminal device 10X) and the wearer Y (terminal device 10Y) is m1, the speech signal transmitted from the terminal device 10X as shown in FIG. The terminal device 10Y is reached. Then, the signal receiving unit 17 of the terminal device 10Y receives the speech signal from the terminal device 10X.

  On the other hand, since the distance between the wearer X (terminal device 10X) and the wearer Z (terminal device 10Z) is m2 (> m1), the utterance transmitted from the terminal device 10X as shown in FIG. The signal does not reach the terminal device 10Z, and the terminal device 10Z does not receive the speech signal from the terminal device 10X.

Subsequently, in the terminal device 10X, the utterance information is transmitted to the host device 20 (step 509), and the series of processes is terminated.
Here, since the host device 20 is usually installed at a position away from the wearers X to Z who wear the terminal devices 10X to 10Z, the radio wave intensity for transmitting the speech information to the host device 20 is , It is set to be larger than the radio wave intensity at which the speech signal is transmitted.
The utterance information transmitted from the terminal device 10X to the host device 20 includes the analysis result of the utterance voice of the wearer X acquired by the terminal device 10X, the acquisition time of the utterance voice of the wearer X, and the wearer X Information such as the sound pressure of the uttered voice, ID information of the terminal device 10X, and the like are included.

  In this example, since the wearer X is speaking to the wearer Y and the wearer Y and the wearer Z are not speaking, the utterance signal is transmitted from the terminal device 10Y and the terminal device 10Z. The terminal device 10X does not receive the speech signal (NO in step 507). Therefore, in this example, the terminal device 10X does not transmit the reception information in step 508.

(Specific example of processing performed in terminal device 10Y)
Next, in the example illustrated in FIG. 8, processing performed by the terminal device 10 </ b> Y worn by the wearer Y when the wearer X speaks to the wearer Y who is a conversation partner will be described with reference to FIG. 5. This will be described based on the procedure.
When the wearer X utters in the example shown in FIG. 8, in the terminal device 10Y, the first microphone 11 and the second microphone 12 (see FIG. 1 respectively) acquire the uttered voice (step 501).
Subsequently, the speech analysis unit 15 (see FIG. 1) performs processing such as amplification of the speech signal on the acquired speech speech (step 502), and then calculates the sound pressure of the speech speech based on the amplified speech signal. (Step 503).

  Next, the voice analysis unit 15 of the terminal device 10Y performs self-other identification of the uttered voice by the method as described above (step 504). In this example, since the wearer X is uttering, it is determined that the uttered voice acquired by the terminal device 10Y is from someone other than the wearer Y (NO in step 504). Accordingly, the terminal device 10Y does not calculate the radio field intensity of the speech signal in step 505 and does not transmit the speech signal in step 506.

Subsequently, as described above, the terminal device 10Y receives the speech signal transmitted from the terminal device 10X (YES in step 507). Here, as described above, the speech signal from the terminal device 10X includes information such as the ID information of the terminal device 10X and the acquisition time of the speech sound by the terminal device 10X.
Then, the terminal device 10Y transmits the reception information to the host device 20 by the data transmission unit 16 (step 508). Here, the reception information transmitted from the terminal device 10Y toward the host device 20 includes the ID information of the terminal device 10Y and the ID information of the terminal device 10X and the terminal device included in the speech signal received from the terminal device 10X. Information such as the acquisition time of the speech voice by 10X.

Next, the terminal device 10Y transmits the utterance information to the host device 20 (step 509), and the series of processing ends.
At this time, the speech information transmitted by the terminal device 10Y includes the analysis result of the speech of the wearer X acquired by the terminal device 10Y, the acquisition time of the speech of the wearer X, and the sound of the speech of the wearer X. Information such as pressure, ID information of the terminal device 10Y, and the like are included.

(Processing performed in the terminal device 10Z)
Next, in the example shown in FIG. 8, processing performed by the terminal device 10 </ b> Z worn by the wearer Z when the wearer X speaks to the wearer Y who is a conversation partner will be described with reference to FIG. 5. This will be described based on the procedure.
When the wearer X speaks in the example shown in FIG. 8, in the terminal device 10Z, the first microphone 11 and the second microphone 12 (refer to FIG. 1 respectively) acquire the speech voice (step 501).
Subsequently, the speech analysis unit 15 (see FIG. 1) performs processing such as amplification of the speech signal on the acquired speech speech (step 502), and then calculates the sound pressure of the speech speech based on the amplified speech signal. (Step 503).

