Information processing apparatus, method, and program

Abstract

To promote a conversation between a user and a conversation device by acquiring data about the user and area environment data via the conversation device.SOLUTION: An information processing device transmits, to a conversation content automatic generation unit, a set of first voice information of a conversation device and second voice information of a user, and a first question sentence for generating a first voice information candidate to be continued to the set; receives a first answer sentence generated by the conversation content automatic generation unit in response to the first question sentence; reads conversation log information associated with the user; and transmits, to the conversation content automatic generation unit, the set, the conversation log information, the first answer sentence, and the first answer sentence. And transmit a second question sentence for verifying content of the first answer sentence to the conversation content automatic generation unit, receive a second answer sentence generated by the conversation content automatic generation unit in response to the second question sentence, and transmit the second answer sentence to the conversation device as first voice information of the conversation device.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present invention relates to an information processing device, method, and program that acquires data about a user and area environment data via a conversation device, controls a conversation between the user and the conversation device, and manages the confidentiality and publicity of the data acquired through the conversation. [Background technology]

[0002] Japan is experiencing a decline in the labor force due to a declining birthrate, an aging population, and health issues. As a result, problems such as intensifying competition for recruitment due to a labor shortage, chronic long working hours, and increased corporate burdens for health insurance premiums have arisen, leading to companies turning their attention to "health and productivity management," which considers employee health management from a management perspective and puts it into strategic practice. Health and productivity management is expected to lead to improved performance and stock prices by investing in employee health based on the company's philosophy, which will revitalize the organization by increasing employee vitality and productivity. For companies, promoting employee health is seen as an "investment for the future," rather than a "cost."

[0003] In response to this social situation, Reference 1 introduces a system that aims to provide mental health care for each individual throughout the workplace and create an environment in which employees can work with peace of mind. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2023-094205 Summary of the Invention [Problem to be solved by the invention]

[0005] In recent years, companies have been promoting health and productivity management to address labor shortages. To achieve health and productivity management, it is necessary to understand the status of employees and the local environment, such as the workplace and nursing care facilities, analyze the results, and implement countermeasures. Traditionally, companies have considered employee health by conducting regular health checkups, questionnaires, and interviews and consultations with supervisors. However, these measures make it difficult to grasp employees' status in a timely manner, and implementing them places a significant burden on both supervisors and employees. Furthermore, interviews and questionnaires with supervisors often require employees to provide personal information, such as their physical condition, dissatisfaction, and causes of stress, making it difficult for companies to accurately grasp the information and status of their employees.

[0006] The present invention has been made to solve these problems, and aims to provide an information processing device, method, and program that acquires data about a user and area environment data via a conversation device such as an autonomously operable robot, a personal computer (PC), or a smartphone, promotes conversation between the user and the conversation device, and manages the confidentiality and openness of the data acquired through the conversation between the user and the conversation device. [Means for solving the problem]

[0007] An information processing device according to one aspect of the present invention is an information processing device including a communication unit for communicating with an external network, a storage unit, and a control unit, The storage unit stores a user master and a conversation log DB, The control unit Transmitting a set of first voice information associated with the conversation device and second voice information associated with the user, and a first question sentence for generating a first voice information candidate to be followed by the set, to an automatic conversation content generation unit; receiving a first answer sentence generated by the automatic conversation content generation unit in response to the first question sentence; reading conversation log information associated with the user from the conversation log DB; transmitting the set, the read conversation log information, the first answer sentence, and a second question sentence for verifying the content of the first answer sentence to an automatic conversation content generation unit; receiving a second answer sentence generated by the automatic conversation content generation unit in response to the second question sentence; transmitting the second reply sentence to the conversation device via the communication unit as first voice information associated with the conversation device; is configured to execute [Effects of the Invention]

[0008] According to the present invention, user data and area environmental data can be acquired via a conversation device such as an autonomously operable robot, personal computer (PC), or smartphone, enabling low-cost and timely data collection. Furthermore, according to the present invention, voice information for a user is verified based on the conversation between the user and the conversation device and past conversation log information, making it possible to provide the user with conversations based on more appropriate voice information. Furthermore, according to the present invention, the user converses with the conversation device rather than with a person such as a supervisor. The disclosure of acquired user data is controlled by the user, and a system for providing incentives to users is provided, lowering the barrier to users providing personal information. Furthermore, according to the present invention, user health management and workplace environment improvements can be performed in a timely manner, thereby improving business efficiency and corporate image through health management. [Brief explanation of the drawings]

