JP2023034563A - Matching system and matching method - Google Patents

Abstract

To allow for appropriately matching a user with another user sharing common interests or tastes with the user.SOLUTION: A heard sound extraction unit 61 and an emotional voice extraction unit 63 extract information of heard sounds heard by users 11 and information of emotional voices reflecting emotional states of the users 11 from sound information. A taste specification unit 65 specifies information of tastes of the users 11 from the extracted information of heard sounds and emotional voices. A similarity determination unit 90 determines similarity of tastes between users 11 on the basis of information of tastes of the individual users 11 stored in a database unit 80, and a user extraction unit 90 extracts pairs of users 11 in each of which the similarity is equal to or higher than a determination threshold. A matching information providing unit 100 provides user information of one user 11 of a pair extracted by the user extraction unit 90 to a terminal device 20 of the other user 11 of this pair, as matching information.SELECTED DRAWING: Figure 2

Description

The present invention relates to matching systems and matching methods.

Self-driving vehicles allow users who cannot drive to ride alone. In self-driving vehicles, it is thought that there will be more opportunities for users to get into the vehicle alone. Opportunities for users to travel alone in vehicles may also increase in situations where social distancing is required. A vehicle without a passenger may increase the likelihood that a user will feel lonely in the vehicle compared to a vehicle with a passenger. In particular, in a self-driving vehicle, the user may feel even more lonely because the user does not perform driving operations.

Providing a vehicle user with a means of communication with other vehicle users is considered effective in alleviating the sense of loneliness felt by vehicle users. As a means of communication between vehicle users, Patent Literature 1 proposes a vehicle user matching method.

In the matching method of Patent Document 1, when it is determined that the second user's boarding vehicle, which is determined to be compatible with the first user, may approach the first user's boarding vehicle, Information of the second user is provided to the first user.

Whether the first user and the second user are compatible is determined based on the user information of the vehicle user. The user information of each user includes riding characteristic information of each user. Each user's riding characteristic information is generated based on each user's riding history information.

JP 2020-95475 A

In the matching method of Patent Literature 1, user attributes are determined based on the user's boarding history information, and the determined attributes are used for user matching. By the way, when determining whether the compatibility between users is good or bad, whether or not the tastes and preferences of the users match is an important factor. In order to alleviate the feeling of loneliness that users feel in vehicles, it is considered preferable to determine user attributes based on elements that match the user's tastes and preferences rather than the user's ride history information and use it for user matching. be done.

SUMMARY OF THE INVENTION It is an object of the present invention to appropriately match a user, such as an occupant of a vehicle, with another user who has the same tastes and interests as the user.

In order to achieve the above object, the matching system according to the first aspect of the present invention includes:
a listening sound extracting unit that extracts information about a listening sound heard by a user from voice information;
an emotional sound extracting unit that extracts emotional sound information reflecting the user's emotional state from voice information;
a terminal device having a preference identifying unit that identifies information about the user's preference from the extracted listening sound information and the extracted emotional sound information;
a database unit in which the information on the preferences specified for each of the plurality of users is accumulated for each user;
a similarity determination unit that determines the similarity of preferences between the users based on the information on the preferences for each user in the database unit;
a user extracting unit that extracts a combination of the users whose similarity is equal to or greater than a determination threshold;
a matching information providing unit that provides, as matching information, information of other users in the extracted combination to the terminal device of one of the users in the combination extracted by the user extraction unit; ,
Prepare.

In order to achieve the above object, a matching method according to a second aspect of the present invention includes:
a listening sound extraction step of extracting information of a listening sound heard by a user from voice information;
an emotional sound extraction step of extracting emotional sound information reflecting the user's emotional state from voice information;
a preference identifying step of identifying information about the user's preference from the extracted listening sound information and the extracted emotional sound information;
a similarity determination step of determining the similarity of the preferences of the users based on the information of the preferences for each user that are specified for each of a plurality of users and accumulated in a database unit;
a user extraction step of extracting a combination of the users whose similarity is equal to or greater than a determination threshold;
a matching information providing step of providing, as matching information, information of other users in the extracted combination to a terminal device of one of the users in the combination extracted in the user extraction step;
including.

Advantageous Effects of Invention According to the present invention, it is possible to appropriately match a user with other users who have similar interests and tastes.

FIG. 1 is an explanatory diagram of a matching system according to one embodiment of the present invention. FIG. 2 is a block diagram showing a functional configuration of the matching system of FIG. 1; FIG. 3 is a block diagram showing a more detailed configuration of the environmental sound analysis section of FIG. 2. As shown in FIG. 4A is a graph for explaining an example of an environmental sound generation pattern analyzed by the environmental sound analysis unit in FIG. 2. FIG. 4B is a graph for explaining another example of the environmental sound generation pattern analyzed by the environmental sound analysis unit in FIG. 2; FIG. FIG. 5A is an explanatory diagram showing an example of user preferences specified by the preference specifying unit in FIG. 3 . FIG. 5B is an explanatory diagram showing another example of user preferences specified by the preference specifying unit in FIG. 3 . FIG. 6 is a flow chart showing an example of a procedure of similarity tendency determination processing performed by the similarity tendency determination unit of FIG. 2 . FIG. 7 is a flow chart showing an example of a matching method procedure executed by each part of the matching system of FIG.

