system

Abstract

A system is provided.SOLUTION: A system, comprising: means for receiving information of family members, friends, and pets input by a user; means for analyzing the received information and classifying characters and voice features of the family members, friends, and pets; means for training a AI model based on the analyzed information; means for performing a conversation simulation with the user by using the trained AI model; and means for setting an API endpoint accessible by the user end.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Individuals who feel lonely, isolated, or troubled often seek psychological stability by conversing with a being who understands them, but in reality, such a being is often not nearby. Furthermore, the person they want to talk to may be a deceased relative, an ill person who is unable to speak, or a pet. Furthermore, some people feel reluctant to confide their feelings directly to family or friends. These circumstances make it difficult to obtain psychological support. The present invention aims to solve these problems and provide a virtual being that can provide emotional support. [Means for solving the problem]

[0005] The present invention provides a "virtual family" system in which AI recreates the personalities and voices of family, friends, and pets based on information entered by the user, and responds to everyday conversations and consultations. Specifically, the system includes a means for receiving information about family, friends, and pets entered by the user, a means for analyzing the received information and classifying personality and voice characteristics, a means for training an AI model based on the analyzed data, a means for conducting conversation simulations with the user using the trained AI model, and a means for setting an API endpoint that allows access from the user's device. This allows users to receive psychological support while having natural conversations with their virtual family members.

[0006] "User" refers to each individual who uses the System.

[0007] "Family, friend, and pet information" refers to various information that a user enters into the system about a specific family member, friend, or pet, such as their name, personality, vocal characteristics, and relationships.

[0008] "Means of receiving" refers to the technical methods and processes by which the server receives the information entered by the user.

[0009] "Means of analysis" refers to the technical methods and processes used to classify received information and extract personality and voice characteristics.

[0010] "Classification methods" refers to the technical methods or processes used to classify personality or vocal characteristics into specific categories based on analyzed information.

[0011] An "AI model" refers to an algorithm or data structure that uses machine learning to recreate the personalities and voices of family, friends, and pets.

[0012] "Training means" refers to the technical means by which an AI model is trained using the received and analyzed data.

[0013] "Means for conducting conversational simulation" refers to the technical methods and processes for using a trained AI model to realize natural conversation with a user.

[0014] "API endpoint" refers to a program interface on a server that can be accessed from a user's device.

[0015] "Means enabling access" refers to the technical methods for accessing the AI ​​model on the server from the user's device and sending and receiving the necessary data.

[0016] The term "system" refers to the overall configuration that encompasses each of these means and enables conversations between the user and their virtual family. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0018] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0019] First, the terms used in the following description will be explained.

[0020] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0021] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0022] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0023] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0030] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0035] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0037] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0038] This invention is a system that uses AI technology to provide everyday conversations and consultations with virtual family members to help individuals who are feeling lonely, isolated, or worried to find psychological stability. The system is designed to receive input from users, train an AI model based on that information, and engage in natural conversations with the users.

[0039] System Configuration

[0040] Server-side configuration

[0041] The server has the following features:

[0042] Data reception function: Receives information about family, friends, and pets sent from the user's device.

[0043] Data analysis function: Analyzes received information and classifies personality and voice characteristics.

[0044] Feature extraction function: Extracts important features from the analyzed data.

[0045] AI model training function: Trains an AI model based on feature-extracted data.

[0046] Conversation simulation function: Conducts conversation simulations with users using a trained AI model.

[0047] API endpoint setting function: Set a trained AI model to an API endpoint that can be accessed from the device.

[0048] Terminal configuration

[0049] The terminal has the following features:

[0050] Information input screen: Displays a screen where users can enter information about family, friends, and pets.

[0051] Data transmission function: Sends the entered information to the server.

[0052] Response reception function: Receives a response from the server.

[0053] Content display function: Displays the received response to the user or provides an audio response using speech synthesis.

[0054] Conversation log saving function: Saves conversation logs between the user and their virtual family.

