Conversation control system and conversation control program

Abstract

The AI ​​avatar engages in conversation that praises the user based on their actions. [Solution] A conversation control system according to one aspect of the present disclosure comprises: an action information acquisition unit that acquires user action information from a user terminal; a virtual space control unit that places an avatar and an AI avatar corresponding to the user in a virtual space; and a dialogue processing unit that controls the dialogue between the avatar and the AI ​​avatar and causes the AI ​​avatar to speak an answer to a question entered by the user, wherein the dialogue processing unit causes the AI ​​avatar to speak content praising the user based on the action information.

Description

Technical Field

[0001] The present disclosure relates to a conversation control system and a conversation control program.

Background Art

[0002] In recent years, virtual spaces (metaverses) constructed on networks have been provided, enabling users to perform various actions such as communicating with other users by operating an avatar (their own alter ego). When communicating with other users and receiving praise, the need for approval is satisfied, the quality of life is improved, and self - affirmation increases.

[0003] However, communication (conversation) with other users may sometimes lower self - affirmation by comparing oneself with other users or cause fatigue.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Patent Document 3

Summary of the Invention

Problems to be Solved by the Invention

[0005] An object of the present disclosure is to provide a conversation control system and a conversation control program in which an AI avatar conducts a conversation to praise a user according to the user's actions. Another object of the present disclosure is to provide a conversation control system that transmits a message to praise a user according to the user's actions.

Means for Solving the Problems

[0006] A conversation control system according to one aspect of the present disclosure includes: an action information acquisition unit that acquires user action information from a user terminal; a virtual space control unit that places an avatar and an AI avatar corresponding to the user in a virtual space; and a dialogue processing unit that controls the dialogue between the avatar and the AI ​​avatar and causes the AI ​​avatar to speak an answer to a question entered by the user, wherein the dialogue processing unit causes the AI ​​avatar to speak content praising the user based on the action information.

[0007] A conversation control program according to one aspect of the present disclosure causes a computer to perform the following steps: acquire user behavior information from a user terminal; place an avatar and an AI avatar corresponding to the user in a virtual space; control the dialogue between the avatar and the AI ​​avatar; cause the AI ​​avatar to speak an answer to a question entered by the user; and cause the AI ​​avatar to speak content praising the user based on the behavior information. [Effects of the Invention]

[0008] According to this disclosure, the AI ​​avatar can engage in conversations praising the user or send messages praising the user to the user's device in response to the user's actions. [Brief explanation of the drawing]

[0009] [Figure 1] This is a schematic diagram of a conversation control system according to an embodiment of the present disclosure. [Figure 2] This is a schematic diagram of the virtual space displayed on the user's terminal. [Figure 3] This is a schematic diagram of the virtual space displayed on the user's terminal. [Figure 4] This is a schematic diagram of the response confirmation screen displayed on the user's terminal. [Figure 5] This is a functional block diagram of the server device. [Modes for carrying out the invention]

[0010] The embodiments of this disclosure will be described below with reference to the drawings.

[0011] As shown in Figure 1, the conversation control system comprises a server device 1 and a user terminal 2, and the server device 1 and the user terminal 2 are connected to each other via a network such as the Internet.

[0012] User terminal 2 is an information processing device such as a smartphone or tablet. The user can use user terminal 2 to experience the virtual space (also called the "metaverse") provided by server device 1 and communicate with the AI ​​avatar described later.

[0013] For example, if user terminal 2 is a smartphone, the touch panel displays images of the virtual space and accepts user input to move their avatar. The speaker on user terminal 2 outputs sounds from within the virtual space. This allows the user to experience the virtual space through sight and sound.

[0014] When a user creates their own avatar by operating user terminal 2, they register user information such as age, gender, hobbies, and goals.

[0015] User terminal 2 transmits various information about the user's actions (behavioral information) to server device 1. Behavioral information includes, for example, website browsing history, search history, terminal usage time, alarm setting time, and sensor measurement results from acceleration sensors and barometric pressure sensors.

[0016] Figure 2 schematically shows an example of how the virtual space is displayed on user terminal 2. The virtual space is a space that includes multiple pre-created AI (Artificial Intelligence) avatars 3 and an avatar 4 created by the user. Each of the multiple AI avatars 3 is associated with a character that has a different role assigned to it. Roles include, for example, "good at giving compliments," "advisor," "conversational listener," and "chatty."