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described with reference to the drawings. The same reference numerals are given to the same or equivalent parts or components throughout each drawing.

The embodiments shown below are examples of apparatuses and the like for embodying the technical idea of the present invention. The technical idea of the present invention does not specify the material, shape, structure, arrangement, function, etc. of each component as follows.

FIG. 1 is an explanatory diagram of a matching system according to one embodiment of the present invention. As shown in FIG. 1, the matching system of this embodiment can be configured by connecting a terminal device 20 of a vehicle 10 and a server 30 via a communication network 40 .

The matching system can determine the preferences of the user 11 from the sounds and emotional sounds of the user 11 riding in the vehicle 10, for example, based on voice information acquired by the vehicle 10. FIG. The matching system can provide the user 11 of the vehicle 10 with the information of the user 11 who got on the other vehicle 10 as the matching information in line with the determined preference of the user 11 . In this embodiment, it is assumed that each user 11 rides in a different vehicle 10 .

The vehicle 10 may be an automatically operated vehicle in which part or all of the driving operation is automatically performed by computer control, or may be a non-automatically operated vehicle in which all driving operations are performed by a person riding in the vehicle 10 .

The terminal device 20 may be a vehicle-mounted device mounted on the vehicle 10 or a mobile terminal brought into the vehicle 10 by the user 11 . The terminal device 20 includes a CPU (Central Processing Unit) and an internal storage device (not shown). Onboard equipment includes, for example, a car navigation system and a display audio system. Mobile terminals include, for example, smartphones and tablet terminals. It is assumed that the terminal device 20 has a sound collecting function of acquiring voice information in the room of the vehicle 10 . A sound collection function can be realized using a microphone, for example.

The server 30 can be configured by, for example, a general-purpose computer having a CPU and an internal storage device (not shown). The server 30 can communicate with the terminal device 20 of each vehicle 10 via the communication network 40 .

The communication network 40 can be configured to include, for example, a network that performs wireless communication with the terminal device 20 . Communication network 40 may include the Internet. The network used by the terminal device 20 and the network used by the server 30 may be the same network or different networks.

When the terminal device 20 is a vehicle-mounted device, the user 11 in the vehicle 10 can operate the vehicle-mounted device to input necessary information. The required information may include, for example, information needed by server 30 to match user 11 with users 11 of other vehicles 10 .

When the terminal device 20 is a mobile terminal, the user 11 in the vehicle 10 activates application software (hereinafter referred to as an "app") installed in the mobile terminal brought into the vehicle 10 to use the mobile terminal. You can operate and input necessary information. When the user 11 activates the application of the mobile terminal, which is the terminal device 20, the native application is installed in the mobile terminal of the user 11, and the arithmetic device in the mobile terminal performs arithmetic processing.

Also, instead of installing a native application in the mobile terminal, which is the terminal device 20, for example, PWA (Progressive Web Apps) technology can be used. When PWA technology is used, a website for mobile devices built on a cloud server on the Internet can be used like an application installed on the mobile terminal of the user 11 .

When PWA technology is used, for example, a mobile terminal can access a cloud server on the Internet, and necessary information can be input on the console screen of the cloud server displayed on the mobile terminal. The function of the cloud server may be provided to the server 30, or may be provided to another server (not shown) on the Internet that cooperates with the server 30. FIG.

In this embodiment, a case where the terminal device 20 is a vehicle-mounted device will be described. In the following, the matching system of the present embodiment will be described with reference numeral 20 attached to the vehicle-mounted device as the terminal device 20 .

FIG. 2 is a block diagram showing a functional configuration of the matching system of FIG. 1; As shown in FIG. 2, the matching system of this embodiment includes an environmental sound acquisition unit 50, an environmental sound analysis unit 60, an analysis result transmission unit 70, an analysis result storage unit 80, a similarity tendency determination unit 90, and a communication control unit 100. It can be configured to have.

The environmental sound acquisition unit 50 can be constructed in the vehicle-mounted device 20 . The environmental sound acquisition unit 50 acquires audio information inside the vehicle using a microphone or the like. The in-vehicle audio information may be information about environmental sounds acquired in the vehicle, and it does not matter whether the source of the audio is inside or outside the vehicle. The matching system does not necessarily have the environmental sound acquisition unit 50 .

The environmental sounds include voices uttered by the user 11 inside the vehicle and listening sounds output inside the vehicle for the purpose of listening by the user 11 .

The voice uttered by the user 11 in the car is composed of an emotional sound that reflects the emotional state of the user 11 and a voice that does not necessarily reflect the emotional state of the user 11 and does not correspond to the emotional sound. Including both. Emotional sounds may include, for example, the user's 11 laughter, anger, humming, sighing, conversation, and the like. Sounds that do not correspond to emotional sounds can include, for example, sounds of the user 11 clearing his throat, sneezing, and the like.

Listening sounds include, for example, broadcast sounds of television or radio, and reproduced sounds such as music. The broadcast sound and the reproduced sound may be received or reproduced by a device (not shown) mounted on the vehicle 10 or received or reproduced by a device (not shown) brought into the vehicle 10 by the user 11. good.

