Information processing device and information processing method

Abstract

The present invention makes it possible for a virtual agent to talk to a user with content tailored to the user, using a greeting and a topic suited to the situation of the user.　An information processing device 20 uses a generative artificial intelligence model 30 to generate conversation content of a virtual agent VA for a user. The information processing device comprises: a prompt generation unit 223 that includes user information of the user in a prompt for causing the generative artificial intelligence model to generate the conversation content for the virtual agent to talk to the user; a transmission unit 224 that transmits the prompt to the generative artificial intelligence model; a reception unit 225 that receives the conversation content, as a response to the prompt, generated by the generative artificial intelligence model; and an output unit 226 that outputs the received conversation content as the response to the prompt.

Description

Information Processing Apparatus and Information Processing Method Cross-Reference to Related Applications

[0001] This application claims the benefit of Japanese Patent Application No. 2025-066444 filed in Japan on April 14, 2025, the content of which is hereby incorporated by reference in its entirety.

[0002] The present invention relates to an information processing apparatus and an information processing method.

[0003] Conventionally, there are applications that allow users to converse with AI agents (virtual agents) (for example, Japanese Unexamined Patent Application Publication No. 2024-028844).

[0004] In such applications, when the application is launched, the AI agent cannot determine the current situation of the user, so the AI agent may speak to the user with remarks that do not suit the user's situation, or the content of the speech may become monotonous and uniform.

[0005] In view of the above background, the present invention aims to provide an information processing apparatus and the like that enables a virtual agent to speak to the user with content that is considerate of the user, using greetings and topics suited to the user's situation.

[0006] The information processing apparatus of the present invention is an information processing apparatus that generates conversation content of a virtual agent for a user using a generative artificial intelligence model, and comprises: a prompt generation unit that includes user information of the user in a prompt for causing the generative artificial intelligence model to generate the conversation content for the virtual agent to speak to the user; a transmission unit that transmits the prompt to the generative artificial intelligence model; a reception unit that receives the conversation content generated by the generative artificial intelligence model as a response to the prompt; and an output unit that outputs the conversation content received by the reception unit as the response to the prompt.

[0007] The present invention relates to an information processing method for generating conversation content of a virtual agent to a user using a generative artificial intelligence model, comprising the steps of: including the user information of the user in a prompt for the generative artificial intelligence model to generate the conversation content for the virtual agent to speak to the user; transmitting the prompt to the generative artificial intelligence model; receiving the conversation content generated by the generative artificial intelligence model as a response to the prompt; and outputting the conversation content as a response to the received prompt.

[0008] According to the present invention, a virtual agent can engage in conversation with the user in a way that is tailored to the user's situation, using appropriate greetings and topics.

[0009] Figure 1 is a diagram showing an example of the system configuration of the information processing system S according to this embodiment. Figure 2 is a diagram showing an example of information processing realized by the information processing system S according to this embodiment. Figure 3 is a diagram showing an example of the hardware configuration of the user terminal 10 according to this embodiment. Figure 4 is a diagram showing an example of the hardware configuration of the distribution server 20 according to this embodiment. Figure 5 is a diagram showing the functional configuration of the distribution server 20. Figure 6 is a diagram showing an example of user information stored in the user information storage unit 211 according to this embodiment. Figure 7 is a diagram showing an example of information stored in the status command storage unit 212 according to this embodiment. Figure 8 is a diagram showing an example of conversation history information stored in the conversation history information storage unit 213 according to this embodiment. Figure 9 is a diagram showing the functional configuration of the user terminal 10. Figure 10 is a flowchart showing an example of processing executed in the information processing system S according to this embodiment. Figure 11 is a flowchart showing an example of processing executed by the distribution server 20 in step S103 of Figure 10, which generates a prompt for the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user.

[0010] The information processing system according to this embodiment will be described below with reference to the drawings. Note that the following description is merely an example of a preferred embodiment and is not intended to limit the invention as described in the claims. In all the figures illustrating this embodiment, common components are denoted by the same reference numerals, and repeated explanations are omitted.

[0011] [Overview of Information Processing System S] Figure 1 is a diagram showing an example of the system configuration of the information processing system S according to this embodiment. As shown in Figure 1, the information processing system S according to this embodiment is an information processing system that uses an artificial intelligence system (generative artificial intelligence model) 30 to generate conversation content of a virtual agent (AI agent) to a user.

[0012] The information processing system S according to this embodiment comprises a user terminal 10, a distribution server 20, and an artificial intelligence system 30. The information processing system S may also include other servers, terminals, and other equipment. The user terminal 10 and the distribution server 20 are connected to each other via a network NW. The distribution server 20 and the artificial intelligence system 30 are also connected to each other via a network NW. The network NW consists of the Internet, LAN (Local Area Network), various mobile communication systems constructed by wireless base stations, etc. For example, the network includes 3G, 4G, 5G mobile communication systems, LTE (Long Term Evolution), and wireless networks that can connect to the Internet via predetermined access points (e.g., Wi-Fi®). When connecting wirelessly, communication protocols include, for example, Z-Wave®, ZigBee®, Bluetooth®, etc. When connecting via a wired connection, the network also includes those directly connected by USB (Universal Serial Bus) cables, etc.

[0013] The user terminal 10 is an information processing device used by the user. The user terminal 10 accepts operations from the user. The user terminal 10 launches an application according to this embodiment (hereinafter sometimes simply referred to as "app") based on the user's operation. The app according to this embodiment is a virtual agent app that provides a conversation service with a virtual agent. The user terminal 10 requests content to be displayed or played by the app from the distribution server 20 based on the user's operation. The user terminal 10 may display the content using any app, as long as it is capable of displaying or playing the content distributed by the distribution server 20. For example, the user terminal 10 may display or play the content using a dedicated application such as the virtual agent app according to this embodiment, or it may display or play the content using a browser app (hereinafter sometimes simply referred to as "browser").

[0014] The user terminal 10 can be implemented as, for example, a smartphone, tablet, notebook PC (Personal Computer), desktop PC, mobile phone, PDA (Personal Digital Assistant), etc. In this embodiment, the user terminal 10 is shown as a smartphone with touch panel functionality.

[0015] In this embodiment, the user terminal 10 has the function of a voice response system (speech recognition), detects voice information (hereinafter also referred to as "utterances") spoken by the user, and accepts the detected utterances as input information. For example, the user terminal 10 is a terminal device that processes the user's utterances. When the user terminal 10 detects an utterance by the user, it sends the input information based on the user's utterance to the distribution server 20, receives content corresponding to the user's utterance, and presents it to the user.

[0016] The distribution server 20 receives a distribution request from the user's user terminal 10 and distributes content related to the virtual agent that interacts with the user in the virtual agent application launched on the user terminal 10. The distribution server 20 uses an artificial intelligence system 30 to generate the conversation content of the virtual agent to the user and distributes the generated conversation content to the user terminal 10. The distribution server 20 is composed of at least a computer having communication and program execution functions. The distribution server 20 may be implemented with one information processing device or with multiple information processing devices.

[0017] Furthermore, the distribution server 20 may have a speech recognition function. Also, the distribution server 20 may be able to obtain information from a speech recognition server that provides speech recognition services. In this case, the information processing system S may include a speech recognition server. In this embodiment, it is assumed that the user terminal 10, distribution server 20, and speech recognition server recognize user speech and identify the user who made the speech using various conventional technologies as appropriate, and therefore the explanation is omitted.