[0055] User Actions

[0056] 1. Log in

[0057] The user launches the application on their device and enters their authentication information on the login screen. The first time they log in, they are required to create a new account.

[0058] 2. Enter information

[0059] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[0060] 3. Start a conversation

[0061] After the information is sent to the server, the user can initiate everyday conversations such as "How was your day?"

[0062] Specific examples

[0063] Example 1: Everyday conversation

[0064] The user launches the application and enters information about their mother. For example, the "mother"'s personality can be "kind," her voice characteristic can be "high-pitched," and her relationship status can be "understanding." If the user says, "I had a hard time at work today," the virtual mother will respond, "Really? What happened?" If the user continues, "My boss scolded me," the virtual mother will empathize, saying, "That must have been tough. But you did a great job."

[0065] Example 2: Consultation

[0066] The user inputs information about a friend, and for example, sets the friend's personality to "energetic," the voice characteristics to "mid-tone," and the relationship to "good at encouraging others." If the user says, "I've been feeling down lately," the virtual friend will show an accepting attitude by saying, "What happened? Tell me." If the user says, "A lot of things have been happening at once...," the virtual friend will encourage them by saying, "That's tough, but I know you can get through it!"

[0067] In this way, the present invention makes full use of AI technology to provide users with an experience that makes them feel as if they are conversing with real family, friends, or pets, providing psychological support.

[0068] The processing flow will be explained below.

[0069] Program processing flow

[0070] Server-side processing

[0071] Step 1:

[0072] The server receives information about family, friends, and pets entered by the user, including their names, personalities, vocal characteristics, and relationships.

[0073] Step 2:

[0074] The server analyzes the received information and classifies the personality and voice characteristics into specific categories. For example, personality can be classified as "friendly," "strict," or "energetic," while voice characteristics can be classified as "gentle," "mid-range," or "high-pitched."

[0075] Step 3:

[0076] The server extracts important features from each piece of analyzed information, allowing the AI ​​to more accurately recreate the personalities and characteristics of family, friends, and pets.

[0077] Step 4:

[0078] The server trains an AI model based on the feature-extracted data using a machine learning algorithm (e.g., a natural language processing model).

[0079] Step 5:

[0080] The server uses the trained AI model to simulate a conversation with the user, using natural language processing based on the user's input.

[0081] Step 6:

[0082] The server deploys the trained AI model to an API endpoint, making it accessible from user devices.

[0083] Terminal side processing

[0084] Step 1:

[0085] The user launches the application on their device and enters their authentication information on the login screen. The first time they log in, they are required to create a new account.

[0086] Step 2:

[0087] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[0088] Step 3:

[0089] The device sends the entered information to the server using HTTPS, ensuring secure communication.

[0090] Step 4:

[0091] The device receives a response from the server. The response is the content of the conversation generated by the AI ​​model (text data, voice data, etc.).

[0092] Step 5:

[0093] The device displays the received response to the user. In addition to displaying it in text format, it can also respond in voice using speech synthesis. The device uses a speech synthesis library or service (e.g., Google Text-to-Speech).

[0094] Step 6:

[0095] The device will store conversation logs between the user and their virtual family, which will then be sent to a server for analysis and to improve the AI ​​model.

[0096] User Actions

[0097] Step 1:

[0098] The user launches the app and enters their credentials on the login screen. The first time they log in, they are required to create a new account.

[0099] Step 2:

[0100] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[0101] Step 3:

[0102] The user initiates a conversation, for example, "How was your day?", and the virtual family member responds, "You did a great job today. Did anything special happen?"

[0103] Step 4:

[0104] For example, if a user says, "Work has been tough lately," their virtual family member will empathize and respond, "That must be tough. Tell me more about what happened. I hope it makes you feel a little better."

[0105] In this way, the system can provide psychological support to the user through each step.