The environmental sound includes the voice of the user 11 and the listening sound described above, as well as the voice outside the vehicle that is generated around the vehicle 10 and reaches the inside of the vehicle. The sound outside the vehicle includes, for example, noise outside the vehicle. The noise outside the vehicle includes, for example, the sound of the horn of the vehicle 10, street noise, and the like. It does not matter whether the sound of the horn is the sound of the vehicle 10 in which the user 11 is riding or the sound of another vehicle 10 . The noise on the street includes, for example, noise caused by people walking around the vehicle 10, noise caused by means of transportation other than the vehicle 10 such as railways, and the like.

The environmental sound acquisition unit 50 can start acquiring voice information or stop acquiring voice information, for example, when the user 11 sets the function of the matching system on and off in the vehicle-mounted device 20. . The environmental sound acquisition unit 50 can start acquiring audio information, for example, when the user 11 changes the setting of the matching system function from off to on. The environmental sound acquisition unit 50 can stop acquiring voice information, for example, when the user 11 changes the setting of the matching system function from on to off.

The environmental sound analysis unit 60 and the analysis result transmission unit 70 can be constructed in either the vehicle-mounted device 20 or the server 30 . Since environmental sounds include conversation, it is desirable to construct the environmental sound analysis unit 60 and the analysis result transmission unit 70 in the vehicle-mounted device 20 from the viewpoint of protecting privacy. When the environmental sound analysis unit 60 and the analysis result transmission unit 70 are configured in the server 30, the environmental sound acquisition unit 50 transmits the acquired voice information from the vehicle-mounted device 20 to the environmental sound analysis unit 60 of the server 30 via the communication network 40. can be sent to

The environmental sound analysis unit 60 analyzes the audio information acquired by the environmental sound acquisition unit 50 and quantifies it. The environmental sound analysis unit 60 does not use the audio information acquired by the environmental sound acquisition unit 50, but audio information output by a playback device (not shown) such as an earphone, or a narrow directivity microphone directed toward the mouth of the user 11. Sound information collected by a sound collector (not shown) may be analyzed.

The environmental sound analysis unit 60 may analyze the audio information in real time. The environmental sound analysis unit 60 may analyze the audio information acquired by the environmental sound acquisition unit 50 within a certain period of time at regular intervals. The fixed cycle length may be, for example, 5 seconds or 5 minutes. When the sound information is analyzed periodically, the environmental sound analysis unit 60 compares the analysis result of the current period with the analysis result of the previous period, and based on the similarity between the two analysis results, It is possible to judge whether the same environmental sound continues or not.

As shown in FIG. 3 , the environmental sound analysis unit 60 has a listening sound extraction unit 61 , an emotional sound extraction unit 63 and a preference identification unit 65 .

The listening sound extracting unit 61 analyzes the audio information acquired as the environmental sound information by the environmental sound acquiring unit 50, and extracts the listening sound that the user 11 heard in the vehicle 10 in which the user 11 boarded from the acquired audio information. Extract information. The listening sound extracting unit 61 may extract the listening sound of the user 11 not from the audio information acquired by the environmental sound acquiring unit 50 but from audio information output by a playback device (not shown) such as an earphone. The listening sound extracting unit 61 may extract the listening sound of the user 11 from the audio information including the sound heard by the user 11 .

The emotional sound extracting unit 63 analyzes the voice information acquired as environmental sound information by the environmental sound acquiring unit 50, and extracts the emotional sound reflecting the emotional state of the user 11 riding in the vehicle 10 from the acquired voice information. extract the information of The emotional sound extraction unit 63 extracts the sound of the user 11 not from the sound information acquired by the environmental sound acquisition unit 50 but from sound information collected by a sound collecting device (not shown) such as a narrow directional microphone aimed at the mouth of the user 11 . Eleven emotional sounds may be extracted. The listening sound extracting unit 61 may extract the listening sound of the user 11 from the audio information including the sound reflecting the emotional state of the user 11 .

The preference identification unit 65 identifies information about the preference of the user 11 from the listening sound information extracted by the listening sound extraction unit 61 and the emotional sound information extracted by the emotional sound extraction unit 63 .

In the example shown in the graph of FIG. 4A, the environmental sound ES1 is continuously acquired by the environmental sound acquisition unit 50 during the period from t0 to t4. Further, in the example of FIG. 4A, another environmental sound ES3 is acquired by the environmental sound acquisition unit 50 only during the period from t1 to t2.

The listening sound extraction unit 61 in FIG. 3 analyzes the audio information acquired by the environmental sound acquisition unit 50 . For example, the listening sound extraction unit 61 determines that the environmental sound ES1 that continues during the period from t0 to t4 in FIG. , can be determined as an environmental sound without continuity. The listening sound extracting unit 61 normalizes the waveform of the audio information of the environmental sound ES1 determined to be the environmental sound with continuity for the period from t0 to t4, and determines that it is the listening sound. The listening sound extracting unit 61 determines that the audio information of the environmental sound ES3 determined as the environmental sound without continuity is the environmental sound other than the listening sound.