[0009] A more detailed understanding of the embodiments disclosed herein can be had from the following description, taken in conjunction with the accompanying drawings, in which: [Figure 1] 1 is a diagram showing the configuration of an entire system including an information processing device 10 according to the present invention. [Figure 2] 1 is a system configuration diagram of an information processing device 10 according to the present invention. [Figure 3]3 is a diagram illustrating an example of the data structure of a user master 106. FIG. [Figure 4] FIG. 2 is a diagram illustrating an example of the data structure of a conversation log DB 107. [Figure 5] 10 is a flow diagram illustrating processing executed by information processing device 10 and conversation device 13 at the start of a conversation with a user. [Figure 6] 1 is a flow diagram illustrating a process of a conversation with a user executed by the information processing device 10 and the conversation device 13. FIG. [Figure 7] FIG. 10 is a flow diagram illustrating a process of registering a set of voice information of conversation device 13 and voice information of a user in association with a DID and a private key. [Figure 8] FIG. 10 is a flow diagram illustrating a process in which a user can control whether or not a conversation log stored in conversation log DB 107 is made public. [Figure 9] FIG. 9 is a diagram showing an example of a conversation log disclosure permission control screen 900. [Figure 10] FIG. 10 is a flow diagram illustrating a process in which a manager or an industrial physician at a workplace or the like refers to a conversation log of a user stored in the conversation log DB 107 and grants an incentive. [Figure 11] FIG. 11 is a diagram showing an example of a conversation log information display screen 1100. DETAILED DESCRIPTION OF THE INVENTION

[0010] (Overall structure and functions of each component) Fig. 1 is a configuration diagram of an entire system including an information processing device 10 according to the present invention. The information processing device 10 is connected to a user terminal 11 and an administrator terminal 12 via wired or wireless communication so that they can communicate with each other. The information processing device 10 is connected to a conversation device 13 via wired or wireless communication so that they can communicate with each other. Although Fig. 1 shows only one user terminal 11, one administrator terminal 12, and one conversation device 13, there may be a plurality of these.

[0011] In this specification, "user" refers to a user who receives the mental health support service described in this specification, such as an employee belonging to a specific community such as a company, an individual in a nursing home, a single person, or a delivery person. Mental health support services can include services that monitor the user's health status and services that monitor single people. "Company" refers not only to for-profit companies but also to any entity such as a local government or a non-profit organization. "Administrator" refers to a person who supports a user who receives the mental health support service described in this specification, such as a manager or industrial physician at a company, nursing home staff, welfare provider, or operations manager.

[0012] The information processing device 10 can acquire from the conversation device 13 environmental information of a specific area where the conversation device 13 is located or patrolled, such as information on temperature, humidity, illuminance, vibration, wind, etc. acquired by various sensors. The information processing device 10 can calculate the comfort level of the environment of an area, such as a workplace, based on the information acquired by the various sensors. The comfort level may be, for example, the "discomfort index" established by the Japan Weather Association, or a value compared to the standard values ​​for office illuminance, noise, and wind established by the Ministry of Health, Labor and Welfare.

[0013] The information processing device 10 acquires the user ID or facial image of the user with whom the conversation device 13 is attempting to have a conversation, performs an authentication process, determines a conversation scenario based on at least one of the authenticated user's personal information, the user's conversation log information, and environmental information of a specific area, and transmits audio information based on the conversation scenario to the conversation device 13.

[0014] For example, if the conversation scenario is based on the user's personal information, it may be a conversation scenario that asks about the user's hobbies and preferences, if it is based on the user's conversation log, it may be a conversation scenario related to the content discussed in previous conversations, and if it is based on environmental information for a specific area, it may be a conversation scenario related to general matters such as the weather and temperature. The conversation scenario is information determined to confirm the user's status, such as the user's mental health, and includes information that can serve as a conversation trigger (hint, clue) for the user. When information for confirming the user's status is obtained, a conversation that deviates from the conversation scenario may be conducted. For example, if the conversation begins with talking about the weather and the user then mentions the difficulties of work, the conversation may develop to confirm the details of the difficulties of work, without sticking to environmental topics such as the weather.

[0015] The information processing device 10 can promote a conversation between the conversation device 13 and the user by receiving voice information (and, optionally, a facial image) from the conversation device 13 and transmitting the voice information to the conversation device 13. The information processing device 10 can analyze the voice information (and, optionally, the facial image) received from the conversation device 13 and calculate a relevance score and an interest score of the conversation content with respect to the conversation scenario. The relevance score and the interest score may be expressed numerically, or may be expressed as a level such as high, medium, or low, or may be expressed as an intensity value of any value, and can be determined arbitrarily.

[0016] The facial image may be analyzed according to, for example, the six basic emotions (anger, disgust, fear, joy, sadness, and surprise) based on Ekman's theory. Based on the voice information and facial image received from the conversation device 13, the information processing device 10 analyzes the user's facial expression to determine what emotion the user is feeling, whether the tone of voice is high or low, whether the voice is loud or soft, and whether the speech is calm, gentle, harsh, or fast, thereby estimating the user's emotion and / or the reliability of the conversation content. The reliability of the conversation content can be determined based on the degree to which the tone, volume, and manner of speech differ from the tone, volume, and manner of speech in past conversations.

[0017] The information processing device 10 can identify changes in the user over time based on the voice information and / or facial images received from the conversation device 13. For example, the information processing device 10 can identify changes in the user's situation by reading the conversation log information of the user from the conversation log DB 107 and identifying whether the content of the user's utterances is different from previous ones, or whether the tone of the user's voice is higher or lower than a predetermined threshold.