[0017] The user operates the user terminal 2, moves their own avatar 4, and talks to the AI avatar 3. As shown in FIG. 3, the role of the AI avatar 3 is displayed before talking to it. When the talk button B is pressed, an answer confirmation screen as shown in FIG. 4 is displayed, and the conversation starts.

[0018] When the user operates the user terminal 2, enters characters, numbers, symbols, etc. in the input field on the screen, and presses the send button, the user terminal 2 transmits the input information to the server device 1 as the user's speech information.

[0019] When the server device 1 receives the speech information transmitted from the user terminal 2, it determines the content of the speech and returns a response. For example, when the content of the speech is a question, it returns an answer to the question, and when it is a casual conversation, it returns a casual conversation response. Therefore, the server device 1 can also be called a chatbot.

[0020] In a conversation with the AI avatar 3 with the role of being good at praising, the server device 1 makes the AI avatar 3 praise the user from the action information obtained in advance.

[0021] For example, from the search history information of the user searching for the meaning of English words on the user terminal 2 or the browsing history information of viewing English conversation videos, it says "You're really working hard on learning English." From the search history information of searching for things related to child-rearing, it says "It must be tough. I really respect you for working hard every day." From the history information of purchasing a present for a friend on an e-commerce site, it says "You have good taste." or "You have a kind heart."

[0022] From the information on the daily alarm setting time of 4:30, it says "It's amazing that you wake up early every day." or "It's amazing that you can wake up in the cold." When it is determined from the sensor information that the user has climbed a lot of stairs, it says "It's amazing that you can climb the stairs even though you're tired."

[0023] When users receive praise, their need for recognition is fulfilled, and their self-esteem increases. Furthermore, because the conversation is with an AI avatar, there is no emotional strain like in conversations with other people, and there is no comparison to others, allowing users to enjoy the conversation at their own pace.

[0024] Furthermore, since each of the multiple AI avatars 3 is assigned a different role, various conversations can be held. For example, in a conversation with AI avatar 3 assigned the role of advisor, server device 1 causes AI avatar 3 to speak advice to the user based on behavioral information and user information such as goals that have been acquired in advance. If the goal is weight loss and sensor information determines that the number of steps taken that day was low, it will say, "Tomorrow, you should get off the train one stop earlier and walk the rest of the way."

[0025] In conversations with AI avatar 3, which is set to act as a listener or conversationalist, server device 1 causes AI avatar 3 to speak to the user based on user information such as hobbies obtained in advance and weather information obtained in conjunction with an external API.

[0026] Advertisements may be displayed in the virtual space. Preferably, the advertisements should be related to the user's goals and interests. For example, if the user's goals or interests are language-related, advertisements for English conversation classes or language learning apps should be displayed. If the user's goal is weight loss, advertisements for gyms, Pilates, yoga, or protein supplements should be displayed. If the user's hobby is cooking, advertisements for cooking classes or food delivery services should be displayed.

[0027] If a user does not have hobbies or goals registered in their user information, advertisements for job search sites, detergents, beverages, etc., will be displayed based on their search history. Advertisements for clothing and home appliance online stores may also be displayed based on age and gender.

[0028] Next, the configuration of server device 1 will be described. Server device 1 is a computer equipped with a communication interface, CPU, memory, etc. The CPU executes a conversation control program stored in the memory, thereby realizing the functions of the user registration unit 11, the behavior information acquisition unit 12, the virtual space control unit 13, the dialogue processing unit 14, and the advertisement display unit 15, as shown in Figure 5.

[0029] The user registration unit 11 communicates with the user terminal 2 and accepts user registration. During user registration, it accepts input such as user ID, password, age, gender, hobbies, and goals, and stores this user information in the storage unit 10. The user registration unit 11 also accepts settings for information that can be acquired (tracked) from the user terminal 2. In addition, the user registration unit 11 accepts the user's own avatar settings (body parts, clothing, etc.) and stores the avatar information in the storage unit 10.

[0030] The memory unit 10 also stores avatar information for multiple AI avatars created in advance by the service provider. Each of the multiple AI avatars has a different role assigned to it.

[0031] The memory unit 10 includes non-volatile memory and volatile memory. Non-volatile memory is, for example, ROM, EPROM, EEPROM, HDD, etc. Volatile memory is, for example, RAM.

[0032] The behavior information acquisition unit 12 acquires behavior information from the user terminal 12, including website browsing history, search history, alarm setting time, and various sensor information, and stores it in the storage unit 10.