[0106] Example 1

[0107] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0108] In modern society, many individuals suffer from loneliness, isolation, and worries, and there is a need for a means to receive psychological support. However, due to the limited time and opportunity to consult or talk with family and friends, there is a lack of effective systems to provide such psychological support. Another issue is the lack of systems that can provide dialogue tailored to the individual needs of users.

[0109] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0110] In this invention, the server includes means for receiving information about family, friends, and pets input by the user, means for analyzing the received information and classifying the personalities and vocal characteristics of the family, friends, and pets, means for training an AI model based on the analyzed and feature-extracted data, means for conducting a conversation simulation with the user using the trained AI model, means for setting an API endpoint that enables access from the user terminal, means for the user to input authentication information into a login screen for authentication, means for displaying the information input by the user as text or voice or responding using voice synthesis, and means for saving a conversation log between the user and the virtual family. This enables the user to receive psychological support through natural conversations with their virtual family and friends.

[0111] A "user" is an individual who uses the system to enter information about family, friends, and pets and engage in virtual interactions.

[0112] A "server" is a computer system that receives information sent by users and performs analysis, feature extraction, AI model training, conversation simulation, etc.

[0113] A "terminal" is a device through which a user inputs information and engages in virtual interactions via communication with a server. Examples include smartphones and tablets.

[0114] The "data receiving means" is a function that allows the server to receive information about family, friends, and pets sent by the user.

[0115] "Data analysis means" is a function for analyzing received information and classifying the personalities and vocal characteristics of family, friends, and pets.

[0116] "Feature extraction means" is a function that extracts important features from the analyzed data and uses them to train the AI ​​model.

[0117] "AI model training means" is a function for training an artificial intelligence model based on feature-extracted data.

[0118] "Conversation simulation means" is a function for simulating a virtual conversation between a user and the AI ​​model using a trained AI model.

[0119] The "API endpoint setting means" is a function for setting a trained AI model as an API endpoint that can be accessed from a user terminal.

[0120] The "login authentication means" is a function in which a user inputs authentication information into a login screen and the server verifies that information.

[0121] "Information display means" is a function for displaying information entered by the user as text or voice, or for responding using voice synthesis.

[0122] The "conversation log storage means" is a function for storing conversation logs between the user and the virtual family and transmitting them to the server as reusable data later.

[0123] "Virtual family" refers to family, friends, and pets recreated by an AI model trained based on information entered by the user.

[0124] This invention is a system that uses artificial intelligence (AI) technology to provide everyday conversations and consultations with a virtual family, helping individuals experiencing loneliness, isolation, or anxiety to find psychological stability. The system is designed to receive user input, train an AI model based on that information, and have natural conversations with the user.

[0125] System Configuration

[0126] Server-side configuration

[0127] The server has the following features:

[0128] Data reception function: Receives information about family, friends, and pets sent by the user from the device. Specifically, it receives data such as names, personalities, voice characteristics, and relationships entered by the user.

[0129] Data analysis function: Analyzes the received information and classifies the personalities and vocal characteristics of family, friends, and pets. For example, it analyzes personality and vocal characteristics such as "gentle" or "high-pitched."

[0130] Feature extraction function: Extracts important features from the analyzed data, thereby extracting the elements necessary for training the AI ​​model.

[0131] AI model training function: The AI ​​model is trained based on the feature-extracted data, for example, using a deep learning algorithm to replicate the personality and voice of specific family or friends selected by the user.

[0132] Conversation simulation function: A trained AI model is used to simulate a conversation with the user, generating appropriate responses to prompts entered by the user.

[0133] API endpoint setting function: Set the trained AI model to an API endpoint accessible from the device, allowing users to easily access it at any time.

[0134] Login authentication function: A function that allows users to enter authentication information on the login screen and perform authentication.

[0135] Terminal configuration

[0136] The terminal has the following features:

[0137] Information entry screen: Displays a screen where users can enter information about family, friends, and pets, including details such as their names, personalities, vocal characteristics, and relationships.

[0138] Data transmission function: Sends the entered information to the server. For example, it sends data to the server using an HTTP POST request.