[0018] The information processing device 10 can analyze the voice information received from the conversation device 13 to determine whether the content of the user's conversation is related to the conversation scenario and calculate a relevance score, i.e., a score indicating how much the user conversation contains content related to the conversation scenario. The information processing device 10 can also analyze the voice information received from the conversation device 13 to calculate an interest score for the user's conversation content, i.e., a score indicating how much the user conversation is interested in the conversation of the conversation device 13, based on how many positive words, negative words, and neutral words the user conversation contains. The information processing device 10 can add the calculated relevance score and interest score to the user's voice information and store it as log information.

[0019] The information processing device 10 can generate candidates for conversational voice information to be conveyed to the user next by transmitting the voice information uttered by the conversation device 13 and the user's voice information to which the relevance score and interest score have been added to the automatic conversation content generation unit. The automatic conversation content generation unit has a large-scale language model generated by machine learning, and has a function of generating an answer to the content of a question (prompt) in response to receiving the question. The information processing device 10 can generate a prompt based on a set of the voice information of the conversation uttered by the conversation device 13 and the voice information of the user's conversation, and a conversation scenario.

[0020] Machine learning may be performed by learning the results of a user taking a stress check test based on a 57-item questionnaire provided by the Ministry of Health, Labor and Welfare, or by learning the content of the conversation between the user and the conversation device described in this specification, or by using other learning data.

[0021] The information processing device 10 can acquire a conversation log of a user based on a user ID. The information processing device 10 can verify the content of candidate voice information for the next conversation to be conveyed to the user and generate a revised candidate by transmitting the conversation log and response sentences associated with the user to an automatic conversation content generation unit. The revised candidate becomes the voice information for the next conversation to be conveyed to the user. The information processing device 10 can generate a prompt based on a set of voice information emitted by the conversation device 13 and voice information of a voice emitted by the user, candidates for voice information for the next conversation to be conveyed to the user following the set, a conversation scenario, and a conversation log associated with the user.

[0022] The information processing device 10 can transmit the verified audio information of the next conversation to be conveyed to the user to the conversation device 13. By repeating the above-described process, the information processing device 10 can promote the conversation between the user and the conversation device 13.

[0023] The information processing device 10 can provide the content of a conversation between a user and a conversation device 13 to the user terminal 11. The user terminal 11 can select whether the conversation content is public, conditionally public, or private, and transmit the selected user control content to the information processing device 10. The information processing device 10 can receive the user control content for the conversation content and make the conversation content public to the administrator terminal 12 based on the user control content. Control of whether the conversation content is public or private can be set for each individual conversation, or can be set for each set of conversations (for example, five round-trip conversations between the user and the conversation device 13).

[0024] The information processing device 10 can perform a process of granting an incentive to a user who has published the content of a conversation based on a predetermined criterion. The incentive may be a deposit into the user's account, or may be a charge to a cashless means such as electronic money designated by the user.

[0025] The user terminal 11 and the administrator terminal 12 may be any type of device (e.g., a smartphone, a PC, a tablet terminal, etc.) that can operate in a wired or wireless environment, and are not limited to a specific apparatus or device. The user terminal 11 and the administrator terminal 12 can communicate with the information processing device 10 and send and receive various information. The user terminal 11 is a terminal used by a user in a specific area such as a company, a nursing home, or a private room. The administrator terminal 12 is a terminal used by a person who plays a role in supporting the mental health of users, such as a manager of a company, an industrial physician, or a nursing home staff member.

[0026] The user terminal 11 can determine whether a conversation log provided by the information processing device 10 should be made public, public with conditions, or private (user control content), and can transmit the determined information to the information processing device 10. The administrator terminal 12 can receive from the information processing device 10 information about the conversation log for which public / conditionally public / private content has been determined based on the user control content.

[0027] The conversation device 13 may be a device such as an autonomously operable robot, a PC, or a smartphone. The conversation device 13 is an autonomously operable robot equipped with various sensors and based on SLAM (Simultaneous Localization and Mapping) technology, and is capable of patrolling a specific area such as a corporate workplace. The conversation device 13 is a PC or a smartphone equipped with various sensors, and can be located near the user. The conversation device 13 can transmit sensor data regarding environmental information of the area, such as temperature, humidity, illuminance, vibration, and wind, acquired by the various sensors, to the information processing device 10.

[0028] The conversation device 13 may have a short-range wireless communication means, a communication means, a speaker, a microphone, a camera, and a display.

[0029] The conversation device 13 can obtain a user ID from the employee ID presented by the employee via a short-range wireless communication means that functions as an employee ID reading means, and transmit the user ID to the information processing device 10 for user authentication processing.

[0030] Conversation device 13 communicates with information processing device 10 via wired or wireless communication means and can send and receive various information. When conversation device 13 is patrolling a specific area or is located near a user, if the user speaks to conversation device 13 via a microphone or the like, conversation device 13 can acquire a facial image of the user using a camera and transmit the acquired facial image to information processing device 10 for user authentication processing.

[0031] The conversation device 13 can acquire information such as the name, personal information, and voice information based on a conversation scenario of the user who has been authenticated by the information processing device 10, and start a conversation with the employee.