[0018] The artificial intelligence system 30 is an information processing device that internally incorporates an LLM (Large Language Model) and is capable of outputting output sentences (responses to prompts) in response to input sentences called prompts. For example, the artificial intelligence system 30 includes ChatGPT, OpenAI GPT, PerplexityAsk, BingAI, BERT, etc. The artificial intelligence system 30 may be operated by a business that provides the virtual agent application according to this embodiment, or it may be used via an API (Application Programming Interface) provided by an external artificial intelligence system 30 operated by another business.

[0019] The artificial intelligence system 30 is equipped with a dialogue (chat) function, and by providing the artificial intelligence system 30 with any inquiry or command / instruction in text, it is possible to obtain an answer to the inquiry or command / instruction. In this embodiment, by sending a prompt to the artificial intelligence system 30 to generate the content of the conversation between the virtual agent and the user, or the content of the conversation between the virtual agent and the user, the content of the conversation can be obtained as an answer. The content of the conversation between the virtual agent and the user is an example of the content of the conversation between the virtual agent and the user, and is the first thing the virtual agent says to the user.

[0020] Furthermore, in this embodiment, the artificial intelligence system 30 is not limited to text-based dialogue. For example, the artificial intelligence system 30 may be capable of voice-based dialogue. In this case, the artificial intelligence system 30 may receive voice data of the user's utterance (for example, the voice data of the utterance itself, or the voice data of the utterance with noise, etc., canceled out) along with the user's identification information, and generate the conversation content of the virtual agent. Alternatively, the artificial intelligence system 30 may be an image generation AI system such as Midjourney or Stable Diffusion that can generate image data.

[0021] Hereinafter, an example of information processing realized by the information processing system S according to this embodiment will be described using Figure 2. Figure 2 is a diagram showing an example of information processing realized by the information processing system S according to this embodiment. Figure 2 shows the case where the user is identified by user ID "U1" (hereinafter sometimes referred to as "user U1"). In Figure 1, user U1 is assumed to be a user using the conversation service with the virtual agent VA provided by the distribution server 20.

[0022] In step S1, user U1 operates user terminal 10 to launch the virtual agent application and sends a request to distribution server 20 for the distribution of conversation content (i.e., the content of what the virtual agent VA says to user U1) for the virtual agent VA to speak to user U1. Specifically, in response to user U1's operation, user terminal 10 launches the virtual agent application provided by distribution server 20 and requests distribution server 20 to distribute conversation content for the virtual agent to speak to user U1. For example, user U1 may launch the virtual agent application by tapping the virtual agent application icon Appl displayed on user terminal 10 and send a request to distribution server 20 for the distribution of conversation content for the virtual agent to speak to user U1. Alternatively, user U1 may launch the virtual agent application by making a predetermined utterance such as "Launch the virtual agent application." Note that this distribution request may also be sent to distribution server 20 when the virtual agent application is brought back to the foreground from the background.

[0023] In step S2, the distribution server 20, having received a request from the user terminal 10 to distribute conversation content for the virtual agent VA to speak to user U1, includes user information of user U1 in a prompt for the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to user U1, and generates the prompt.

[0024] Here, the user information for user U1 includes at least one of the following: fixed information about user U1, extracted information about user U1 obtained from the conversation history between user U1 and virtual agent VA, and current information about user U1.

[0025] Fixed information about user U1 includes information about user U1 such as age, gender, place of residence, occupation, workplace, working hours, break times, and holidays. Extracted information about user U1 from conversation history includes information about user U1 such as hobbies and interests, recent events, emotions, current interests, physical condition, and upcoming plans. Current information about user U1 includes information about user U1 such as time, location information, and location movement.

[0026] By including user information of user U1 in the above prompt, the artificial intelligence system 30 can generate conversational content that is appropriate for user U1's situation based on user information of user U1. As a result, the virtual agent VA can engage in conversation that is empathetic to user U1, using greetings and topics that are appropriate for user U1's situation.

[0027] Furthermore, although details will be described later, it is preferable to include, in addition to user information, commands to the virtual agent VA (in other words, the virtual agent's conversational stance towards the user) linked to the current status of user U1 as determined by the determination unit 222 described later, in the above prompt.

[0028] By including commands for the virtual agent VA, which are linked to the current status of the determined user U1, in addition to user information, the artificial intelligence system 30 can generate conversation content based on the commands linked to the current status of user U1. As a result, the virtual agent VA can engage in conversation that is more tailored to the user's current situation, using greetings and topics that are more appropriate to the user's current state.

[0029] In step S3, the distribution server 20 sends the generated prompt to the artificial intelligence system 30.

[0030] In step S4, the artificial intelligence system 30, having received a prompt from the distribution server 20, generates conversation content as a response to the prompt. The conversation content generated by the artificial intelligence system 30 is text data, but may also be audio data. In addition to the text data or audio data of the conversation content, the artificial intelligence system 30 may also generate image data related to the conversation content, motion data of the virtual agent VA related to the conversation content, and so on.

[0031] In step S5, the distribution server 20 receives the conversation content generated by the artificial intelligence system 30 from the artificial intelligence system 30.

[0032] In step S6, the distribution server 20, having received the conversation content generated by the artificial intelligence system 30 (the conversation content for virtual agent VA to speak to user U1), outputs the received conversation content to the user terminal 10. At this time, if the received conversation content is text data, the distribution server 20 may convert the text data into audio data and send it to the user terminal 10. The conversion from text data to audio data may also be performed on the user terminal 10 side.

[0033] In step S7, the user terminal 10, having received the conversation content, outputs the conversation content. Specifically, the user terminal 10 outputs the audio data of the conversation content from its speaker and / or displays the text data of the conversation content on its display unit. The user terminal 10 may also play motion data of the virtual agent VA related to the conversation content on its display unit. In the example shown in Figure 2, the user terminal 10 outputs the audio data and text data of the conversation content of the virtual agent VA speaking to user U1, which reads, "Good work today. Shall we have a drink tonight?"

[0034] In step S8, user U1, having recognized the voice data and / or text data "Good work today. Shall we have a drink tonight?", speaks, for example, "Yeah. Tomorrow's a day off, so maybe I'll drink a lot." into the microphone icon MI displayed on the user terminal 10's display. As a result, user terminal 10 acquires text data corresponding to "Yeah. Tomorrow's a day off, so maybe I'll drink a lot." through voice recognition.

[0035] In step S9, the user terminal 10 sends a distribution request to the distribution server 20, including the text data corresponding to the acquired "That's right. Tomorrow is a holiday, so maybe I'll drink a lot." in the distribution request for the virtual agent VA to user U1.

[0036] When the distribution server 20 receives a distribution request containing text data corresponding to "Yeah. Tomorrow is a day off, so maybe I'll drink a lot," it includes the text data as speech information of user U1 in the prompt, generates the prompt, and sends the generated prompt to the artificial intelligence system 30. When generating the prompt, similar to step S2, the prompt may also include a command to the virtual agent VA linked to user information and / or the current status of user U1 as determined by the determination unit 222 described later.

[0037] The subsequent processing is the same as in steps S4 to S7, and the process is repeated until user U1 stops talking to virtual agent VA.

[0038] As described above, in the information processing system S according to this embodiment, by including user information of user U1 in the prompts for the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to user U1, the artificial intelligence system 30 can generate conversation content that is appropriate to user U1's situation based on user information of user U1. Therefore, the virtual agent VA can speak to user U1 with greetings and topics that are appropriate to user U1's situation and are empathetic to user U1.