The listening sound extraction unit 61 may have a function of analyzing the frequency of the environmental sound using a known technique and determining whether or not it is the environmental sound output from the speaker of the in-vehicle audio device. When having this function, the listening sound extracting unit 61 can determine, for example, an environmental sound that is determined to be output from a speaker of an in-vehicle audio device to be a listening sound.

The listening sound extraction unit 61 in FIG. 3 analyzes the information of the extracted listening sound, and identifies the music, program, etc., that the user 11 has listened to. For example, the listening sound extracting unit 61 compares the waveform of the normalized listening sound with the waveform of music data registered in a database (not shown) or the like, and detects music data including matching or similar waveforms, so that the user can 11 can identify the music that has been listened to. The listening sound extracting unit 61 can identify the program that the user 11 listened to through the music that was played during the program, for example, by checking the waveform of the music data of the music in the playlist of the broadcasted program. .

In the example shown in the graph of FIG. 4B, the environmental sound ES2 different from the environmental sounds ES1 and ES3 is continuously acquired by the environmental sound acquisition unit 50 during the same period of t0 to t4 as the environmental sound ES1.

The emotional sound extraction unit 63 of FIG. 3 analyzes the audio information that the listening sound analysis unit 61 has determined to be environmental sounds other than listening sounds, among the audio information acquired by the environmental sound acquiring unit 50 . For example, the emotional sound extraction unit 63 can determine that the environmental sound ES2 that continues during the period from t0 to t4 in FIG. 4B is an environmental sound that has continuity like the environmental sound ES1. The environmental sound analysis unit 60 normalizes the waveform of the audio information of the environmental sound ES2 judged to be the continuous environmental sound for the period from t0 to t4, and normalizes the audio information of the environmental sound ES1 for the same period from t0 to t4. Evaluate the similarity with the waveform of The similarity of waveforms can be evaluated based on, for example, the degree of synchronism of amplitude fluctuations of the waveforms shown on the vertical axes in FIGS. 4A and 4B, the degree of approximation of the frequencies of the waveforms, and the like.

The emotional sound extraction unit 63 of FIG. It can be determined that the emotional sound reflects a positive emotional state.

The emotional sound extraction unit 63, for example, analyzes the voice information acquired by the environmental sound acquisition unit 50, acquires the utterance content and utterance volume of the user 11, and determines the negative emotional state of the user 11 from the characteristics of the content. It can be determined that it is a reflected emotional sound. A known method can be used as a method of acquiring the utterance content and the utterance volume from the voice information and determining that the emotional sound reflects the emotional state of the user 11 .

For example, the emotional sound extraction unit 63 may suspend the analysis of the audio information acquired by the environmental sound acquisition unit 50 while determining that the emotional sound reflects the negative emotional state of the user 11. . This temporary suspension makes it possible to prevent matching information of users 11 having similar emotional states from being provided from other vehicles 10 and users 11 having negative emotional states to be matched with each other.

As a result of analyzing the voice information acquired by the environmental sound acquiring unit 50, the emotional sound extracting unit 63 extracts environmental sounds that are determined not to fit into any of the types of auditory sounds and emotional sounds as general sounds other than auditory sounds and emotional sounds. It can be judged as an environmental sound.

For example, when the emotional sound extracted by the user 11 during the extraction of the listening sound by the listening sound extracting unit 61 is extracted by the emotional sound extracting unit 63, the preference identifying unit 65 determines the content of the extracted listening sound and the content of the emotional sound. can be specified as information of user 11's preference.

FIG. 5A is an explanatory diagram showing an example of preferences of the user 11 identified by the preference identification unit 65. FIG. When the emotional sound extracted by the user 11 while listening to the song A specified by the listening sound extracting unit 61 is determined to be positive by the emotional sound extracting unit 63, the preference information of the user 11 specified by the preference specifying unit 65 is Preference information No. in FIG. 5A. Like 1, it is a combination of song A and positive.

When the emotional sound extraction unit 63 determines that the emotional sound emitted by the user 11 while listening to the song B specified by the listening sound extraction unit 61 is negative, the preference information of the user 11 specified by the preference specification unit 65 is Preference information No. in FIG. 5A. 2, it is a combination of music B and negative.

If the emotional sound extraction unit 63 does not extract the emotional sound of the user 11 while listening to the song C specified by the listening sound extraction unit 61, the preference specification unit 65 selects the preference information No. 1 shown in FIG. 5A. 3, only the song C may be specified as the user's 11 preference information.

The preference identification unit 65 can include information indicating the emotional state of the user 11 reflected in the emotional sound that is chronologically linked to the listening sound in the preference information as the analysis result of the emotional sound. The information indicating the emotional state of the user 11 to be included in the preference information is not limited to the positive or negative emotional state reflected in the emotional sound extracted by the emotional sound extractor 63 . Information to be included in the preference information may be, for example, information obtained by quantifying the emotional state reflected in the emotional sound extracted by the emotional sound extraction unit 63 .