[0139] Response reception function: Receives responses from the server, analyzes the received responses, and displays them in an easy-to-understand manner for the user.

[0140] Content display function: A function that displays the received response to the user, and can display text or output audio using voice synthesis.

[0141] Conversation log saving function: The function to save the conversation log between the user and the virtual family, which can be used to review the conversation later.

[0142] Usage example

[0143] Example 1: Everyday conversation

[0144] The user launches the application and enters information about their mother. For example, the "mother"'s personality can be "kind," her voice characteristic can be "high-pitched," and her relationship status can be "understanding." If the user says, "I had a hard time at work today," the virtual mother will respond, "Really? What happened?" If the user continues, "My boss scolded me," the virtual mother will empathize, saying, "That must have been tough. But you did a great job."

[0145] Example 2: Consultation

[0146] The user enters information about a friend, setting the friend's personality to "energetic," voice characteristics to "mid-tone," and relationship to "good at encouraging others." If the user says, "I've been feeling down lately," the virtual friend will respond with an accepting attitude, "What happened? Tell me." If the user says, "A lot of things have been happening at once...," the virtual friend will encourage them, saying, "That's tough, but I know you can get through it!"

[0147] With the above configuration, the system provides users with an experience that makes them feel as if they are having a conversation with their real family, friends, or pets, and can provide psychological support.

[0148] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0149] Step 1:

[0150] User login: The user launches the application and enters authentication information on the login screen. When the user enters their email address and password and presses the "Login" button, the authentication information is sent from the device to the server. The server compares this with the database, and if authentication is successful, the user is taken to the home screen. The input is the user's authentication information, and the output is the authentication result.

[0151] Step 2:

[0152] Information input: An information input screen is displayed for the user to enter information about family, friends, and pets. Here, information such as "name," "personality," "voice characteristics," and "relationship" are entered in the input fields. When the user confirms the input and presses the "send" button, the data is sent from the device to the server. The input is the information about family members, etc. entered by the user, and the output is the sent data.

[0153] Step 3:

[0154] Data transmission: The device sends the information entered by the user to the server. Specifically, the device's data transmission function converts the input data into JSON format and sends it to the server as an HTTP POST request. The input is the user's input data, and the output is the data sent to the server.

[0155] Step 4:

[0156] Data reception: The server receives the data sent from the terminal. The data reception function analyzes the HTTP request, extracts the user data, and stores it in the database. The input is the data in the HTTP request format, and the output is the stored data.

[0157] Step 5:

[0158] Data analysis: The server analyzes the received data. The data analysis function analyzes information such as name, personality, vocal characteristics, and relationships, and performs classification. For example, characteristics such as "kind," "high-pitched voice," and "understanding" are analyzed. The input is the saved user data, and the output is the analyzed classified data.

[0159] Step 6:

[0160] Feature Extraction: The server extracts important features from the analyzed data. The feature extraction function quantifies the user data and generates parameters for use in training the AI ​​model. The input is the analyzed classified data, and the output is the feature-extracted data.

[0161] Step 7:

[0162] AI model training: The server trains the AI ​​model based on the feature-extracted data. The AI ​​model training function uses a deep learning algorithm to build a model that matches the user's specific data. The input is the feature-extracted data, and the output is the trained AI model.

[0163] Step 8:

[0164] Conversation simulation: The server uses a trained AI model to simulate a conversation with the user. The conversation simulation function receives the user's utterances and generates appropriate responses. The input is the user's utterance prompt, and the output is the generated response.

[0165] Step 9:

[0166] Displaying and responding to conversation: The device displays the response received from the server to the user. The response receiving function analyzes the received data and presents it to the user in text or audio format. The input is the response data from the server, and the output is the content displayed to the user.

[0167] Step 10:

[0168] Conversation log storage: The device stores conversation logs between the user and virtual family members. The conversation log storage function records the conversation content in a database so that the user can review the conversation later. The input is the conversation data, and the output is the saved conversation log.