[0032] Conversation device 13 can converse with a user by recognizing the user's voice received via a microphone, generating voice information based on the recognized voice, transmitting the generated voice information to information processing device 10, and emitting from a speaker a voice synthesized based on the voice information received from information processing device 10. Conversation device 13 can optionally display the voice information on a display, allowing the user to know what they have said and the content of the voice emitted by conversation device 13 as text information. Conversation device 13 can optionally transmit conversation information input by the user to the display to information processing device 10. In this specification, the terms voice information and conversation information can be used interchangeably.

[0033] The conversation device 13 may be equipped with a noise reduction function to recognize the user's voice, a double microphone to detect the direction to identify the location where the voice is coming from, and a lidar to measure the distance to the user.

[0034] The conversation device 13 can transmit a facial image of the user acquired by the camera to the information processing device 10 for processing to determine the user's mood, which is performed by the information processing device 10. The video information can include not only a facial image but also images of other parts of the body or the entire body.

[0035] The order in which conversation device 13 speaks to users may be determined arbitrarily. For example, conversation device 13 may move to the position of a conversation target based on conversation target information and seat position information of the target received from information processing device 10 and start a conversation, or may start a conversation with people encountered in the area being patrolled in order, or may start a conversation with a user when it is determined that the user has spoken to it, and the selection of a conversation target is not particularly limited.

[0036] (System Configuration) 2 is a system configuration diagram of an information processing device 10 according to the present invention. Similar to a general computer, the information processing device 10 includes a control unit 101, a main memory unit 102, an auxiliary memory unit 103, an IF unit 104, and an output unit 105, which are interconnected by a bus 120 or the like. The information processing device 10 includes a user master 106 and a conversation log DB 107 in the form of storage means such as a file / database in the auxiliary memory unit 103. The file / database stored in the auxiliary memory unit 103 may be configured to be managed based on a distributed ledger technology such as blockchain, and may function as a Verifiable Data Registry (a public DID storage location for issuers) in DID (Decentralized Identifier) / VC (Verifiable Credentials) technology, for example.

[0037] The control unit 101, also called a central processing unit (CPU), controls each component of the information processing device 10 and performs data calculations, and also reads various programs stored in the auxiliary storage unit 103 into the main storage unit 102 and executes them. The main storage unit 102, also called a main memory, can store various received data, computer-executable instructions, and data after calculation processing based on those instructions. The auxiliary storage unit 103 is a storage device such as a hard disk drive (HDD) or solid-state drive (SSD), and is used for long-term storage of data and programs.

[0038] 2 illustrates an embodiment in which the control unit 101, main memory unit 102, and auxiliary memory unit 103 are provided in the same computer, but in another embodiment, the information processing device 10 can be configured to realize parallel distributed processing by a plurality of computers by using a plurality of control units 101, main memory units 102, and auxiliary memory units 103. In another embodiment, it is also possible to provide an embodiment in which a plurality of servers are installed for the information processing device 10, and the plurality of servers share a single auxiliary memory unit 103.

[0039] The IF unit 104 serves as an interface (IF) when sending and receiving data to and from other systems and devices, and also provides an interface for accepting various commands and input data (various masters, tables, etc.) from a system operator. The output unit 105 provides a display screen for displaying processed data and printing means for printing the data.

[0040] Components with functions similar to those of the control unit 101, main memory unit 102, auxiliary memory unit 103, IF unit 104, and output unit 105 also exist in the user terminal 11 and the administrator terminal 12, but descriptions of these will be omitted in this specification.

[0041] The user master 106 stores information about users who receive the mental health support services described herein. FIG. 3 is a diagram illustrating an example of the data structure of the user master 106. The user master 106 may include a user ID 301, user information 302, and a user image 303, but is not limited to these data items and may also include other data items. For example, the user master 106 may be configured to include an identifier (e.g., a company ID) that identifies a community in order to distinguish between employees of different companies. The user master 106 may also be managed by using separate storage means such as files or databases for each community.

[0042] User ID 301 is an identifier for identifying a user. User information 302 indicates information about the user, such as the user's name, affiliation, gender, date of birth, hobbies, and account to which incentives are awarded for disclosing conversation content. User image 303 includes at least a facial image of the user and may also include a full-body image.

[0043] Returning to Fig. 2, the conversation log DB 107 stores the contents of the conversation between the user and the conversation device 13. Fig. 4 is a diagram illustrating an example of the data structure of the conversation log DB 107. The conversation log DB 107 can include a user ID 301, a conversation start time 401, a conversation end time 402, conversation log information 403, a distributed ID 404, key information 405, a disclosure flag 406, an anonymous 407, a disclosure target user 408, and a VP disclosure count 409, but is not limited to these data items and can also include other data items.

[0044] As described above, the user ID 301 is an identifier that identifies a user. The conversation start time 401 indicates the time when the user and the conversation device 13 started a conversation. The conversation end time 402 indicates the time when the user and the conversation device 13 ended a conversation. The conversation log information 403 indicates a set of one or more conversation contents between the user and the conversation device 13, as well as a relevance score and an interest score calculated for the conversation contents.