When the emotional sound of the user 11 is extracted by the emotional sound extracting unit 63 while the listening sound is being extracted by the listening sound extracting unit 61, the preference specifying unit 65 stores the information including the time point of occurrence of the emotional sound as the preference of the user 11. It may be specified as information. The emotional sound of the user 11 extracted by the emotional sound extraction unit 63 during listening to one listening sound may be performed once or multiple times. The time point at which the emotional sound is generated can be represented, for example, by the elapsed time from the start of the listening sound. By representing the time point at which the emotional sound is generated by the elapsed time from the start of the listening sound, the preference specifying unit 65 can determine the preference information of the user 11 from the information indicating the time-series correlation between the listening sound and the emotional sound. can be specified.

FIG. 5B is an explanatory diagram showing another example of the preferences of the user 11 specified by the preference specifying unit 65. As shown in FIG. Preference information No. in FIG. 5B. 10, the emotional sound extracting unit 63 detects the emotional sounds uttered by the user 11 after 10 seconds, 1 minute and 13 seconds, and 2 minutes and 8 seconds after starting listening to the song A specified by the listening sound extracting unit 61 as positive sounds. , positive and negative, respectively. Preference information No. At 11, the emotional sounds uttered by the user 11 after 5 seconds and after 3 minutes and 28 seconds from the start of listening to the song B specified by the listening sound extracting unit 61 are determined as negative by the emotional sound extracting unit 63, respectively. indicate the case.

If the emotional sound extraction unit 63 does not extract the emotional sound of the user 11 while listening to the song C specified by the listening sound extraction unit 61, the information of the preference of the user 11 specified by the preference specification unit 65 is shown in FIG. 5B. Preference information No. As shown in 12, the information is only for the song C, which does not have emotional sound and its occurrence time.

The preference specifying unit 65 can provide preference information that enables more detailed evaluation of the preference of the user 11 by specifying the information including the time point of occurrence of the emotional sound as the preference information of the user 11 .

Each time the environmental sound analysis unit 60 analyzes the audio information acquired by the environmental sound acquisition unit 50, the environmental sound analysis unit 60 converts the preference information of the user 11 identified by the preference identification unit 65 into the analysis result of the audio information by the environmental sound analysis unit 60. It is handed over to the analysis result transmission unit 70 as digital data.

Each time the environmental sound analysis unit 60 delivers the digital data of the analysis result, the analysis result transmission unit 70 stores the delivered digital data of the analysis result in the analysis result holding unit 80 . When constructing the environmental sound analysis unit 60 and the analysis result transmission unit 70 in the vehicle-mounted device 20, the analysis result transmission unit 70 transmits the digital data of the analysis result handed over from the environmental sound analysis unit 60 from the vehicle-mounted device 20 to the communication network 40. can be transmitted to the analysis result holding unit 80 of the server 30 via the .

The analysis result holding unit 80 can be constructed using an internal storage device of the server 30 . The internal storage device can be composed of, for example, an SSD (Solid State Drive) or HDD (Hard Disk Drive). The analysis result holding unit 80 stores the digital data of the analysis result transmitted by the analysis result transmission unit 70 .

The analysis result holding unit 80 can function as a database unit that accumulates information on the preferences of the users 11 extracted for each of the plurality of vehicles 10 for each user 11 .

For each user 11, the similarity tendency determination unit 90 selects another user 11 having a similarity tendency (preference). The similarity tendency determination unit 90 periodically refers to the digital data of the analysis results stored for each user 11 in the analysis result holding unit 80, and selects users 11 having similar tendencies. The similarity tendency determination unit 90 can be constructed in the server 30, for example. The similarity tendency determination unit 90 can select the user 11 having a similarity tendency, for example, by performing the processing of the procedure shown in the flowchart of FIG. 6 .

The similarity tendency determination unit 90 may perform the processing of the flowchart of FIG. 6, for example, each time the analysis result digital data transmitted by the analysis result transmission unit 70 is stored in the analysis result storage unit 80, or may be performed at a constant cycle. You can go every The constant cycle may be, for example, 5 minutes or 1 hour.

The similarity tendency determination unit 90 refers to the digital data of the analysis results stored in the analysis result holding unit 80, extracts similar digital data for all the users 11, and creates a group sorted by preference tendency. (Step S1). Sorting by tendency can be performed using, for example, the genre of the music or program listened to, the emotional state of the user 11 reflected in the emotional sound, or the like as an index.

When the analysis result holding unit 80 stores the digital data of the analysis result transmitted by the analysis result transmission unit 70 up to the past for a certain period, the similarity tendency determination unit 90 also refers to the digital data of the analysis result for the past. By doing so, it is possible to create a group with a stronger similarity in taste.

The similarity tendency determination unit 90 identifies the digital data of the user 11 who selects a matching partner and the other user 11 who has the highest similarity tendency regarding taste within the created group.

When constructing the environmental sound analysis unit 60 and the analysis result transmission unit 70 in the vehicle-mounted device 20, the analysis result transmission unit 70 adds attribute information of the user 11 riding in the vehicle 10 in which the vehicle-mounted device 20 is mounted, Digital data of the analysis result can be transmitted to the analysis result holding unit 80 . The analysis result transmission unit 70 can dynamically or statically add the attribute information of the user 11 .