[0169] The above is the specific processing flow of the program of this system.

[0170] (Application example 1)

[0171] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0172] In modern society, many people suffer from loneliness, isolation, and worries, and are seeking ways to achieve psychological stability. There is also a growing need for virtual assistance with shopping-related concerns and product selection. However, effective technology to solve these problems is not yet available. In particular, there is no system that combines AI technology that can hold natural conversations based on user information with a virtual assistant that provides product information and purchasing advice.

[0173] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0174] In this invention, the server includes means for receiving information about family, friends, and pets input by the user, means for analyzing the received information and classifying the personalities and vocal characteristics of the family, friends, and pets, means for training an AI model based on the analyzed data, means for conducting a conversation simulation with the user using the trained AI model, means for setting an API endpoint accessible from the user's terminal, means for the user to converse with the virtual assistant and receive product information and purchasing consultation support, and means for the virtual assistant to recommend appropriate products based on the received information. This allows the user to not only receive psychological support but also virtually receive shopping-related consultation and product selection support.

[0175] "User" refers to an individual or group that uses the system.

[0176] "Inputted Information" refers to data about the personalities and vocal characteristics of family members, friends, pets, etc. that a user provides to the system.

[0177] "Means for receiving" refers to the technical means by which the system receives information entered by the user.

[0178] "Means for analyzing" refers to technical means for analyzing received information and extracting specific patterns or attributes.

[0179] "Means for classifying personality and voice characteristics" refers to technical means for categorizing personality and voice characteristics based on the analysis results.

[0180] "Training means" refers to the technical means by which an AI model is trained using classified data.

[0181] "Means for conducting conversational simulation" means technical means for conducting a conversation with a user using a trained AI model.

[0182] "Means for configuring API endpoints" means the technical means for defining API endpoints for accessing a trained AI model.

[0183] A "virtual assistant" is a virtual support character that uses AI technology to interact with users and provide various types of assistance.

[0184] "Providing product information" refers to the act of a virtual assistant presenting information about a specific product to a user.

[0185] "Means for assisting with purchasing consultation" are technological means by which a virtual assistant can provide advice to a user when selecting a product.

[0186] "Means for making product recommendations" refers to the technical means by which the virtual assistant recommends appropriate products based on the user's input information.

[0187] "Means for storing conversation logs" means technical means for storing the history of conversations between a user and a virtual assistant.

[0188] "Server" is a remote data processing system that processes information entered by the user and assists in the operation of the virtual assistant.

[0189] This invention is a system that uses AI technology to interact with virtual family and virtual assistants. This system trains an AI model based on information about family, friends, and pets entered by the user, and provides natural conversations to the user. Furthermore, this system also has a shopping support function in a virtual store.

[0190] Server-side configuration

[0191] The server has the following features:

[0192] 1. Information reception function: Receives information about family, friends, and pets sent from the user's device. This information includes names, personalities, voice characteristics, relationships, etc.

[0193] 2. Information analysis function: Analyzes received data and classifies personality and voice characteristics. Specifically, it uses Python and TensorFlow to preprocess the data and then analyzes it using natural language processing technology.

[0194] 3. Feature extraction: Extracting important features from the analyzed data. This process uses machine learning algorithms.

[0195] 4. AI model training function: Trains an AI model based on the extracted features. Model training is performed using Python and TensorFlow.

[0196] 5. Conversation Simulation: Simulate conversations with users using a trained AI model. Build a RESTful API using Flask and generate responses based on user requests.

[0197] 6. API endpoint configuration function: Set the trained AI model as an API endpoint accessible from the device. Deploy the API using AWS or GCP.

[0198] Terminal configuration

[0199] The terminal has the following features:

[0200] 1. Information input screen: Provides a screen where users can enter information about family, friends, and pets. Build a cross-platform application using React Native.

[0201] 2. Data transmission function: Sends the input information to the server. Axios is used as the HTTP client library.