[0045] The distributed ID 404 indicates the distributed ID (DID) assigned to the conversation log. The key information 405 indicates the key (private key, public key) that controls encryption and decryption of the conversation log. The disclosure flag 406 is a flag indicating whether the contents of the conversation log selected by the user are to be made public. The pseudonym 407 indicates a name used when a user wishes to disclose information anonymously rather than using their real name. The disclosure target users 408 includes information specifying users who can access information that has been made public. In other words, users with user IDs registered in the disclosure target users 408 can access the information that has been made public, but users with user IDs that are not registered cannot access the information that has been made public. The VP disclosure count 409 indicates the number of times the conversation log to which the distributed ID has been assigned has been made public to a verifier, such as a workplace supervisor or an industrial physician.

[0046] (Processing flow at the start of conversation between the information processing device 10 and the conversation device 13) 5 is a flow diagram illustrating the processing at the start of a conversation with a user, which is executed by the information processing device 10 and the conversation device 13. In this processing flow, the conversation device 13 is described as an autonomously operable robot, but any operation of this processing flow can be executed even if the conversation device 13 is a PC or a smartphone.

[0047] In S501, the information processing device 10 transmits a user search request and user location information (e.g., seat location) to the conversation device 13. The information processing device 10 can also transmit conversation target information generated by querying the user master 106 and the conversation log DB 107 to the conversation device 13. The conversation target may be, for example, a person for whom a conversation log has not been accumulated for a predetermined period (e.g., no conversation log has been accumulated for the past month). Upon receiving the user search request, the conversation device 13 patrols an area such as a workplace based on the user's location information. While patrolling the area, the conversation device 13 acquires environmental information using various sensors and transmits the acquired environmental information to the information processing device 10. The information processing device 10 stores the received environmental information.

[0048] In S502, conversation device 13 acquires facial image data of the conversation target user using a camera, and transmits the acquired facial image data to information processing device 10. Information processing device 10 can authenticate the user by querying user master 106 based on the received facial image data.

[0049] In another embodiment, the conversation device 13 may transmit employee identification information acquired from the employee ID reader to the information processing device 10. The employee identification information may be the same as the user ID. The information processing device 10 can authenticate the user by querying the user master 106 based on the received employee identification information.

[0050] In S503, the information processing device 10 can query the conversation log DB 107 based on the user ID of the authenticated user and acquire the conversation log information of the user. The information processing device 10 can determine a conversation scenario based on at least one of the authenticated user's personal information, the user's conversation log information, and the area's environmental information. The user's personal information may include information obtained by analyzing the user's voice information and / or face image received from the conversation device 13, in addition to the information indicated in the user information 302.

[0051] In S504, the information processing device 10 can generate greeting information (voice information) for starting a conversation with the user based on the conversation scenario. The greeting information may be a greeting prepared in advance for a specific conversation scenario (e.g., a greeting related to the weather or temperature), may be generated based on information such as hobbies stored in the user information 302, or may be generated by analyzing the contents of the user's past conversation log (e.g., "Hello, Mr. / Ms. X. Did you have fun at XY the last time you visited?").

[0052] The information processing device 10 can transmit greeting information to the conversation device 13 together with the user ID 301 and user information 302 of the authenticated user. The conversation device 13 can emit a synthesized voice from a speaker based on the received greeting information. The conversation device 13 can also display the received greeting information on a display. A user who is greeted by the conversation device 13 returns some kind of response. This response may be words or an attitude that actively attempts to start a conversation, or words or an attitude that rejects the conversation, and the content of the response will be analyzed as described below.

[0053] In S505, the conversation device 13 converts the user's response voice into voice information and transmits the voice information to the information processing device 10. The conversation device 13 can also acquire the user's attitude, for example, a facial image of the user's expression, and transmit the acquired information to the information processing device 10. The information processing device 10 can analyze the received voice information (and, optionally, the facial image) and determine whether the user intends to continue the conversation. This determination may be made based on whether the voice information contains positive language or negative language (e.g., "I'm busy now"). If it is determined that the user intends to continue the conversation, the process proceeds to the processing of FIG. 6. If it is determined that the user does not intend to continue the conversation, the information processing device 10 transmits voice information of a standard greeting to the user (e.g., "I understand. See you, Mr. / Ms. X"), causing the conversation device 13 to speak the voice, and then returns to S502.

[0054] (Processing flow at the start of conversation between the information processing device 10 and the conversation device 13) FIG. 6 is a flow diagram illustrating the processing of a conversation with a user executed by the information processing device 10 and the conversation device 13. In this embodiment, a conversation scenario between the user and the conversation device 13 will be described as an example related to the user's health. Therefore, the predetermined category that serves as the basis for calculating the relevance score described below may be something that can affect health, such as "work, workplace," "physical condition, mental health," or "human relationships." Furthermore, keywords belonging to a predetermined category (e.g., "work, workplace") may be, for example, "busy at work," "deadline," or "overtime," while keywords belonging to another category (e.g., "physical condition, mental health") may be, for example, "stress," "tired," or "anxiety." The categories and keywords may be based on data items defined in the above-mentioned stress check test or the like.