[0039] The following describes the detailed configuration of the user terminal 10 and the distribution server 20.

[0040] Figure 3 shows an example of the hardware configuration of the user terminal 10 according to this embodiment. The user terminal 10 is a computer such as a smartphone, tablet, notebook PC (Personal Computer), desktop PC, mobile phone, or PDA (Personal Digital Assistant).

[0041] The user terminal 10 is physically configured as a computer including a processor 101, memory 102, storage 103, communication device 104, input device 105, output device 106, and a bus connecting them. Each of these devices operates on power supplied from a battery (not shown). In the following description, the term "device" can be read as a circuit, device, unit, etc. The hardware configuration of the user terminal 10 may include one or more of the devices shown in Figure 2, or some of the devices may be omitted. Alternatively, multiple devices with different enclosures may be connected via communication to constitute the user terminal 10.

[0042] Each function in the user terminal 10 is realized by loading predetermined software (programs) onto hardware such as the processor 101 and memory 102, which allows the processor 101 to perform calculations, control communication by the communication device 104, and control at least one of the reading and writing of data in the memory 102 and storage 103.

[0043] The processor 101 controls the entire computer, for example, by running the operating system. The processor 101 may be composed of a central processing unit (CPU) that includes interfaces with peripheral devices, control devices, arithmetic units, registers, etc. Also, for example, a baseband signal processing unit or a call processing unit may be implemented by the processor 101.

[0044] The processor 101 reads programs (program code), software modules, data, and the like from at least one of the storage 103 and the communication device 104 into the memory 102, and executes various processes in accordance therewith. As the program, a program that causes a computer to execute at least part of the operations described later is used. The functional blocks of the user terminal 10 may be implemented by a control program stored in the memory 102 and operating on the processor 101. The various processes may be executed by a single processor 101, or may be executed simultaneously or sequentially by two or more processors 101. The processor 101 may be implemented by one or more chips. Note that the program may be transmitted to the user terminal 10 via an electric communication line.

[0045] The memory 102 is a computer-readable recording medium, and may be configured by at least one of, for example, ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), RAM (Random Access Memory), and the like. The memory 102 may also be referred to as a register, a cache, a main memory (main storage device), or the like. The memory 102 can store executable programs (program code), software modules, and the like for implementing the method according to the present embodiment.

[0046] The storage 103 is a computer-readable recording medium, and may be configured by at least one of, for example, optical discs such as CD-ROM (Compact Disc ROM), hard disk drives, flexible disks, magneto-optical discs (e.g., compact discs, digital versatile discs, Blu-ray (registered trademark) discs), smart cards, flash memories (e.g., cards, sticks, key drives), floppy (registered trademark) disks, magnetic strips, and the like. The storage 103 may also be referred to as an auxiliary storage device.

[0047] The communication device 104 enables communication with the distribution server 20. The communication device 104 may be configured to include a high-frequency switch, a duplexer, a filter, a frequency synthesizer, etc., for example, to implement at least one of Frequency Division Duplex (FDD) and Time Division Duplex (TDD). For example, a transmitting / receiving antenna, an amplifier unit, a transmitting / receiving unit, a transmission path interface, and the like may be implemented by the communication device 104. The transmitting / receiving unit may be implemented as a physically or logically separated transmitting unit and receiving unit.

[0048] The input device 105 is an input device that accepts external inputs (for example, various sensors such as keys, microphones, cameras, switches, buttons, position information sensors, motion sensors, etc.). The output device 106 is an output device that performs output to the outside (for example, a display, a speaker, an LED lamp, etc.). Note that the input device 105 and the output device 106 may have an integrated configuration (for example, a touch panel).

[0049] The respective devices such as the processor 101 and the memory 102 are connected by a bus for communicating information. The bus may be configured using a single bus, or may be configured using different buses for each pair of devices.

[0050] The user terminal 10 may be configured to include hardware such as a microprocessor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Programmable Logic Device (PLD), and a Field Programmable Gate Array (FPGA), and part or all of each functional block may be implemented by the hardware. For example, the processor 101 may be implemented using at least one of these types of hardware.

[0051] Figure 4 shows an example of the hardware configuration of the distribution server 20 according to this embodiment. Note that the hardware configuration of the artificial intelligence system 30 according to this embodiment is the same as the hardware configuration of the distribution server 20, so its description is omitted. The distribution server 20 is an information processing device realized by, for example, a computer, server device, cloud system, etc. Physically, the distribution server 20 is configured as a computer device including a processor 201, memory 202, storage 203, communication device 204, input device 205, output device 206, and a bus connecting these. The processor 201, memory 202, storage 203, input device 205, and output device 206 are the same hardware as the processor 101, memory 102, storage 103, input device 105, and output device 106 of the user terminal 10, so their description is omitted.

[0052] The communication device 204 is hardware (transceiver / receiver device) for communicating between computers via at least one of a wired network and a wireless network, and is also called a network device, network controller, network card, communication module, etc. The communication device 204 enables communication with the user terminal 10.

[0053] [Functional Configuration of Distribution Server 20] Next, the functional configuration of the distribution server 20 according to this embodiment will be described. Figure 5 is a diagram showing the functional configuration of the distribution server 20. The distribution server 20 has a storage unit 21 and a control unit 22.

[0054] The memory unit 21 stores various types of data. For example, the memory unit 21 stores programs executed by the control unit 22. The memory unit 21 stores programs that cause the control unit 22 to function as an acquisition unit 221, a determination unit 222, a prompt generation unit 223, a transmission unit 224, a reception unit 225, an output unit 226, and an extraction unit 227. The memory unit 21 also stores various types of information related to content, including the conversation content of the virtual agent VA, which is distributed to the user terminal 10.

[0055] The storage unit 21 includes a user information storage unit 211, a status command storage unit 212, and a conversation history information storage unit 213.

[0056] (User Information Storage Unit 211) The user information storage unit 211 stores various user information about the user. Figure 6 is a diagram showing an example of user information stored in the user information storage unit 211 according to this embodiment. As shown in Figure 6, the user information storage unit 211 stores user information for each "user ID" regarding items such as "fixed information," "extracted information," and "current information." The user information storage unit 211 may store at least one of "fixed information," "extracted information," and "current information" as user information for each "user ID." By referring to this information as user information, the virtual agent VA can speak to the user with greetings and topics that are more appropriate to the user's situation and are more empathetic to the user.

[0057] The "User ID" indicates identification information used to identify a user. For example, the user identified by User ID "U1" corresponds to User U1 shown in the example in Figure 2.

[0058] "Fixed information" refers to fixed information about a user identified by a user ID. "Fixed information" includes information such as "age," "gender," "place of residence," "occupation," "workplace," "working hours," "break times," and "holidays." "Age" indicates the age of the user identified by the user ID. "Gender" indicates the gender of the user identified by the user ID. "Place of residence" indicates the place of residence of the user identified by the user ID. "Occupation" indicates the occupation of the user identified by the user ID. "Workplace" indicates the workplace of the user identified by the user ID. "Working hours" indicates the working hours of the user identified by the user ID. "Break times" indicates the break times of the user identified by the user ID. "Holidays" indicates the holidays of the user identified by the user ID. "Fixed information" can be provided in advance by the user when the user identified by the user ID starts using the virtual agent application according to this embodiment. The user can also edit or update the fixed information when using the virtual agent application.