[0202] 3. Response reception function: A function that receives responses from the server. It displays them on the UI using React Native.

[0203] 4. Content display function: Displays the received response to the user and, if necessary, provides a voice response using speech synthesis. For speech synthesis, Google's Text-to-Speech API is used.

[0204] 5. Conversation log storage function: A function to save conversation logs between users and virtual assistants. Saved in cloud storage using Firebase.

[0205] Specific examples

[0206] As a concrete example, we present a scenario in which a user converses with a virtual assistant to shop.

[0207] The user opens the app and selects a "friend" character. For example, the "friend"'s personality can be set to "friendly," the voice characteristics to "mid-tone," and the relationship to "likes shopping." If the user says, "I want a new smartphone," the virtual friend will ask, "What features do you want?" If the user answers, "I want one with a good camera," the virtual friend will suggest a specific product, saying, "This model has an excellent camera."

[0208] Prompt Sentence Examples

[0209] The user says, "I want a new smartphone." The virtual assistant asks, "What features do you want?" If the user replies, "I want one with a good camera," the virtual assistant suggests a specific product, saying, "This model has an excellent camera."

[0210] In this way, we provide a system that allows users to receive not only psychological support but also shopping-related advice and product selection through a virtual assistant. This system is expected to alleviate loneliness and isolation in modern society and improve the shopping experience.

[0211] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0212] Step 1:

[0213] The user starts the application and enters the required data into the information input screen.

[0214] Specifically, you enter information in text format, such as the names, personalities, vocal characteristics, and relationships of family members, friends, and pets.

[0215] The input data is temporarily saved on the device. The user checks whether the input is correct and then presses the send button.

[0216] Step 2:

[0217] The terminal receives information entered by the user and sends it to the server.

[0218] Specifically, input data is sent as an HTTP request using Axios.

[0219] The input data is sent to the server, and the server completes receiving the data.

[0220] Step 3:

[0221] The server parses the received data.

[0222] Specifically, using Python and TensorFlow, text data is analyzed using natural language processing technology to classify names, personalities, voice characteristics, and relationships.

[0223] The classified data is temporarily stored in an internal database, and the analysis results are extracted as features such as personality type, voice characteristics, and relationships.

[0224] Step 4:

[0225] The server trains the AI ​​model based on the analyzed data.

[0226] Specifically, features are input into the model using TensorFlow, and the model parameters are updated.

[0227] The input is feature data, and the output is a trained AI model.

[0228] Step 5:

[0229] The server uses a trained AI model to simulate a conversation with the user.

[0230] Specifically, user requests are sent to a RESTful API built using Flask, and the AI ​​model generates a response.

[0231] The input is the user's question or comment, and the output is the generated response message.

[0232] Step 6:

[0233] The server sets the API endpoint to be accessible from the device.

[0234] Specifically, the trained AI model is placed on an API endpoint deployed using AWS or GCP.

[0235] The input is an AI model and API configuration information, and the output is an accessible API endpoint.

[0236] Step 7:

[0237] The terminal receives the response from the server and displays the content.

[0238] Specifically, the response data received as a response to an HTTP request is displayed on the screen using React Native.

[0239] It also uses a speech synthesis function to output responses in voice. The input is response data from the API, and the output is displayed on the screen and output as voice.

[0240] Step 8:

[0241] The device stores a log of the conversation between the user and the virtual assistant.

[0242] Specifically, the content of the conversation is saved in real time to cloud storage such as Firebase.

[0243] The input is the text data of the conversation, and the output is the saved conversation log.

[0244] In this way, the entire system processes each step in order, enabling natural conversations and shopping consultations with users.

[0245] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0246] This invention is a system that uses AI technology to enable individuals who are feeling lonely, isolated, or worried to have everyday conversations and seek advice from a virtual family member, in order to achieve psychological stability.The system incorporates an emotion engine that recognizes the user's emotions, providing a more natural and empathetic conversation experience.