[0055] In S601, the information processing device 10 analyzes the voice information (and optionally the facial image) received from the conversation device 13 and determines whether the user intends to continue the conversation. This determination may be made based on whether the voice information contains positive or negative words. If it is determined that the conversation should be continued, the process proceeds to S602, and if it is determined that the conversation should not be continued, this processing flow ends. When this processing flow ends, the information processing device 10 can transmit voice information of a standard greeting to the user and cause the conversation device 13 to speak the voice.

[0056] In S602, the information processing device 10 analyzes the voice information received from the conversation device 13, extracts keywords belonging to a predetermined category, and calculates the number of extracted keywords to calculate a relevance score. Table 1 shows examples of categories and keywords related to the user's health.

[0057] [Table 1]

[0058] If the conversation between the user and the conversation device 13 is as follows: Conversational Device: "How was your week?" User: "I had a deadline yesterday and it was busy. I worked overtime every day. I was stressed out because I was worried I wouldn't be able to meet the deadline. My boss put a lot of pressure on me, and it was tough." The information processing device 10 analyzes the user's voice information and extracts four keywords, "deadline," "busy work," "overtime," and "delivery date," from the "work, workplace" category; two keywords, "anxiety" and "stress," from the "physical condition, mental health" category; one keyword, "boss," from the "human relationships" category; and one keyword, "it was difficult," from the "negative words" category.

[0059] The information processing device 10 can calculate the number of extracted keywords and calculate a relevance score based on the calculated number of keywords. Table 2 shows an example of a criterion for calculating the relevance score.

[0060] [Table 2]

[0061] In the above example, the information processing device 10 can determine the relevance score as 5 points because there are eight keywords (=4+2+1+1). In this example, each keyword is weighted equally, but the keywords may be weighted and the number of keywords for specific keywords may be doubled or tripled in the calculation. The information processing device 10 may be configured to learn the relevance scores associated with the accumulated conversation logs and calculate weights for categories and keywords.

[0062] The information processing device 10 can analyze the voice information (and, optionally, the facial image) received from the conversation device 13 and calculate the user's interest score for the conversation content. When using the facial image, the information processing device 10 can determine which of the six basic emotions (anger, disgust, fear, joy, sadness, and surprise) the user's facial image corresponds to. The interest score is used to determine whether to continue the conversation with the user according to the current conversation scenario or switch to another conversation scenario.

[0063] If the conversation between the user and the conversation device 13 is as follows: Conversational Device: "How was your week?" User: "I had a deadline yesterday and it was busy. I worked overtime every day. I was stressed out because I was worried I wouldn't be able to meet the deadline. My boss put a lot of pressure on me, and it was tough." The information processing device 10 analyzes the user's voice information and determines whether the user is interested in the conversation scenario based on a predetermined threshold. In this example, the user's response to the conversation device conveys the details of the work and the user's feelings about it, so it can be determined that the user is interested. The predetermined threshold may be divided into two values, "interested" and "not interested," or into three or more values. For example, it may be five values, such as "very interested," "somewhat interested," "neither interested nor not interested," "not very interested," and "not at all interested."

[0064] As another example, if the conversation between the user and the conversation device 13 was as follows: Conversational Device: "Did you have any fun today?" User: "Um, nothing special. Nothing special." The information processing device 10 analyzes the user's voice information and determines whether the user is interested in the conversation scenario based on a predetermined threshold. In this example, the user's response can be determined to indicate a lack of interest in the conversation content based on the predetermined threshold.

[0065] The information processing device 10 adds the calculated relevance score and interest score to a set of speech information of speech uttered by the conversation device and speech information of speech uttered by the user, and stores the result in the conversation log DB 107.

[0066] In S603, the information processing device 10 generates a question sentence (prompt) to be sent to an automatic conversation content generation unit to generate candidates for the conversation device's voice information following the set of voice information of the voice uttered by the conversation device and voice information of the voice uttered by the user. That is, the information processing device 10 generates candidates for the conversation content of the conversation device 13 in response to the conversation of the conversation device 13 and the user's conversation in response to the conversation. The automatic conversation content generation unit has a large-scale language model generated by machine learning and has a function of generating an answer sentence to the content of the question sentence in response to receiving the question sentence (prompt). The automatic conversation content generation unit may be present inside the information processing device 10 or may be present in an external system, and is not particularly limited.

[0067] The content of the prompt is generated so as to be related to the conversation scenario, and may be, for example, the following content: "As an employee health manager or industrial physician, please generate the conversation content for the conversation device that should continue based on the set of stored voice information." The information processing device 10 transmits a set of audio information of the voice uttered by the conversation device 13 and audio information of the voice uttered by the user, as well as a prompt for generating candidates for the voice information of the conversation device, to the automatic conversation content generation unit, and receives a response sentence generated by the automatic conversation content generation unit in response to the prompt.

[0068] In S604, information processing device 10 queries conversation log DB 107 based on the user ID and reads out the conversation log associated with the user.