[0059] "Extracted information" is extracted information about the user from the conversation history between the user and the virtual agent VA stored in the conversation history information storage unit 213. "Extracted information" includes information such as "hobbies and preferences," "recent events," "physical condition," and "upcoming plans." "Hobbies and preferences" indicates the user's hobbies and preferences extracted from the conversation history between the user identified by the user ID and the virtual agent VA. "Recent events" indicates the user's recent events extracted from the conversation history between the user identified by the user ID and the virtual agent VA. "Physical condition" indicates the user's physical condition extracted from the conversation history between the user identified by the user ID and the virtual agent VA. "Upcoming plans" indicates the user's upcoming plans extracted from the conversation history between the user identified by the user ID and the virtual agent VA.

[0060] "Current information" refers to the current information about the user. "Current information" includes information such as "location information" and "location movement." "Location information" indicates the current location of the user identified by the user ID. "Location movement" indicates the movement of the user's current location identified by the user ID. This information is acquired by the location information sensor of the user terminal 10 and transmitted from the user terminal 10 to the distribution server 20, for example, when a request for distribution of conversation content is sent.

[0061] Furthermore, the user information storage unit 211 may store various types of information depending on the purpose, not limited to the information described above. For example, in addition to the information described above, the user information storage unit 211 may store other demographic attribute information or psychographic attribute information. For example, the user information storage unit 211 may store information such as name, interests, family structure, income, lifestyle, current obsessions, and emotions.

[0062] (Status command storage unit 212) The status command storage unit 212 stores the current status of multiple predetermined users and commands to multiple predetermined virtual agents VA, linking them together. Figure 7 is a diagram showing an example of the information stored in the status command storage unit 212 according to this embodiment. As shown in Figure 7, the information stored in the status command storage unit 212 consists of two items: "the user's current status" and "commands to the virtual agent." "Commands to the virtual agent" can also be said to be "the virtual agent's conversational stance towards the user."

[0063] The "user's current status" consists of a predetermined set of patterns, and the "commands to the virtual agent" also consist of the same number of predetermined commands as the "user's current status" patterns. These are all linked to each other. In this embodiment, as shown in Figure 7, the "user's current status" consists of seven patterns, but the number of patterns is not limited to this.

[0064] In the example in Figure 7, the "current user status" for pattern No. 1 indicates that today is a "workday," the user is currently "outside of working hours," and the user is currently "at home." This "current user status" for pattern No. 1 is linked to the "command to the virtual agent," which is "to make empathetic remarks to the user based on the given information."

[0065] In pattern No. 2, the "user's current situation" indicates that today is a "working day," the user is currently "outside of working hours," and the user is currently "on their way to work." This "user's current situation" in pattern No. 2 is linked to the "command to the virtual agent," which is "to make empathetic remarks to the user based on the given information."

[0066] In pattern No. 3, the "current user status" indicates that today is a "workday," the user is currently "within working hours," and the user is currently "on break." This "current user status" in pattern No. 3 is linked to the "command to the virtual agent," which is "to praise the user based on the given information."

[0067] In pattern No. 4, the "current user status" indicates that today is a "working day," it is currently "within working hours," and the user is currently "on duty." This "current user status" in pattern No. 4 is linked to the "command to the virtual agent," which is "to praise the user based on the given information."

[0068] In pattern No. 5, the "user's current status" indicates that today is a "holiday," the user has "no plans," and it is currently "AM." This "user's current status" in pattern No. 5 is linked to the "command to the virtual agent," which is "suggest how the user should spend their holiday based on the given information."

[0069] In pattern No. 6, the "user's current status" indicates that today is a "holiday," that there are "no plans" today, and that it is currently "PM." The "command to the virtual agent" associated with this "user's current status" in pattern No. 6 is "ask the user about their day based on the given information."

[0070] In pattern No. 7, the "user's current status" indicates that today is a "holiday" and that the user has "plans to make." The "command to the virtual agent" associated with this "user's current status" in pattern No. 7 is "reconfirm today's schedule with the user based on the given information."

[0071] The information (data table) stored in the status command storage unit 212 shown in Figure 7 is based on the assumption that the user of the virtual agent application according to this embodiment is an office worker. Different data tables may be prepared for each user attribute (for example, office worker, university student, high school student, housewife, househusband, etc.), and the data table used may be changed according to the user's attribute.

[0072] (Conversation History Information Storage Unit 213) The conversation history information storage unit 213 stores various information relating to the conversation history between the user and the virtual agent VA. Figure 8 is a diagram showing an example of conversation history information stored in the conversation history information storage unit 213 according to this embodiment. As shown in Figure 8, the information stored in the conversation history information storage unit 213 consists of two items: "User ID" and "Conversation History Information". The "Conversation History Information" includes items such as "Type", "Date and Time", "Conversation Content", and "Other".

[0073] "User ID" indicates identification information for identifying the user. "Conversation History Information" indicates the conversation history between the user and the virtual agent VA in the virtual agent application according to this embodiment. "Type" indicates the type of content in the virtual agent application. In the example in Figure 8, "Type" stores either "Output," which indicates that content was output to the user, or "Input," which indicates that speech information from the user's utterance was input.

[0074] "Date and Time" indicates the date and time when each piece of content was input or output. In the example in Figure 8, "Date and Time" is represented by an abstract code such as "Date and Time DA-1". "Conversation Content" indicates the text data contained in the content output at the corresponding date and time. "Other" indicates various information such as audio data and image data contained in the content displayed at the corresponding date and time.

[0075] For example, as shown in Figure 8, it indicates that content was output by the distribution server 20 to the user terminal 10 of the user identified by user ID "U1" (user U1) on the date and time "Date and Time DA-1". For example, it indicates that the content output by the distribution server 20 on the date and time "Date and Time DA-1" includes the conversation "Good work today. Shall we have a drink tonight?".

[0076] Furthermore, it indicates that content corresponding to user U1's speech information has been input from user terminal 10 to the date and time "Date and Time DA-2". For example, it indicates that the content input to the distribution server 20 for the date and time "Date and Time DA-2" includes the conversation "Yeah. Tomorrow is a holiday, so maybe I'll drink a lot."

[0077] Furthermore, the user terminal 10 of user U1 is shown that content was output by the distribution server 20 on the date and time "Date and Time DA-3". For example, the content output by the distribution server 20 on the date and time "Date and Time DA-3" includes the conversation "You worked hard this week. Let's drink a lot!".

[0078] Returning to Figure 5, the control unit 22 functions as an acquisition unit 221, a determination unit 222, a prompt generation unit 223, a transmission unit 224, a reception unit 225, an output unit 226, and an extraction unit 227 by executing the program stored in the storage unit 21.

[0079] The acquisition unit 221 acquires various types of information. The acquisition unit 221 acquires the content of conversations between the virtual agent VA and the user, or requests for the distribution of conversation content for the virtual agent VA to speak to the user. The acquisition unit 221 acquires speech information from the user terminal 10 based on the user's utterances. For example, the acquisition unit 221 acquires the user's utterance information from the user terminal 10 regarding the conversation content for the virtual agent VA to speak to the user. The acquisition unit 221 also acquires fixed information about the user and current information about the user from the user terminal 10.

[0080] The determination unit 222 determines the user's current status based on the user's user information. Specifically, the determination unit 222 determines the user's current status by selecting one of the user's current statuses from among several predetermined user current statuses based on the user's user information. For example, the determination unit 222 uses the user's user information (fixed information, extracted information, current information) to determine which of the seven patterns shown in Figure 7 the user's current status corresponds to. For example, if the determination unit 222 determines from the user's user information that today is a "working day," the user is currently "outside working hours," and the user is currently "on their way to work," it determines that this corresponds to pattern No. 2 in Figure 7 and selects pattern No. 2 as the user's current status.