[0247] System Configuration

[0248] Server-side configuration

[0249] The server has the following features:

[0250] Data reception function: Receives information about family, friends, and pets sent from the user's device.

[0251] Data analysis function: Analyzes received information and classifies personality and voice characteristics.

[0252] Feature extraction function: Extracts important features from the analyzed data.

[0253] AI model training function: Trains an AI model based on feature-extracted data.

[0254] Conversation simulation function: Conducts conversation simulations with users using a trained AI model.

[0255] Emotion analysis function: Analyzes emotions from the user's voice, text tone, facial expressions, etc.

[0256] Conversation adjustment function: Adjusts the content and tone of conversations based on analyzed emotional data.

[0257] API endpoint setting function: Set the trained AI model and emotion engine to an API endpoint accessible from the device.

[0258] Terminal configuration

[0259] The terminal has the following features:

[0260] Information input screen: Displays a screen where users can enter information about family, friends, and pets.

[0261] Data transmission function: Sends the entered information to the server.

[0262] Emotion data collection function: Collects user voice, text, and facial expression data.

[0263] Response reception function: Receives a response from the server.

[0264] Content display function: Displays the received response to the user or provides an audio response using speech synthesis.

[0265] Conversation log saving function: Saves conversation logs between the user and their virtual family.

[0266] User Actions

[0267] 1. Log in

[0268] The user launches the application on their device and enters their authentication information on the login screen. The first time they log in, they are required to create a new account.

[0269] 2. Enter information

[0270] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[0271] 3. Start a conversation

[0272] After the information is sent to the server, the user can initiate everyday conversations such as "How was your day?"

[0273] Specific examples

[0274] Example 1: Everyday conversation

[0275] The user launches the application and enters information about their mother. For example, the "mother"'s personality could be "kind," her voice characteristic "high-pitched," and her relationship with the virtual mother "understanding." If the user talks about a difficult time at work, the virtual mother responds, "Really? What happened?" If the user continues, "Your boss scolded me," the virtual mother empathizes, saying, "That must have been tough. But you did a great job." The user's emotions are then analyzed, and the AI ​​adjusts the expressions of empathy and encouragement appropriately.

[0276] Example 2: Consultation

[0277] The user inputs information about a friend and sets the "friend's" personality to "energetic," the voice characteristics to "mid-tone," and the relationship to "good at encouraging people." When the user says, "I've been feeling down lately," the virtual friend will respond with an accepting attitude, saying, "What happened? Tell me." When the user says, "A lot of things have been happening...," the emotion engine analyzes the user's emotional state, and based on that information, the virtual friend will dynamically adjust the words of encouragement and respond, saying, "That's tough, but I know you can get through it!"

[0278] Program processing

[0279] Server-side processing

[0280] The server receives and analyzes input information and emotion data from the user. It trains an AI model based on the analysis results and works with the emotion engine to simulate a conversation with the user. The results are returned to the user's device via an API endpoint.

[0281] Terminal side processing

[0282] The device collects input information and emotion data from the user and sends it to the server, receives responses from the server and displays them to the user in text or audio format, and saves the conversation log and sends it to the server for later reuse.

[0283] Through this operational flow and each function, the system provides consistent psychological support to users, realizing a natural and empathetic conversation experience.

[0284] The processing flow will be explained below.

[0285] Program processing flow

[0286] Server-side processing

[0287] Step 1:

[0288] The server receives information about family, friends, and pets entered by the user, including their names, personalities, vocal characteristics, and relationships.

[0289] Step 2:

[0290] The server analyzes the received information and classifies the personality and voice characteristics into specific categories. For example, personality can be classified as "friendly," "strict," or "energetic," and voice characteristics can be classified as "gentle," "mid-range," or "high-pitched."

[0291] Step 3:

[0292] The server extracts important features from each piece of analyzed information, allowing the AI ​​to more accurately recreate the personalities and characteristics of family, friends, and pets.

[0293] Step 4:

[0294] The server trains an AI model based on the feature-extracted data using a machine learning algorithm (e.g., a natural language processing model).