[0033] The virtual space control unit 13 provides virtual space services. The virtual space control unit 13 uses known methods to display the virtual space on the user terminal 2, allows the user to move their own avatar within the virtual space by operating the user terminal 2, and enables the user to communicate with an AI avatar through their own avatar.

[0034] The dialogue processing unit 14 performs response processing according to the content of the user's utterance. The dialogue processing unit 14 receives the utterance entered by the user into the user terminal 2 and generates a response to the utterance. For generating the response, natural language processing using rule-based or large-scale language models, AI chatbot technology, etc., can be used. As an example of a large-scale language model, a text generation language model called GPT-3 (Generative Pre-trained Transformer 3) can be applied. In rule-based natural language processing, the response content of the AI ​​avatar is generated based on rules that pre-describe the response content in relation to the intent of the user's utterance.

[0035] For example, the dialogue processing unit 14 understands the content of the utterance and the user's intent through morphological analysis, dependency parsing, phrase structure analysis, semantic analysis, and contextual analysis, and then generates an appropriate response. Alternatively, the user's utterance may be input into a machine learning model that has been pre-trained to interpret the user's intent, and the intent of the user's utterance may be interpreted.

[0036] Here, machine learning is a technique for enabling computers to acquire human-like learning abilities. It refers to a technique in which a computer autonomously generates algorithms necessary for data identification and other judgments from pre-inputted training data, and then applies these algorithms to new data to make predictions. The learning method for machine learning can be supervised learning, unsupervised learning, semi-supervised learning, reinforcement learning, or deep learning, or a combination of these learning methods; the learning method for machine learning is not restricted.

[0037] Furthermore, the dialogue processing unit 14 uses the user's behavior information and user information acquired in advance to generate response content corresponding to the role set for the AI ​​avatar.

[0038] The response content includes verbal responses representing the content spoken by the AI ​​avatar, and nonverbal responses representing the AI ​​avatar's facial expressions, gestures, etc. The dialogue processing unit 14 performs speech synthesis processing to convert the verbal responses into speech using speech synthesis technology. The dialogue processing unit 14 also reflects facial expressions, gaze, posture, or emotions in the AI ​​avatar according to the nonverbal responses. Furthermore, the dialogue processing unit 14 performs lip-syncing, drawing the mouth movements in accordance with the AI ​​avatar's speech.

[0039] The dialogue processing unit 14 sends the response content as text, audio, and a dialogue screen to the user terminal 2. The user terminal 2 displays the text and dialogue screen and outputs audio. Through this dialogue processing, conversations and casual chats between the user and the AI ​​avatar are realized.

[0040] The advertising display unit 15 displays advertisements in the virtual space. The advertisements are preferably related to the user's goals and interests within the user information.

[0041] In the above embodiment, a configuration was described in which the user logs into a virtual space with an avatar and receives praise by conversing with an AI avatar. However, the server device 1 may create a message praising the user from the behavioral information obtained from the user terminal 2 and send the message as a push notification to the user terminal 2 on which the app is installed.

[0042] Although this disclosure has been described in detail using specific embodiments, it will be apparent to those skilled in the art that various modifications are possible without deviating from the intent and scope of this disclosure. [Explanation of symbols]

[0043] 1 Server device 2 User terminals

Claims

1. A behavior information acquisition unit that acquires user behavior information from the user terminal, A virtual space control unit that places avatars and AI avatars corresponding to the user in a virtual space, A dialogue processing unit controls the dialogue between the avatar and the AI ​​avatar, and causes the AI ​​avatar to speak an answer to a question entered by the user. Equipped with, The dialogue processing unit is a conversation control system that causes the AI ​​avatar to speak content praising the user based on the behavioral information.

2. The virtual space control unit places a plurality of AI avatars in the virtual space, The conversation control system according to claim 1, wherein each of the plurality of AI avatars is assigned a different role.

3. The conversation control system according to claim 1, wherein the behavioral information includes the user terminal's website browsing history, search history, alarm setting time, or sensor measurement results.

4. Steps include obtaining user behavior information from the user's terminal, The steps include placing an avatar and an AI avatar corresponding to the user in a virtual space, The steps include controlling the dialogue between the avatar and the AI ​​avatar, The steps include causing the AI ​​avatar to speak an answer to the question entered by the user, The steps include causing the AI ​​avatar to speak a message praising the user based on the aforementioned behavioral information, A conversational control program that causes a computer to execute a command.

5. A conversation control system that acquires user behavior information from a user terminal and sends a message praising the user to the user terminal based on the behavior information.

Images