[0069] In S605, the information processing device 10 generates a question (prompt) for verifying the content of the answer sentence based on a set of audio information of the voice uttered by the conversation device 13 and audio information of the voice uttered by the user, the read conversation log, the conversation scenario, and the answer sentence, and transmits the generated question sentence together with the set of audio information, the answer sentence, and the conversation log to the automatic conversation content generation unit. The content of the prompt is generated so as to be content related to the conversation scenario, and may be, for example, the following content: "As a manager or industrial physician responsible for managing the health of employees, please consider the flow of conversation in the read conversation log and determine whether the content of the reply sentences is appropriate as the conversation content for the conversation device that should continue with the stored set of voice information. If it is not appropriate, please correct that part and generate the conversation content for the conversation device." The information processing device 10 receives the verified response sentence generated by the automatic conversation content generation unit in response to the prompt.

[0070] In one embodiment of the present invention, if the interest score is lower than a certain reference value, for example, in the above example, if the interest is "none," "not very much," or "none at all," the information processing device 10 may be configured to modify the conversation scenario, such as by specifying the selection of a different category in the prompt for generating the answer sentence.

[0071] In S606, information processing device 10 transmits the verified answer sentence as voice information of the conversation device to conversation device 13. Conversation device 13 synthesizes voice based on the received voice information and emits the voice through a speaker.

[0072] In S607, the conversation device 13 receives the voice of the user's conversation and transmits voice information generated based on the received voice (and optionally a facial image of the user acquired by a camera) to the information processing device 10. The information processing device 10 receives the user's voice information (and optionally a facial image) from the conversation device 13. Thereafter, the process returns to S601, and the information processing device 10 determines whether the user intends to continue the conversation. If it is determined that the conversation should be continued, the process proceeds to S602, and if it is determined that the conversation should not be continued, this processing flow ends. By repeating the process of FIG. 6, the conversation between the user and the conversation device 13 continues.

[0073] (DID issuance process when storing conversation logs) The process of storing the set of voice information of conversation device 13 and the user's voice information in the conversation log in S602 may be performed based on DID (Decentralized Identifier) / VC (Verifiable Credentials) technology. In such a case, the set of voice information of conversation device 13 and the user's voice information is treated as one piece of data in conversation log DB 107, and for example, if conversation device 13 and the user have five conversations (five round trips), five pieces of data are generated.

[0074] FIG. 7 is a flow diagram illustrating the process of registering a set of voice information of conversation device 13 and voice information of a user in association with a DID and a private key.

[0075] In S701, the information processing device 10 issues a DID to the user ID in response to the addition of a relevance score and an interest score to a set of the voice information of the conversation device 13 and the voice information of the user.

[0076] In S702, the information processing device 10 encrypts the contents of the conversation log information 403 with a private key associated with the user ID, associates the conversation log with the issued DID, and key information used for encryption, and stores them in the conversation log DB 107. The key information may include the private key used for encryption and a public key to be used when referencing the conversation log.

[0077] (User selection process for whether or not to make conversation logs public) FIG. 8 is a flow diagram illustrating a process in which a user can control whether or not a conversation log stored in the conversation log DB 107 is made public.

[0078] In S801, the information processing device 10 provides a conversation log publication permission control screen 900 in response to a request from the user terminal 11. Fig. 9 is a diagram showing an example of the conversation log publication permission control screen 900. The conversation log publication permission control screen 900 may include a calendar selection button 901, a conversation summary button 902, and a registration button 903.

[0079] The conversation log disclosure control screen 900 can display summary data of the conversation log for the year and month selected by the user, and can also display data of each conversation log of the conversation summary selected by the user in a log data display window 904. These display controls are executed by the information processing device 10.

[0080] Users can select conversation summaries that they are willing to disclose to managers such as their superiors at work or verifiers such as industrial physicians using check boxes on the screen, and can further select which individual conversations within the conversation summary they are willing to disclose. Users can choose to disclose information anonymously rather than using their real name, and can enter the pseudonym on the screen. Users can specify specific users to disclose information to; for example, they can choose not to disclose information to managers such as their superiors at work, but to disclose information to industrial physicians.

[0081] In S802, the information processing device 10 can register the above information inputted through the conversation log disclosure permission control screen 900 in the conversation log DB 107.

[0082] (Administrators can view conversation logs) FIG. 10 is a flow diagram illustrating a process in which a manager or an industrial physician in a workplace or the like refers to a conversation log of a user stored in the conversation log DB 107 and grants an incentive.

[0083] In S1001, the information processing device 10 provides a conversation log information display screen 1100 in response to a request from the administrator terminal 12. Fig. 11 is a diagram showing an example of the conversation log information display screen 1100. The conversation log information display screen 1100 includes a conversation party information display field 1101 and a statistical data display field 1102, and may further include a user data display window 1105 that is displayed when a specific user 1103 is selected in the conversation party information display field 1101, and a log data display window 1106 that is displayed when conversation log information 1104 is selected.

[0084] The conversation participant information display field 1101 can display information about users with whom the conversation device 13 conversed on any given day. The displayed user information may be user 1103 (user ID, name) or anonymous. The statistical data display field 1102 can display the number of users with whom the conversation device 13 conversed on any given day and the number of conversation log data items. The user data display window 1105 displays the conversation log or vital data of a selected user on that day. In one embodiment of the present invention, the conversation device 13 can acquire the user's body temperature and other vital data using a sensor, and can store the sensor data as a type of conversation log. The log data display window 1106 can display the content of a conversation between the conversation device 13 and the user, data that the user has permitted to be made public.