[0081] The prompt generation unit 223 generates prompts to cause the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user. The prompt generation unit 223 includes the user's user information in the prompts to cause the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user.

[0082] Preferably, the prompt generation unit 223 generates a prompt that includes the user's user information and the command to the virtual agent VA (the virtual agent VA's conversational stance towards the user) associated with the user's current status as determined by the determination unit 222, in order to cause the artificial intelligence system 30 to generate conversational content for the virtual agent VA to speak to the user.

[0083] In other words, the prompt generation unit 223 may obtain from the status command storage unit 212 a command to the virtual agent VA associated with one of the user's current situations selected by the determination unit 222 based on the user's user information, and include this command, in addition to the user information, in the prompt for the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user. Specifically, if the determination unit 222 selects pattern No. 2 in Figure 7 as the "user's current situation" based on the user information, the prompt generation unit 223 may obtain the command "Make a user-friendly statement based on the given information" as a "command to the virtual agent" associated with the "user's current situation" of pattern No. 2, and include this command in the prompt for the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user.

[0084] By including the user's user information and commands to the virtual agent VA linked to the user's current status in the prompts for the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user, the artificial intelligence system 30 can generate conversation content based on commands linked to the user's current status. As a result, the virtual agent VA can speak to the user in a way that is more appropriate to the user's current situation, using greetings and topics that are more empathetic.

[0085] The prompt generation unit 223 may also generate a prompt for the artificial intelligence system 30 to generate the conversation content of the virtual agent VA to the user, including the user's user information and the command to the virtual agent VA (the virtual agent VA's conversational stance towards the user) associated with the user's current status as determined by the determination unit 222.

[0086] Furthermore, it can be said that the commands given to the virtual agent VA (the virtual agent VA's conversational stance towards the user) define how the virtual agent VA should respond to the user's current situation, based on the information provided.

[0087] By issuing commands to the virtual agent VA as defined in this way, the virtual agent VA can engage in conversation with the user in a way that is more tailored to the user's current situation, using greetings and topics that are more appropriate to the user's circumstances.

[0088] Furthermore, if the acquisition unit 221 acquires user utterance information, the prompt generation unit 223 includes the utterance information in the prompt for generating the conversation content of the virtual agent VA for the user. At this time, the prompt generation unit 223 may also include in the prompt a command to the virtual agent VA that is linked to the user information and / or the user's current status as determined by the determination unit 222.

[0089] This allows the virtual agent VA to expand on the topic of conversation with the user, starting from the initial message from the virtual agent VA.

[0090] Furthermore, the prompt generation unit 223 may include a command in the prompts for generating the conversation content of the virtual agent VA to the user that causes the artificial intelligence system 30 to determine the user's current status based on the user's user information. In this case, the user information may be modified based on the response from the artificial intelligence system 30 to the prompt.

[0091] This allows the artificial intelligence system 30 to accurately determine the user's current status. If the user information (especially fixed information) is not updated on the distribution server 20, the distribution server 20 may make a determination that does not match the actual user's current status. However, by having the artificial intelligence system 30 determine the user's current status, it can make an appropriate determination with high accuracy, even when there is a contradiction between the extracted information from the conversation history and the fixed information.

[0092] The transmitting unit 224 sends prompts to the artificial intelligence system 30. For example, the transmitting unit 224 sends prompts to the artificial intelligence system 30 to cause the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user. The transmitting unit 224 sends prompts to the artificial intelligence system 30 to cause the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user.

[0093] The receiving unit 225 receives the conversation content of the virtual agent VA to the user, which is generated by the artificial intelligence system 30 as a response to the prompt generated by the prompt generation unit 223.

[0094] The output unit 226 outputs the conversation content received by the receiving unit 225 as a response to a prompt to the user terminal 10. If the conversation content is in the form of text data, the output unit 226 may convert the text data into audio data and output it to the user terminal 10 together with or in place of the text data.

[0095] The extraction unit 227 extracts user information from the conversation history between the user and the virtual agent VA stored in the conversation history information storage unit 213. The information extracted by the extraction unit 227 is added to or overwritten in the "extracted information" field of the user information storage unit 211. Alternatively, the information extracted by the extraction unit 227 may be added to or overwritten in the "fixed information" and "current information" fields of the user information storage unit 211. The artificial intelligence system 30 may decide whether or not to add or overwrite the information extracted by the extraction unit 227 in the "fixed information" and "current information" fields of the user information storage unit 211.

[0096] By the extraction unit 227 extracting information from the conversation history, the distribution server 20 can refer to information other than the fixed information provided by the user in advance as user information, and the virtual agent VA can speak to the user with greetings and topics that are more appropriate to the user's situation and are more attentive to the user's needs.

[0097] [Functional Configuration of User Terminal 10] Next, the functional configuration of the user terminal 10 according to this embodiment will be described. Figure 9 is a diagram showing the functional configuration of the user terminal 10. The user terminal 10 has a storage unit 11, a control unit 12, a display unit 13, and an audio output unit 14.

[0098] The display unit 13 is, for example, a display that shows various kinds of information, and is an example of the output device 106 shown in Figure 3. Based on the control of the display control unit 124, the display unit 13 displays various images related to the virtual agent application according to this embodiment. For example, based on the control of the display control unit 124, the display unit 13 displays text data of the conversation content of the virtual agent VA. Alternatively, based on the control of the display control unit 124, the display unit 13 may display an image of the virtual agent VA corresponding to the motion data of the virtual agent VA related to the conversation content of the virtual agent VA.

[0099] The audio output unit 14 is, for example, a speaker that outputs (plays) various types of sound information, and is an example of the output device 106 shown in Figure 3. Based on the control of the audio control unit 125, the audio output unit 14 plays various sounds related to the virtual agent application according to this embodiment. For example, based on the control of the audio control unit 125, the audio output unit 14 plays audio data of the conversation content of the virtual agent VA.

[0100] The memory unit 11 stores various types of data. For example, the memory unit 11 stores programs executed by the control unit 12. The memory unit 11 stores programs that cause the control unit 12 to function as a receiving unit 121, a transmitting unit 122, a receiving unit 123, a display control unit 124, and a voice control unit 125. The memory unit 11 stores various image data of the virtual agent VA.

[0101] The control unit 12 functions as a receiving unit 121, a transmitting unit 122, a receiving unit 123, a display control unit 124, and a voice control unit 125 by executing a program stored in the storage unit 11.

[0102] The reception unit 121 receives input from the user. For example, the reception unit 121 receives input from the user via an operation unit such as a touch panel provided on the surface of the display unit 13, or via a microphone. When the reception unit 121 receives an operation to start the virtual agent application, it starts the virtual agent application and notifies the transmission unit 122 of the details of that operation. When the reception unit 121 receives speech information from the user to the virtual agent VA, it notifies the speech information to the transmission unit 122.

[0103] The transmitting unit 122 sends a content distribution request to the distribution server 20 via the communication device 104. For example, when the transmitting unit 122 receives notification from the receiving unit 121 regarding the operation content for launching the virtual agent application, it sends a distribution request to the distribution server 20 via the communication device 104 for the conversation content that the virtual agent VA will use to speak to the user. Also, when the transmitting unit 122 receives notification from the receiving unit 121 of the user's speech information to the virtual agent VA, it sends a distribution request to the distribution server 20 that includes text data corresponding to the speech information.