The user attribute information can include information such as the user's sex, age, living area, or biometric data, for example. When the emotional sound extraction unit 63 determines that the voice information acquired by the environmental sound acquisition unit 50 is the emotional sound reflecting the emotional state, the attribute of the user 11 is reflected in the emotional sound determined by the emotional sound extraction unit 63. may include the emotional state of the

When the attribute information of the user 11 is added to the digital data in the analysis result holding unit 80, the similarity tendency determination unit 90 selects the digital data with the highest similarity tendency regarding taste in the group created by the similarity tendency determination unit 90. may be specified in consideration of the attributes of the user 11 . By taking into account the attributes of the user 11, the similarity tendency determination unit 90 can identify digital data with the highest similarity tendency regarding taste with high accuracy.

The similarity tendency determination unit 90 determines that the digital data of other users 11 whose similarity in taste with the user 11 who selects a matching partner exceeds a predetermined determination threshold exists in the same group created in step S1. It is determined whether or not (step S3). If it is determined that the digital data of another user 11 whose similar tendency regarding preference exceeds the determination threshold does not exist in the same group (NO in step S3), the similar tendency determination unit 90 skips the subsequent steps. Due to this skip, the similarity tendency determination unit 90 suspends the selection of the user 11 who has a similarity in taste until the procedure of FIG. 6 is performed in the next cycle.

If the similarity tendency determination unit 90 determines that the digital data of the other user 11 whose preference similarity tendency exceeds the determination threshold exists in the same group as the user 11 who selects the matching partner (YES in step S3), the following to step S5. The procedure of step S5 may be skipped, and when the procedure of step S5 is skipped, the procedure proceeds to step S7, which will be described later.

In step S5, the similarity tendency determination unit 90 selects digital data of users 11 whose emotional states of emotional sounds exhibit a positive tendency in the group determined in step S3 as having digital data of other users 11 exceeding the determination threshold. Determine whether data exists.

When the similar tendency determination unit 90 determines that the digital data of the user 11 who shows a positive tendency does not exist in the group (NO in step S5), only the digital data of the user 11 whose emotional state shows a negative tendency is sorted. Recognize that it is a group that has been In this case, the similarity tendency determination unit 90 skips the subsequent procedures and suspends the selection of the user 11 who has a similarity in taste until the next cycle. When the similarity tendency determination unit 90 determines that the digital data of the user 11 showing a positive tendency exists in the group (YES in step S5), the procedure proceeds to the next step S7.

In step S7, the similarity tendency determination unit 90 extracts the digital data of the other user 11 who has the highest similarity tendency regarding taste in the same group as the digital data of the user 11 who selects the matching partner as the digital data of the matching partner. do.

The digital data of the user 11 with the highest similarity tendency extracted by the similarity tendency determination unit 90 may literally be a single digital data with the 'highest' similarity tendency, and the degree of similarity range applicable to the concept of 'highest'. It may be one or more digital data belonging to it. Through the process of step S7, the similarity tendency determination unit 90 can extract a one-to-one combination or a one-to-many combination of users 11 whose digital data representing information on preferences are similar.

The similarity tendency determination unit 90 transmits the information of the other user 11 corresponding to the extracted digital data of the matching partner or the information of the vehicle 10 in which the other user 11 rides to the communication control unit 100 as connection information with the matching partner. (step S9).

The similarity tendency determination unit 90 repeatedly executes the above steps S1 to S9 in the server 30. FIG.

The similarity tendency determination unit 90 can function as a similarity determination unit that determines the similarity of preferences between the users 11 based on the information on the preferences of each user 11 in the database unit (analysis result storage unit 80). . The similarity tendency determination unit 90 can function as a user extraction unit that extracts a combination of users 11 whose similarity is equal to or greater than a determination threshold.

Based on the information of the other user 11 corresponding to the connection information transmitted by the similarity tendency determination unit 90 or the information of the vehicle 10 in which the other user 11 has ridden, the communication control unit 100 determines whether the vehicle 10 in which the other user 11 has ridden is installed. Identify the vessel 20 . The communication control unit 100 provides the vehicle-mounted device 20 of the vehicle 10 in which the user 11 who selects a matching partner gets on with information necessary to establish communication with the specified vehicle-mounted device 20 .

The information provided by the communication control unit 100 may be information for connecting the vehicle-mounted devices 20 via the communication network 40 and the server 30 . It may be information for making a connection (P2P connection). Information provided by the communication control unit 100 includes communication connection information with a terminal device (vehicle device 20 ) operated by the user of the other vehicle 10 in the other vehicle 10 .

It is assumed that the communication control unit 100 holds information necessary for establishing communication with respect to the vehicle-mounted device 20 of the vehicle 10 in which each user rides. The communication control unit 100 can be built in the server 30 together with the analysis result storage unit 80 and the similarity trend determination unit 90, for example. The communication control unit 100 can provide information to the vehicle-mounted device 20 of each vehicle 10 by communication via the communication network 40, for example.

The communication control unit 100 can function as a matching information providing unit. The communication control unit 100 as a matching information providing unit provides matching information to the vehicle-mounted device 20 of the vehicle 10 in which one of the users 11 in the combination extracted by the similarity tendency determination unit 90 as a user extraction unit is riding. . The matching information provided by the communication control unit 100 includes information on other users 11 in combinations extracted by the similarity tendency determination unit 90 as a user extraction unit.