  Next, the voice analysis unit 15 of the terminal device 10Z performs self-other identification of the uttered voice by the method as described above (step 504). In this example, since the wearer X is uttering, it is determined that the uttered voice acquired by the terminal device 10Z is by someone other than the wearer Z (NO in step 504). Therefore, the terminal device 10Y does not calculate the radio field intensity of the speech signal in step 505 and does not transmit the speech signal in step 506.

  Subsequently, as described above, since the distance from the terminal device 10X is m2 (> m1), the terminal device 10Z does not reach the utterance signal transmitted from the terminal device 10X, and the terminal device 10X No speech signal is received (NO in step 507). Therefore, the terminal device 10Z does not transmit the reception information in step 508.

Next, the terminal device 10Z transmits the utterance information to the host device 20 (step 509), and the series of processing ends.
At this time, the utterance information transmitted by the terminal device 10Z includes the analysis result of the utterance voice of the wearer X acquired by the terminal device 10Z, the acquisition time of the utterance voice of the wearer X, and the sound of the utterance voice of the wearer X. Information such as pressure, ID information of the terminal device 10Z, and the like are included.

In the above specific example, when the wearer X and the wearer Y have a dialogue, the process when the wearer X speaks to the wearer Y has been described. The processing when the wearer Y speaks toward the wearer X can be considered in the same manner.
That is, when the wearer Y utters toward the wearer X, the terminal device 10Y acquires the utterance voice of the wearer Y and transmits the utterance signal with the radio wave intensity based on the sound pressure of the utterance voice. To do. Further, the terminal device 10 </ b> Y transmits utterance information to the host device 20.

Further, when the wearer Y utters toward the wearer X, the terminal device 10X acquires the utterance voice of the wearer Y and receives the utterance signal transmitted from the terminal device 10Y, and the host device 20 A reception signal is transmitted toward. In addition, the terminal device 10 </ b> X transmits the utterance information to the host device 20.
Furthermore, when the wearer Y utters toward the wearer X, the terminal device 10Z acquires the utterance voice of the wearer Y and transmits the utterance information to the host device 20. Since the distance between the terminal device 10Y (wearer Y) and the terminal device 10Z (wearer Z) is m3 (> m1), the speech signal transmitted from the terminal device 10Y cannot reach the terminal device 10Z. The terminal device 10Z does not receive the speech signal.

(Specific example of processing performed in host device 20)
Subsequently, when processing as described above is performed in the terminal device 10X, the terminal device 10Y, and the terminal device 10Z, the host device 20 connected to the terminal devices 10X to 10Z via a wireless communication line (see FIG. 1). The process performed in is described based on the procedure described in FIG.

As described above, utterance information is transmitted from each of the terminal devices 10X to 10Z to the host device 20, and the utterance information is received by the data receiving unit 21 of the host device 20 (step 701).
Then, the host device 20 accumulates the received utterance information separately for each wearer (wearers X, Y, Z) in the data accumulation unit 22 based on the ID information included in the utterance information (step 702). ).

Subsequently, the data reception unit 21 of the host device 20 acquires the reception information transmitted from the terminal device 10Y and the reception information transmitted from the terminal device 10X as described above (YES in step 703). The acquired reception information is sent to the data storage unit 22 and stored (step 704).
Here, the reception information from the terminal device 10Y received by the data receiving unit 21 is based on the utterance signal from the terminal device 10X received by the terminal device 10Y. The received information from the terminal device 10Y includes the ID information of the terminal device 10Y, the ID information of the terminal device 10X included in the utterance signal from the terminal device 10X, and the information on the acquisition time of the uttered voice of the wearer X Etc. are included.
Similarly, the reception information from the terminal device 10X is based on the speech signal from the terminal device 10Y received by the terminal device 10X. The received information from the terminal device 10X includes the ID information of the terminal device 10X, the ID information of the terminal device 10X included in the utterance signal from the terminal device 10Y, the information on the acquisition time of the uttered voice of the wearer Y, etc. And are included.