[0104] The receiving unit 123 receives the conversation content transmitted from the distribution server 20.

[0105] The display control unit 124 displays the text data of the conversation content of the virtual agent VA on the display unit 13. The display control unit 124 may also display an image of the virtual agent VA on the display unit 13 that corresponds to the motion data of the virtual agent VA related to the conversation content of the virtual agent VA.

[0106] The audio control unit 125 plays back the audio data of the conversation content of the virtual agent VA to the audio output unit 14.

[0107] [Operation Flow] Figure 10 is a flowchart showing an example of a process performed in the information processing system S according to this embodiment.

[0108] In step S101, the user terminal 10 starts the virtual agent application. Note that starting the virtual agent application here also includes bringing the virtual agent application back to the foreground from the background.

[0109] In step S102, the transmission unit 122 of the user terminal 10 sends a request to the distribution server 20 for the distribution of conversation content for the virtual agent VA to speak to the user.

[0110] In step S103, the prompt generation unit 223 of the distribution server 20, which has received a request from the user terminal 10 to distribute conversation content for the virtual agent to speak to the user, generates a prompt for the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user. Details of the process for generating the prompt in step S103 will be described later in Figure 11.

[0111] In step S104, the transmission unit 224 of the distribution server 20 transmits the prompt generated in step S103 to the artificial intelligence system 30.

[0112] In step S105, the artificial intelligence system 30, having received a prompt from the distribution server 20, generates conversation content (conversation content for virtual agent VA to speak to user U1) as a response to the prompt.

[0113] In step S106, the receiving unit 225 of the distribution server 20 receives the conversation content generated by the artificial intelligence system 30 in step S105 from the artificial intelligence system 30.

[0114] In step S107, the output unit 226 of the distribution server 20, which has received the conversation content generated by the artificial intelligence system 30, outputs the received conversation content to the user terminal 10.

[0115] In step S108, the user terminal 10, having received the conversation content, outputs the conversation content. Specifically, the voice control unit 125 of the user terminal 10 causes the voice output unit 14 to play the audio data of the conversation content, and / or the display control unit 124 displays the text data of the conversation content on the display unit 13.

[0116] A user who has recognized the audio and / or text data of the conversation speaks a response to the conversation towards the microphone icon MI displayed on the display unit 13 of the user terminal 10. As a result, in step S109, the reception unit 121 of the user terminal 10 acquires speech information, which is text data corresponding to the utterance, through speech recognition.

[0117] In step S110, the transmission unit 122 of the user terminal 10 sends a distribution request to the distribution server 20, including the acquired speech information, in the request for distribution of the conversation content of the virtual agent VA to the user.

[0118] When the acquisition unit 221 of the distribution server 20 receives a distribution request containing speech information, in step S111, the prompt generation unit 223 of the distribution server 20 generates a prompt that includes the speech information. When generating the prompt, the prompt may also include user information and / or commands to the virtual agent VA linked to the user's current status as determined by the determination unit 222.

[0119] In step S112, the transmission unit 224 of the distribution server 20 transmits the prompt generated in step S111 to the artificial intelligence system 30.

[0120] In step S113, the artificial intelligence system 30, having received a prompt from the distribution server 20, generates conversation content (the conversation content of the virtual agent VA in response to the user's utterance) as a response to the prompt.

[0121] In step S114, the receiving unit 225 of the distribution server 20 receives the conversation content generated by the artificial intelligence system 30 in step S113 from the artificial intelligence system 30.

[0122] In step S115, the output unit 226 of the distribution server 20, which has received the conversation content generated by the artificial intelligence system 30, outputs the received conversation content to the user terminal 10.

[0123] In step S116, the user terminal 10, having received the conversation content, outputs the conversation content. Specifically, the voice control unit 125 of the user terminal 10 causes the voice output unit 14 to play the audio data of the conversation content, and / or the display control unit 124 causes the text data of the conversation content to display the display unit 13.

[0124] After that, steps S109 to S116 are repeated until the user stops talking to the virtual agent VA.

[0125] Figure 11 is a flowchart illustrating an example of the process executed by the distribution server 20 in step S103 of Figure 10, which generates a prompt for the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user. In the flowchart of Figure 11, an example will be explained in which the prompt for the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user includes not only the user's user information but also commands to the virtual agent VA linked to the user's current status as determined by the determination unit 222.

[0126] In step S201, the acquisition unit 221 acquires a request from the user terminal 10 for the distribution of conversation content for the virtual agent VA to speak to the user.

[0127] In step S202, the determination unit 222 determines whether today is a workday or a holiday based on the user information of the user stored in the user information storage unit 211. If today is a workday, the flow proceeds to step S203. If today is a holiday, the flow proceeds to step S204.

[0128] In step S203 (if today is a working day), the determination unit 222 determines whether it is currently working hours or not based on the user information stored in the user information storage unit 211. If it is currently outside of working hours, the flow proceeds to step S205. If it is currently within working hours, the flow proceeds to step S206.

[0129] In step S205 (if it is currently outside working hours), the determination unit 222 determines whether the user is currently at home or on their way to work, based on the user information stored in the user information storage unit 211. If the user is currently at home, the flow proceeds to step S207. If the user is currently on their way to work, the flow proceeds to step S208. "On their way to work" means that the user is traveling to the company or has finished work and is traveling from the company to home.

[0130] In step S207, the determination unit 222 selects pattern No. 1 as the user's current status from the status command storage unit 212 based on the determination results in steps S202, S203, and S205. Then, the prompt generation unit 223 obtains from the status command storage unit 212 an instruction for the virtual agent VA (in this case, "make a statement that is empathetic to the user based on the given information") that is associated with the current status of one user (in this case, pattern No. 1) selected by the determination unit 222 based on the user information.

[0131] In step S209, the prompt generation unit 223 generates a prompt by including the user information stored in the user information storage unit 211, as well as the command obtained in step S207, in the prompt for the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user. At this time, the user information to be included in the prompt may be all of the user information stored in the user information storage unit 211, or it may be at least one of the user information used in the determination made by the determination unit 222, and the user information identified by the result of the determination. The flow then proceeds to step S104 in Figure 10.

[0132] In step S208, the determination unit 222 selects pattern No. 2 from the status command storage unit 212 as the user's current status based on the determinations made in steps S202, S203, and S205. The prompt generation unit 223 then obtains from the status command storage unit 212 an instruction for the virtual agent VA (in this case, "make a statement that is empathetic to the user based on the given information") that is associated with the current status of one user (in this case, pattern No. 2) selected by the determination unit 222 based on the user information.

[0133] In step S210, the prompt generation unit 223 generates a prompt that includes the user information stored in the user information storage unit 211, as well as the command obtained in step S208, to cause the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user. At this time, the user information to be included in the prompt may be all of the user information stored in the user information storage unit 211, or it may be at least one of the user information used in the determination made by the determination unit 222, and the user information identified by the result of the determination. The flow then proceeds to step S104 in Figure 10.

[0134] In step S206 (if it is currently working hours), the determination unit 222 determines whether the user is currently on break or not based on the user information stored in the user information storage unit 211. If the user is currently on break, the flow proceeds to step S211. If the user is not currently on break (i.e., it is outside of break time), the flow proceeds to step S212.