The vehicle-mounted device 20 that has received the information from the communication control unit 100 may automatically establish communication with the vehicle-mounted device 20 of the other vehicle 10 in which the matching partner user 11 rides using the received information. Alternatively, when the information from the communication control unit 100 is received by the vehicle-mounted device 20, the user 11 who got on the vehicle 10 of the vehicle-mounted device 20 that received the information is the vehicle-mounted device 20 of the other vehicle 10 on which the user 11 of the matching partner got on. may decide whether or not to actually perform the communication. When the user 11 decides whether or not to communicate, for example, the user 11 operates the vehicle-mounted device 20 to communicate with the vehicle-mounted device 20 of another vehicle 10 in which the matching partner user 11 rides. communication can be initiated.

In the matching system of this embodiment having the configuration described above, the matching method shown in the flowchart of FIG. 7 can be executed. This matching method includes steps (steps S21 to S33) of obtaining environmental sounds, extracting listening sounds, extracting emotional sounds, identifying preferences, determining similarities, extracting users, and providing matching information.

In the environmental sound acquisition step (step S21), the environmental sound acquisition unit 50 acquires audio information as environmental sound information of the vehicle 10 in which the user 11 has boarded. The environmental sound acquisition step (step S21) can also be skipped. In the listening sound extraction step (step S23), the listening sound extraction unit 61 of the environmental sound analysis unit 60 analyzes the acquired environmental sound information and extracts listening sound information from the acquired environmental sound information. In the listening sound extracting step (step S23) when skipping the environmental sound obtaining step (step S21), the listening sound of the user 11 is extracted from the voice information including the sound heard by the user 11. FIG.

In the emotional sound extraction step (step S25), the emotional sound extraction unit 63 of the environmental sound analysis unit 60 analyzes the acquired environmental sound information, and the emotional state of the user 11 is reflected from the acquired environmental sound information. Extract the information of the emotional sounds. In the emotional sound extraction step (step S25) when the environmental sound acquisition step (step S21) is skipped, the emotional sound of the user 11 is extracted from the voice information containing the sound reflecting the emotional state of the user 11.

In the preference identification step (step S27), the preference identification unit 65 of the environmental sound analysis unit 60 identifies information on the preference of the user 11 from the extracted listening sound information and emotional sound information.

In the similarity determination step (step S29), the similarity tendency determination unit 90 extracts the information on the preferences of each user 11 accumulated in the database unit (analysis result holding unit 80) and the preference identification unit in the preference identification step (step S27). The preference information specified by 65 is collated. Information on preferences for each user 11 is specified for each of a plurality of users 11 and accumulated in the database section (analysis result holding section 80). In the similarity determination step (step S29), the similarity tendency determination unit 90 determines the similarity between the collated preferences.

In the user extraction step (step S31), the similarity tendency determination unit 90 extracts combinations of users 11 whose similarity is equal to or greater than the determination threshold. In the matching information provision step (step S33), the communication control unit 100 selects the vehicle-mounted device 20 of the vehicle 10 on which one of the users 11 in the combination extracted by the similarity tendency determination unit 90 in the user extraction step (step S31) is riding. to provide matching information. Matching information includes information of other users 11 in the extracted combinations.

When the matching system of this embodiment executes the matching method of FIG. Communication can be established with the vehicle-mounted device 20 of another vehicle 10 on which the vehicle is boarded. The user 11 of each vehicle 10 can communicate with the user 11 riding in the vehicle 10 of the vehicle-mounted device 20 with which communication has been established. When the similarity tendency determination unit 90 identifies the digital data of the user with the highest similarity tendency with respect to taste using taste as an index, the communication control unit 100 determines the other vehicle 10 whose taste matches the user 11 riding in the vehicle 10 . of users 11 can be appropriately matched.

By matching the user 11 of the other vehicle 10 having a highly similar emotional state as a matching partner, it is possible to provide the user 11 riding in the vehicle 10 with a comfortable space in the vehicle that is less stressful. In particular, when there are more opportunities for the user 11 to ride the vehicle 10 alone due to the spread of self-driving vehicles and the increase in situations requiring social distancing, the possibility of the user 11 feeling lonely in the vehicle is reduced, A comfortable moving space can be provided in the vehicle. When the user 11 performs a driving operation of the vehicle 10, it is possible to contribute to promoting safe driving behavior of the user 11 as well.

In the above embodiment, an example in which the vehicle-mounted device 20 is the terminal device 20 of the vehicle 10 has been described. When a portable terminal such as a smart phone or a tablet terminal brought into the vehicle 10 by the user 11 is used as the terminal device 20 of the vehicle 10, an application installed in the portable terminal is used to realize a matching system and a matching method. can also

For example, information for accessing a mobile terminal of a user 11 of another vehicle 10 having a high similarity in preference tendency on an application installed in the mobile terminal is sent from the communication control unit 100 as matching information to the mobile terminal of the user 11. may be provided to the terminal.

For example, the user 11 can receive information for accessing the mobile terminal of the matching partner user 11 on the text chat application installed on the mobile terminal from the communication control unit 100 as matching information. The user 11 who received the matching information can start text chat by transmitting a text message to the mobile terminal of the matching partner user 11 using the matching information on a text chat application started on the mobile terminal.