[0085] In S1002, the information processing device 10 increments the value of the VP publication number 409 for the conversation log data referenced via the conversation log information display screen 1100. The conversation log information display screen 1100 displays the conversation log information of the user to be made public.

[0086] The information processing device 10 can add up the values ​​of the VP disclosure numbers 409 of each user at a predetermined timing and grant incentives to the users based on the added value. For example, the information processing device 10 may be configured to read out the account information of the user information 302 and execute a process of depositing an amount based on the incentive value.

[0087] Although the principles of the present invention have been described above with reference to exemplary embodiments, it will be understood by those skilled in the art that various modifications in configuration and detail can be made without departing from the spirit of the present invention. That is, the present invention can be embodied as, for example, a system, an apparatus, a method, a program, or a storage medium. [Explanation of symbols]

[0088] 10. Information processing equipment 11 User terminal 12 Administrator terminal 13 Conversation Device 101 Control section 102 Main memory 103 Auxiliary storage 104 IF Section 105 Output section 106 User Master 107 Conversation Log DB

Claims

1. An information processing device including a communication unit for communicating with an external network, a storage unit, and a control unit, The storage unit stores a user master and a conversation log DB, The control unit Transmitting a set of first voice information associated with the conversation device and second voice information associated with the user, and a first question sentence for generating a first voice information candidate to be added to the set, to an automatic conversation content generation unit; receiving a first answer sentence generated by the automatic conversation content generation unit in response to the first question sentence; reading conversation log information associated with the user from the conversation log DB; transmitting the set, the read conversation log information, the first answer sentence, and a second question sentence for verifying the content of the first answer sentence to an automatic conversation content generation unit; receiving a second answer sentence generated by the automatic conversation content generation unit in response to the second question sentence; transmitting the second reply sentence to the conversation device via the communication unit as first voice information associated with the conversation device; An information processing device configured to execute the above.

2. The control unit receiving environmental information from the conversation device via the communication unit; reading out conversation log information of the authenticated user from the conversation log DB; determining a conversation scenario based on at least one of personal information of the authenticated user, the read conversation log information, and the environmental information; generating the first voice information for starting a conversation with the user based on the conversation scenario, and transmitting the generated first voice information to the conversation device; and further configured to execute the first question sentence and the second question sentence are generated so as to be related to the conversation scenario. The information processing device according to claim 1 .

3. The control unit analyzing the second voice information received from the conversation device via the communication unit, extracting keywords associated with a predetermined category, and calculating a relevance score based on the number of the extracted keywords; analyzing the second voice information received from the conversation device via the communication unit and calculating an interest score of the user with respect to the conversation content; storing the received second voice information, the calculated relevance score, and the calculated interest score in the conversation log DB as the conversation log information; The information processing apparatus of claim 1 , further configured to perform the following:

4. The control unit In response to the received second voice information being associated with the calculated relevance score and the interest score and being stored in the conversation log DB as the conversation log information, issuing a distributed ID associated with the user; encrypting the stored conversation log information, and storing the encrypted conversation log information in the conversation log DB in association with the distributed ID and key information used for encryption; The information processing apparatus of claim 3 , further configured to execute:

5. The information processing device according to claim 3 , wherein the analysis of the second voice information is performed based on at least one of tone, volume, and manner of speaking associated with the second voice information, and a facial image of the user received from the conversation device.

6. The control unit providing the conversation log information associated with the user read from the conversation log DB to a user terminal; receiving publicly-available related information selected by the user for each of the provided conversation log information, the publicly-available related information including information on whether each of the conversation log information is subject to disclosure and information on users subject to disclosure; The information processing apparatus of claim 1 , further configured to perform the following:

7. 7. The information processing device according to claim 6, wherein the control unit is further configured to increment a value of the number of publications associated with the user in response to the conversation log information to be made public being referenced by a user other than the user.

8. The information processing device according to claim 1 , wherein the conversation device is an autonomously operable robot.

9. A method executed by an information processing device including a communication unit for communicating with an external network, a storage unit, and a control unit, The storage unit stores a user master and a conversation log DB, the control unit transmits to an automatic conversation content generation unit a set of first voice information associated with the conversation device and second voice information associated with the user, and a first question sentence for generating a first voice information candidate to be added to the set; the control unit receives a first answer sentence generated by the automatic conversation content generation unit in response to the first question sentence; the control unit reads conversation log information associated with the user from the conversation log DB; the control unit transmits the set, the read conversation log information, the first answer sentence, and a second question sentence for verifying the content of the first answer sentence to an automatic conversation content generation unit; the control unit receives a second answer sentence generated by the automatic conversation content generation unit in response to the second question sentence; the control unit transmits the second reply sentence to the conversation device via the communication unit as first voice information associated with the conversation device; A method for providing the above.

10. A program that causes a computer to execute the method according to claim 9.

Images