[0135] In step S211, the determination unit 222 selects pattern No. 3 as the user's current status from the status command storage unit 212 based on the determination results in steps S202, S203, and S206. Then, the prompt generation unit 223 obtains from the status command storage unit 212 an instruction for the virtual agent VA (in this case, "congratulate the user based on the given information") that is associated with the current status of one user (in this case, pattern No. 3) selected by the determination unit 222 based on the user information.

[0136] In step S213, the prompt generation unit 223 generates a prompt that includes the user information stored in the user information storage unit 211, as well as the command acquired in step S211, to cause the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user. At this time, the user information to be included in the prompt may be all of the user information stored in the user information storage unit 211, or it may be at least one of the user information used in the determination made by the determination unit 222, and the user information identified by the result of the determination. The flow then proceeds to step S104 in Figure 10.

[0137] In step S212, the determination unit 222 selects pattern No. 4 as the user's current status from the status command storage unit 212 based on the determinations made in steps S202, S203, and S206. The prompt generation unit 223 then obtains a command from the status command storage unit 212 to the virtual agent VA associated with the current status of one user (in this case, pattern No. 4) selected by the determination unit 222 based on the user information (in this case, "make a statement that is empathetic to the user based on the given information").

[0138] In step S214, the prompt generation unit 223 generates a prompt by including the user information stored in the user information storage unit 211, as well as the command obtained in step S213, in the prompt for the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user. At this time, the user information to be included in the prompt may be all of the user information stored in the user information storage unit 211, or it may be at least one of the user information used in the determination made by the determination unit 222, and the user information identified by the result of the determination. After that, the flow proceeds to step S104 in Figure 10.

[0139] In step S204 (if today is a holiday), the determination unit 222 determines whether or not there are any appointments for today based on the user information stored in the user information storage unit 211. If there are no appointments for today, the flow proceeds to step S215. If there are appointments for today, the flow proceeds to step S216.

[0140] In step S215 (when there are no plans for today), the determination unit 222 determines whether it is currently AM or PM based on the user information stored in the user information storage unit 211. If it is currently AM, the flow proceeds to step S217. If it is currently PM, the flow proceeds to step S218.

[0141] In step S217, the determination unit 222 selects pattern No. 5 as the user's current status from the status command storage unit 212 based on the determinations made in steps S202, S204, and S215. The prompt generation unit 223 then obtains a command from the status command storage unit 212 to the virtual agent VA associated with the current status of one user (in this case, pattern No. 5) selected by the determination unit 222 based on the user information (in this case, "Suggest how to spend your holidays to the user based on the given information").

[0142] In step S219, the prompt generation unit 223 generates a prompt by including the user information stored in the user information storage unit 211, as well as the command obtained in step S217, in the prompt for the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user. At this time, the user information to be included in the prompt may be all of the user information stored in the user information storage unit 211, or it may be at least one of the user information used in the determination made by the determination unit 222, and the user information identified by the result of the determination. After that, the flow proceeds to step S104 in Figure 10.

[0143] In step S218, the determination unit 222 selects pattern No. 6 from the status command storage unit 212 as the user's current status based on the determinations made in steps S202, S204, and S215. The prompt generation unit 223 then obtains a command from the status command storage unit 212 to the virtual agent VA associated with the current status of one user (in this case, pattern No. 6) selected by the determination unit 222 based on the user information (in this case, "Ask the user how their day was based on the information provided").

[0144] In step S220, the prompt generation unit 223 generates a prompt that includes the user information stored in the user information storage unit 211, as well as the command obtained in step S218, to cause the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user. At this time, the user information to be included in the prompt may be all of the user information stored in the user information storage unit 211, or it may be at least one of the user information used in the determination made by the determination unit 222, and the user information identified by the result of the determination. The flow then proceeds to step S104 in Figure 10.

[0145] In step S216, the determination unit 222 selects pattern No. 7 as the user's current status from the status command storage unit 212 based on the determination results in steps S202 and S204. The prompt generation unit 223 then obtains a command from the status command storage unit 212 to the virtual agent VA associated with the current status of one user (in this case, pattern No. 7) selected by the determination unit 222 based on the user information (in this case, "Confirm today's schedule with the user based on the given information").

[0146] In step S221, the prompt generation unit 223 generates a prompt by including the user information stored in the user information storage unit 211, as well as the command obtained in step S216, in the prompt for the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user. At this time, the user information to be included in the prompt may be all of the user information stored in the user information storage unit 211, or it may be at least one of the user information used in the determination made by the determination unit 222, and the user information identified by the result of the determination. After that, the flow proceeds to step S104 in Figure 10.

[0147] As described above, the distribution server 20 of the information processing system S according to this embodiment can include the user's user information in the prompts used to cause the artificial intelligence system 30 to generate conversation content for the virtual agent VA to speak to the user. This allows the artificial intelligence system 30 to generate conversation content that is appropriate to the user's situation based on the user's user information, so that the virtual agent VA can speak to the user in a way that is empathetic, using greetings and topics that are appropriate to the user's situation.

[0148] Furthermore, if the prompts used to generate conversation content for the virtual agent VA to speak to the user include commands to the virtual agent VA that are linked to the user's current situation as determined by the determination unit 222, then the artificial intelligence system 30 can generate conversation content for speaking to the user based on the commands linked to the user's current situation. As a result, the virtual agent VA can engage in conversation that is more tailored to the user, using greetings and topics that are more appropriate to the user's current situation.

[0149] [Other Embodiments] Although the information processing system S of the present invention has been described above with reference to this embodiment, the information processing system S of the present invention is not limited to the above embodiment. Various modifications to the configuration and details of the present invention can be made, which can be understood by those skilled in the art within the technical scope of the present invention. Furthermore, any system or device that combines the separate features included in this embodiment in any way is also within the technical scope of the present invention.

[0150] Furthermore, the present invention may be applied to a system composed of multiple devices or to a single device. Moreover, the present invention is also applicable when an information processing program that realizes the functions of this embodiment is supplied to a system or device and executed by a built-in processor. The technical scope of the present invention includes programs installed on a computer to realize the functions of the present invention on a computer, or the medium on which the program is stored, a server that downloads the program, and a processor that executes the program. In particular, at least a non-transitory computer-readable medium storing a program that causes a computer to execute the processing steps included in this embodiment described above is included in the technical scope of the present invention.

[0151] In the above embodiment, the user terminal 10 could communicate with the artificial intelligence system 30 via the distribution server 20, allowing the user to converse with the virtual agent VA. If the user terminal 10 can possess the functions of the control unit 22 of the distribution server 20 (the functions shown in Figure 5), and if the user terminal 10 can access the data in the storage unit 21 of the distribution server 20, the user terminal 10 can communicate directly with the artificial intelligence system 30, allowing the user to converse with the virtual agent VA.

[0152] Furthermore, while the above embodiment uses the launch of the virtual agent application as a trigger for sending a request to deliver the conversation content, it is not limited to this. For example, the system may be configured to send a request to deliver the conversation content to the delivery server 20 when the user picks up the user terminal 10 and turns their face towards the display unit 13 of the user terminal as a trigger. Alternatively, the system may be configured to send a request to deliver the conversation content to the delivery server 20 when the user starts moving with the user terminal 10 as a trigger.

[0153] The following additional notes are provided regarding each of the embodiments described above.

[0154] [Note 1] The information processing device is an information processing device that generates conversation content of a virtual agent to a user using a generative artificial intelligence model, and comprises: a prompt generation unit that includes the user information of the user in a prompt for the generative artificial intelligence model to generate the conversation content for the virtual agent to speak to the user; a transmission unit that transmits the prompt to the generative artificial intelligence model; a receiving unit that receives the conversation content as a response to the prompt generated by the generative artificial intelligence model; and an output unit that outputs the conversation content as a response to the prompt received by the receiving unit.