For example, the user 11 can receive, as matching information from the communication control unit 100, information for accessing the mobile terminal of the matching partner user 11 on a matching application installed on the mobile terminal. The user 11 who has received the matching information accesses the mobile terminal of the matching partner user using the matching information on the matching application activated on the mobile terminal, and can make an appointment, for example.

When accessing the mobile terminal of the user 11 who is the matching partner on the application of the mobile terminal, the analysis result holding unit 80 stores the information necessary for accessing the matching partner by using the voice information acquired by the vehicle 10 in which the user 11 is riding. It is stored in association with the digital data of the analysis result. The communication control unit 100 can acquire information associated with the digital data of the user with the highest similarity tendency from the analysis result holding unit 80 and provide it as matching information.

For example, the user 11 can receive, as matching information from the communication control unit 100, a "favorite" song list registered by the matching partner user 11 on a music distribution service application installed in the mobile terminal. After receiving the matching information, the user 11 uses the matching information to obtain the "favorite" music list registered by the user 11 who is the matching partner on the application of the music distribution service started on the mobile terminal, and the obtained music list. You can select a song to play from

When receiving, as matching information, a "favorite" song list or the like of a music distribution service registered by the matching partner user 11 on the application of the mobile terminal, the analysis result holding unit 80 stores the received song list or the like as the analysis result. Store in association with digital data. The communication control unit 100 can acquire a “favorite” music list associated with the digital data of the user 11 having the highest degree of similarity from the analysis result holding unit 80 and provide it as matching information.

A matching system and a matching method in which the communication control unit 100 provides a list of "favorites" as matching information can be realized even when an application other than the music distribution service described above is used. For example, information content distribution services such as e-book distribution services and video distribution services can also provide a list of "favorites" as matching information.

The list of "favorites" provided by the communication control unit 100 as matching information is related to information content provided in a terminal device 20 (portable terminal) operated by a user 11 of another vehicle 10 in another vehicle 10. It contains preference information for ten users 11 .

The terminal device 20 may not necessarily be installed in the vehicle 10 or brought into the vehicle 10 by the user 11 . The terminal device 20 may be installed in a place where the user's 11 listening sounds and emotional sounds can be acquired, or may be brought in by the user 11 .

The place where the user's 11 listening sound and emotional sound can be acquired may be the user's room, for example, other than the vehicle 10 . The vehicle 10, the room of the user 11, and the like are all partitioned indoor spaces, but the places where the user's 11 listening sounds and emotional sounds can be acquired are not limited to the indoor spaces. The place where the user's 11 listening sound and emotional sound can be acquired may be an open space such as outdoors.

10 vehicle 11 user 20 vehicle-mounted device (terminal device)
30 server 40 communication network 50 environmental sound acquisition unit 60 environmental sound analysis unit 61 listening sound extraction unit 63 emotional sound extraction unit 65 preference identification unit 70 analysis result transmission unit 80 analysis result storage unit (database unit)
90 similarity tendency determination unit (similarity determination unit, user extraction unit)
100 communication control unit (matching information providing unit)
ES1 Environmental sound (listening sound)
ES2 environmental sound (emotional sound)
ES3 environmental sound (general environmental sound)

Claims (5)

a listening sound extracting unit that extracts information about a listening sound heard by a user from voice information;
an emotional sound extracting unit that extracts emotional sound information reflecting the user's emotional state from voice information;
a terminal device having a preference identifying unit that identifies information about the user's preference from the extracted listening sound information and the extracted emotional sound information;
a database unit in which the information on the preferences specified for each of the plurality of users is accumulated for each user;
a similarity determination unit that determines the similarity of preferences between the users based on the information on the preferences for each user in the database unit;
a user extracting unit that extracts a combination of the users whose similarity is equal to or greater than a determination threshold;
a matching information providing unit that provides, as matching information, information of other users in the extracted combination to the terminal device of one of the users in the combination extracted by the user extraction unit; ,
A matching system with 2. The matching system according to claim 1, wherein the preference identification unit identifies information about the user's preference from information indicating a time-series correlation between the extracted listening sounds and the extracted emotional sounds. The terminal device further includes an environmental sound acquisition unit that acquires voice information as environmental sound information of the user,
The listening sound extraction unit extracts the listening sound from the acquired information of the environmental sound,
3. The matching system according to claim 1, wherein the emotional sound extraction unit extracts the emotional sound from the acquired environmental sound information. 4. The matching system according to claim 3, wherein the environmental sound acquisition unit acquires voice information as the environmental sound information in a vehicle in which the user rides. a listening sound extraction step of extracting information of a listening sound heard by a user from voice information;
an emotional sound extraction step of extracting emotional sound information reflecting the user's emotional state from voice information;
a preference identifying step of identifying information about the user's preference from the extracted listening sound information and the extracted emotional sound information;
a similarity determination step of determining the similarity of the preferences of the users based on the information of the preferences for each user that are specified for each of a plurality of users and accumulated in a database unit;
a user extraction step of extracting a combination of the users whose similarity is equal to or greater than a determination threshold;
a matching information providing step of providing, as matching information, information of other users in the extracted combination to a terminal device of one of the users in the combination extracted in the user extraction step;
Matching methods, including

Images