Then, the data analysis unit 23 of the host device 20 limits the combination of the terminal devices 10 (the combination of the wearer) that performs the sound synchrony determination based on the acquired reception information (step 705).
Here, since the reception information from the terminal device 10Y includes the ID information of the terminal device 10X and the reception information from the terminal device 10X includes the ID information of the terminal device 10Y, the host device The data analysis unit 23 of 20 uses the combination of the wearer X of the terminal device 10X and the wearer Y of the terminal device 10Y as a target for performing the sound synchrony determination.
On the other hand, the data receiving unit 21 has not acquired the reception information from the terminal device 10Z, and the reception information from the terminal device 10Y and the reception information from the terminal device 10X include the ID information of the terminal device 10Z. Not. Therefore, the data analysis unit 23 of the host device 20 does not set the wearer Z of the terminal device 10 </ b> Z as a target for determining the audio synchrony with the wearer X and the wearer Y.

Subsequently, the host device 20 determines audio synchrony for the combination of the target terminal devices 10 (the combination of the wearer) limited in Step 705 (Step 706).
That is, in this example, among the utterance information received in step 701 and accumulated in step 702, the utterance information from the terminal device 10X is compared with the utterance information from the terminal device 10Y. In this example, since the wearer X and the wearer Y are interacting, the utterance information from the terminal device 10X and the utterance information from the terminal device 10Y are switched between the length of the utterance and the speaker. Information indicating the utterance status such as timing approximates. As a result, it is determined that there is audio synchronism between the utterance information from the terminal device 10X and the utterance information from the terminal device 10Y (YES in step 707), and between the wearer X and the wearer Y Is determined to have an interactive relationship (step 708).

On the other hand, between the utterance information from the terminal device 10X and the utterance information from the terminal device 10Z, and between the utterance information from the terminal device 10Y and the utterance information from the terminal device 10Z, the determination of the audio synchrony is performed. Absent.
As a result, it is possible to suppress the complexity of the process related to the determination of the dialog relationship as compared with the case where the present configuration is not adopted, and the determination of the dialog relationship can be performed with high accuracy.
Further, by having the above-described configuration, it is erroneously determined that there is a dialogue relationship between the wearer X and the wearer Z, and between the wearer Y and the wearer Z, although there is actually no dialogue relationship. Can be suppressed.

Subsequently, the present invention will be described in more detail using examples. In addition, this invention is not limited to a following example.
Example 1
A wearer a wearing the terminal device 10a, a wearer b wearing the terminal device 10b, a wearer c wearing the terminal device 10c, and a wearer d wearing the terminal device 10d are arranged as shown in FIG. That is, the distance between the wearer a and the wearer b is 1.5 m, the distance between the wearer b and the wearer c is 4 m, and the distance between the wearer c and the wearer d is 1.5 m. The wearer a to the wearer d are arranged so that the distance between the wearer a and the wearer d is 7 m. Then, a dialogue was performed between the wearer a and the wearer b for a predetermined period, and a dialogue was carried out between the wearer c and the wearer d in the same time zone. Note that the wearers a to d are located within a range where each other's voice can reach.
At this time, each terminal device 10 performs processing as shown in FIG.
And under such a situation, the acquisition situation of the utterance signal in each terminal device 10 was observed, and the result is shown in FIG.

(Comparative Example 1)
The wearer a, the wearer b, the wearer c, and the wearer d are arranged in the same manner as in the first embodiment, and are performed between the wearer a and the wearer b and between the wearer c and the wearer d. Dialogue was conducted as in Example 1. As described above, the wearer a to the wearer d are located within a range where each other's voice can reach.
In the first comparative example, unlike the first embodiment, each terminal device 10 transmits an utterance signal with a predetermined uniform radio wave intensity regardless of the sound pressure of the uttered voice of the wearer. It is assumed that the utterance signal transmitted from the terminal device 10 with this radio wave intensity reaches another terminal device 10 existing at a distance of 7 m or more. That is, the utterance signal transmitted from the terminal device 10 of one of the wearers a to d reaches the terminal devices 10 of all other wearers.
Under such circumstances, the acquisition status of the utterance signal at each terminal device 10 was observed, and the result is shown in FIG.