[0155] With the above configuration, the user's user information can be included in the prompts used by the generative artificial intelligence model to generate conversational content for the virtual agent to speak to the user. This allows the generative artificial intelligence model to generate conversational content that is appropriate to the user's situation based on the user's user information, enabling the virtual agent to engage in conversation that is empathetic to the user, using greetings and topics that are appropriate to the user's situation.

[0156] [Note 2] The information processing device of Note 1 further comprises a determination unit that determines the current status of the user based on the user information of the user, and the prompt generation unit includes the user information and the command to the virtual agent associated with the current status of the user determined by the determination unit in the prompt.

[0157] With the above configuration, the generative artificial intelligence model can generate conversational content based on commands linked to the user's current situation. As a result, the virtual agent can engage in conversation that is more tailored to the user's current situation, using appropriate greetings and topics.

[0158] [Note 3] In the information processing device described in Note 2, the determination unit determines the current status of a user by selecting the current status of one user from among a plurality of predetermined user current statuses based on the user information of the user, and the prompt generation unit obtains an instruction for the virtual agent associated with the current status of the selected user from a status instruction storage unit that stores the current statuses of the plurality of users and instructions for a plurality of predetermined virtual agents linked to each other, and includes the instruction in the prompt in addition to the user information.

[0159] With the above configuration, the generative artificial intelligence model can generate conversational content based on commands linked to the user's current situation. As a result, the virtual agent can engage in conversation that is more tailored to the user's current situation, using appropriate greetings and topics.

[0160] [Note 4] In the information processing device described in Note 2 or 3, the command to the virtual agent specifies how the virtual agent should respond to the determined current status of the user.

[0161] With the above configuration, the virtual agent can engage the user in conversations that are more tailored to the user's current situation, using greetings and topics that are more relevant to the user's needs.

[0162] [Note 5] In any of the information processing devices described in Notes 1 to 4, the user information includes at least one of the following: fixed information about the user, extracted information about the user extracted from the conversation history between the user and the virtual agent, and current information about the user.

[0163] By referring to this information as user information, the virtual agent can engage in conversations with the user that are more tailored to the user's situation, using greetings and topics that are more appropriate to the user's circumstances.

[0164] [Note 6] Any of the information processing devices described in Notes 1 to 5 further comprises an extraction unit that extracts user information from the conversation history between the user and the virtual agent.

[0165] With the above configuration, the information processing device can refer to information other than fixed information provided by the user in advance as user information, and the virtual agent can engage in conversation with the user using greetings and topics that are more appropriate to the user's situation and that are more attentive to the user's needs.

[0166] [Note 7] Any information processing device described in Notes 1 to 6 further includes an acquisition unit that acquires utterance information from the user regarding the conversation content for which the virtual agent speaks to the user, and the prompt generation unit includes the utterance information in a prompt for generating the conversation content of the virtual agent to the user when the acquisition unit has acquired the utterance information.

[0167] With the above configuration, the virtual agent can expand on the topic of conversation with the user based on the initial message from the virtual agent, enabling them to engage in a conversation.

[0168] [Note 8] In any of the information processing devices described in Notes 1 to 7, the prompt generation unit includes in the prompt for generating the content of the conversation between the virtual agent and the user an instruction to cause the generative artificial intelligence model to determine the current status of the user based on the user information.

[0169] With the above configuration, the generative artificial intelligence model can accurately determine the user's current status. If the user information is not updated in the information processing device, the information processing device may make a determination that does not match the actual user's current status. However, by having the generative artificial intelligence model determine the user's current status, it can accurately make an appropriate determination, even when there is a contradiction between the extracted information from the conversation history and the fixed information.

[0170] [Note 9] The information processing method is an information processing method that generates conversation content of a virtual agent to a user using a generative artificial intelligence model, comprising: the steps of: including the user information of the user in a prompt for the generative artificial intelligence model to generate the conversation content for the virtual agent to speak to the user; transmitting the prompt to the generative artificial intelligence model; receiving the conversation content generated by the generative artificial intelligence model as a response to the prompt; and outputting the conversation content as a response to the received prompt.

[0171] This configuration produces the same effects as the information processing device described in Appendix 1.

[0172] 10 User terminal 11 Storage unit 12 Control unit 13 Display unit 14 Audio output unit 121 Reception unit 122 Transmission unit 123 Receiving unit 124 Display control unit 125 Audio control unit 20 Distribution server (information processing device) 21 Storage unit 211 User information storage unit 212 Status command storage unit 213 Conversation history information storage unit 22 Control unit 221 Acquisition unit 222 Determination unit 223 Prompt generation unit 224 Transmission unit 225 Receiving unit 226 Output unit 227 Extraction unit 30 Artificial intelligence system (generative artificial intelligence model)

Claims

1. An information processing device that generates conversation content of a virtual agent to a user using a generative artificial intelligence model, comprising: a prompt generation unit that includes the user information of the user in a prompt for the generative artificial intelligence model to generate the conversation content for the virtual agent to speak to the user; a transmission unit that transmits the prompt to the generative artificial intelligence model; a receiving unit that receives the conversation content as a response to the prompt, which is generated by the generative artificial intelligence model; and an output unit that outputs the conversation content as a response to the prompt, which is received by the receiving unit.

2. The information processing apparatus according to claim 1, further comprising a determination unit that determines the current status of the user based on the user information of the user, wherein the prompt generation unit includes the user information and an instruction to the virtual agent associated with the current status of the user determined by the determination unit in the prompt.

3. The information processing apparatus according to claim 2, wherein the determination unit determines the current status of a user by selecting the current status of one user from among a plurality of predetermined current statuses of users based on the user information of the user, and the prompt generation unit obtains an instruction for the virtual agent associated with the current status of the selected user from a status instruction storage unit that stores the current statuses of the plurality of users and instructions for a plurality of predetermined virtual agents linked to each other, and includes the instruction in the prompt in addition to the user information.

4. The information processing apparatus according to claim 2, wherein the command to the virtual agent specifies how the virtual agent should respond to the determined current status of the user.

5. The information processing apparatus according to claim 1, wherein the user information includes at least one of fixed information about the user, extracted information about the user extracted from the conversation history between the user and the virtual agent, and current information about the user.

6. The information processing apparatus according to claim 1, further comprising an extraction unit for extracting user information from the conversation history between the user and the virtual agent.

7. The information processing apparatus according to claim 1, further comprising an acquisition unit that acquires utterance information from the user regarding the content of the conversation for which the virtual agent speaks to the user, wherein the prompt generation unit includes the utterance information in a prompt for generating the content of the conversation of the virtual agent to the user when the acquisition unit has acquired the utterance information.

8. The information processing apparatus according to claim 1, wherein the prompt generation unit includes a command to cause the generative artificial intelligence model to determine the current status of the user based on the user information in the prompt for generating the content of the conversation between the virtual agent and the user.

9. An information processing method for generating conversation content of a virtual agent to a user using a generative artificial intelligence model, comprising: a step of including the user information of the user in a prompt for causing the generative artificial intelligence model to generate the conversation content for the virtual agent to speak to the user; a step of transmitting the prompt to the generative artificial intelligence model; a step of receiving the conversation content generated by the generative artificial intelligence model as a response to the prompt; and a step of outputting the conversation content as a response to the received prompt.

Images