(Observation results)
As shown in FIG. 10A, in the first embodiment, in the terminal device 10a worn by the wearer a, only the utterance signal transmitted from the terminal device 10b worn by the wearer b who is the conversation partner of the wearer a. It can be seen that the utterance signals from the terminal device 10c of the wearer c and the terminal device 10d of the wearer d who do not interact with the wearer a are not received.
Similarly, the terminal device 10b receives only the utterance signal transmitted from the terminal device 10a, the terminal device 10c receives only the utterance signal transmitted from the terminal device 10d, and the terminal device 10d receives from the terminal device 10c. It can be seen that only the transmitted speech signal is received.

From this result, in the data analysis unit 23 of the host device 20 in the first embodiment, the wearer a to the wearer d (terminal device 10a to terminal device 10d) are combined with the wearer a and the wearer b (terminal device 10a). And a combination of the terminal device 10b) and a combination of the wearer c and the wearer d (combination of the terminal device 10c and the terminal device 10d). Then, in the data analysis unit 23, the audio synchronism is determined for the set of the terminal device 10a and the terminal device 10b and the set of the terminal device 10c and the terminal device 10d. In this case, for example, for the set of the terminal device 10a and the terminal device 10c, the set of the terminal device 10a and the terminal device 10d, etc., it is not necessary to determine the synchrony of voice.
As a result, compared with the case where this configuration is not adopted, it is possible to suppress the processing executed by the data analysis unit 23 from becoming complicated, and the dialogue relation can be accurately determined as compared with the case where this configuration is not adopted. be able to.
Further, since the wearer (terminal device 10) is divided into a plurality of groups in advance before the determination of the synchronicity of the voice, it is less likely to make an erroneous determination regarding the dialogue than in the case where this configuration is not adopted. .

On the other hand, as shown in FIG. 10 (b), in Comparative Example 1, the terminal device 10a worn by the wearer a is not only the utterance signal transmitted from the terminal device 10b of the wearer b interacting with the wearer a. It can be seen that the utterance signal transmitted from the terminal device 10c of the wearer c who does not interact with the wearer a and the terminal device 10d of the wearer d is also received.
Similarly, the terminal device 10b receives not only the utterance signal transmitted from the terminal device 10a but also the utterance signal transmitted from the terminal device 10c and the terminal device 10d, and the terminal device 10c transmits the utterance signal from the terminal device 10d. The terminal apparatus 10d receives not only the utterance signal transmitted from the terminal apparatus 10c but also the utterance signal transmitted from the terminal apparatus 10c and the terminal apparatus 10b. It can be seen that the speech signal transmitted from is also received.

As a result, in the first comparative example, it is possible to separate the target person who determines the tunedness of the uttered voice based on the reception status of the utterance signal and the reception information transmitted to the host device 20 based on the received utterance signal. It has become difficult. And in the data analysis part 23 of the host device 20, in order to determine the dialogue relationship of the wearers a to d, for example, a set of the terminal device 10a and the terminal device 10b, a set of the terminal device 10a and the terminal device 10c, a terminal Judgment of synchronicity is performed for all sets of the device 10a and the terminal device 10d, the set of the terminal device 10b and the terminal device 10c, the set of the terminal device 10b and the terminal device 10d, and the set of the terminal device 10c and the terminal device 10d. There is a need to.
As a result, the processing performed in the data analysis unit 23 becomes complicated, and the accuracy of determination of the dialog relationship is lowered, and there is a concern that erroneous determination of the dialog relationship may occur.

DESCRIPTION OF SYMBOLS 1 ... Speech analysis system, 10 ... Terminal device, 11 ... 1st microphone, 12 ... 2nd microphone, 13 ... 1st amplifier, 14 ... 2nd amplifier, 15 ... Speech analysis part, 16 ... Data transmission part, 17 ... Signal Receiving unit, 20 ... host device, 21 ... data receiving unit, 22 ... data storage unit, 23 ... data analyzing unit, 24 ... output unit

Claims (10)

A plurality of voice acquisition means that are arranged at different positions from the wearer's utterance site and that acquire the voice of the speaker;
Identification means for identifying whether the speaker is the wearer or another person other than the wearer, based on the sound pressure of the sound obtained by at least two of the sound obtaining means;
When the speaker is identified as the wearer by the identification unit, the utterance signal relating to the utterance of the wearer with the radio wave intensity based on the sound pressure of the voice of the wearer acquired by the voice acquisition unit Utterance signal transmission means for transmitting
A speech signal receiving means for receiving the speech signal transmitted from the speech signal transmitting means;
A speech analysis system comprising: a dialog relation determining means for determining a dialog relation of the wearer based on a reception state of the speech signal by the speech signal receiving means and a discrimination result by the discrimination means. When the utterance signal is received by the utterance signal reception means, reception information transmission means for transmitting reception information based on reception of the utterance signal;
Receiving information acquisition means for acquiring the reception information transmitted from the reception information transmission means;
2. The dialog relation determining unit determines the dialog relation of the wearer on the basis of an identification result by the identification unit and the reception information acquired by the reception information acquisition unit. Voice analysis system. Based on the sound pressures of the voices acquired by the plurality of voice acquisition means that are arranged at different distances from the wearer's utterance part and that acquire the voice of the speaker, and at least two of the voice acquisition means. A plurality of identification means for identifying whether the wearer is a person other than the wearer and a communication means for performing communication with the outside via a wireless communication line based on the identification result by the identification means A voice terminal device;
A voice analysis device comprising a dialogue relation determination means for judging a dialogue relation between wearers of each of the voice terminal devices,
The communication means of the voice terminal device, when the speaker is identified as the wearer by the identification means, the radio field intensity based on the sound pressure of the wearer's voice acquired by the voice acquisition means The speech signal transmitting means for transmitting the speech signal related to the utterance of the wearer, and the speech signal receiving means for receiving the speech signal transmitted from the speech signal transmitting means in the other voice terminal device,
The dialogue relation determination unit of the voice analysis device is based on an identification result by the identification unit of each voice terminal device and a reception state of the utterance signal by the utterance signal reception unit of each voice terminal device. A speech analysis system characterized by determining dialogue relations.   The dialogue relation determination unit of the voice analysis device includes the wearer of the voice terminal device that has transmitted the utterance signal by the utterance signal transmission unit, and the utterance signal reception among the wearers of the plurality of voice terminal devices. 4. The speech analysis system according to claim 3, wherein a dialogue relation is determined with respect to a wearer of the voice terminal device that has received the speech signal by means.   The dialogue relation determination unit of the voice analysis device includes the voice acquired by the voice acquisition unit of the voice terminal device that has transmitted the utterance signal by the utterance signal transmission unit, and the utterance signal by the utterance signal reception unit. 5. The voice analysis system according to claim 4, wherein the dialogue relation is determined by comparing synchrony with the voice acquired by the voice acquisition means of the voice terminal device that has received the message. A plurality of voice acquisition means that are arranged at different positions from the wearer's utterance site and that acquire the voice of the speaker;
Identification means for identifying whether the speaker is the wearer or another person other than the wearer, based on the sound pressure of the sound obtained by at least two of the sound obtaining means;
When the speaker is identified as the wearer by the identification unit, the utterance signal relating to the utterance of the wearer with the radio wave intensity based on the sound pressure of the voice of the wearer acquired by the voice acquisition unit Utterance signal transmission means for transmitting
An audio terminal device including an utterance signal receiving means for receiving an utterance signal related to an utterance of another person.   The voice terminal apparatus according to claim 6, further comprising speech information transmitting means for transmitting speech information including information related to the voice acquired by the voice acquisition means and an identification result by the identification means.   8. The voice terminal apparatus according to claim 6, wherein the utterance signal receiving unit receives an utterance signal related to the utterance of the other person transmitted by the utterance signal transmitting unit of the other apparatus worn by the other person. .   9. The utterance signal transmitting means transmits the utterance signal with a higher radio field intensity as the sound pressure of the wearer's voice acquired by the voice acquisition means is larger. The voice terminal device according to claim 1. On the computer,
A function of acquiring voice information from a plurality of voice acquisition means arranged at different positions from the wearer's utterance site and acquiring the voice of the speaker;
A function for identifying whether the speaker is the wearer or another person other than the wearer based on the sound pressure difference between the voices acquired by the at least two voice acquisition means;
A function of transmitting an utterance signal related to the utterance of the wearer with a radio wave intensity based on a sound pressure of the wearer's voice acquired by the voice acquisition unit when the speaker is identified as the wearer. When,
A program that realizes a function of receiving an utterance signal related to another person's utterance.

Images