[0295] Step 5:

[0296] The server uses the trained AI model to simulate a conversation with the user, using natural language processing based on the user's input.

[0297] Step 6:

[0298] The server uses an emotion analysis engine to analyze the user's emotions from their voice, text, and facial expression data. For example, it can identify emotional states such as "angry," "sad," or "happy" from their voice.

[0299] Step 7:

[0300] The server adjusts the content and tone of the AI ​​model's response based on the results of the emotion analysis, generating an appropriate response that matches the emotions the user is feeling.

[0301] Step 8:

[0302] The server transmits data including the tailored conversation response to the user terminal.

[0303] Terminal side processing

[0304] Step 1:

[0305] The user launches the application on their device and enters their authentication information on the login screen. The first time they log in, they are required to create a new account.

[0306] Step 2:

[0307] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[0308] Step 3:

[0309] The device sends the entered information to the server using HTTPS, ensuring secure communication.

[0310] Step 4:

[0311] The device collects voice, text, and facial expression data and transmits it to a server in real time. Voice data is collected through the device's microphone, and facial expression data is collected using the device's camera.

[0312] Step 5:

[0313] The device receives a response from the server, which is the content of the conversation (such as text data or audio data) generated by sentiment analysis and the trained AI model.

[0314] Step 6:

[0315] The device displays the received response to the user. In addition to displaying it in text format, it can also respond verbally using speech synthesis. The device uses a speech synthesis library or service (e.g., Google Text-to-Speech).

[0316] Step 7:

[0317] The device will store conversation logs between the user and their virtual family, which will then be sent to a server for analysis and to improve the AI ​​model.

[0318] User Actions

[0319] Step 1:

[0320] The user launches the app and enters their credentials on the login screen. The first time they log in, they are required to create a new account.

[0321] Step 2:

[0322] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[0323] Step 3:

[0324] The user initiates a conversation, for example, "How was your day?" and their virtual family member responds, "You did a great job today, did anything special happen?"

[0325] Step 4:

[0326] The user expresses their emotions. For example, if the user says, "I've been feeling down lately," the emotion analysis engine analyzes the user's tone of voice and facial expressions, and the server generates an appropriate response. As a result, the virtual family member responds with empathy, saying, "That's tough. Can you tell me more about what happened? I hope it makes you feel a little better."

[0327] In this way, the system provides psychological support to users at each step, enabling a natural and empathetic conversational experience.

[0328] Example 2

[0329] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0330] In modern society, the number of individuals suffering from loneliness, isolation, and anxiety is increasing, and effective methods to achieve psychological stability are needed. However, existing technologies lack systems that provide customized conversations based on individual emotions and personalities, resulting in users being unable to receive sufficient empathy or support. In particular, there is a lack of systems that combine the functionality of emotion analysis and dynamically adjusting appropriate responses, resulting in an insufficient quality of psychological support for users.

[0331] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving information about family, friends, and pets input by the user, means for analyzing the received information and classifying the personalities and vocal characteristics of the family, friends, and pets, means for training an AI model based on the analyzed data, means for conducting a conversation simulation with the user using the trained AI model, means for collecting the user's voice, text, and facial expression data and analyzing emotions, means for adjusting the content and tone of the conversation based on the analyzed emotional data, and means for setting an API endpoint that allows access from the user terminal. This makes it possible to provide consistent psychological support to the user and realize a natural and empathetic conversation experience.

[0332] "Users" refer to individuals who use the system, and who often experience loneliness, isolation, and worries.

[0333] "Information" refers to data about family, friends, and pets, including names, personalities, vocal characteristics, and relationships.

[0334] "Receiving" refers to the process by which the server retrieves information sent from the user terminal.

[0335] "Analysis" refers to the process of analyzing received information and understanding and classifying its content.

[0336] "Personality" is data that indicates the behavioral characteristics and attitudes of humans and pets, and is used in this system for emotion analysis and conversation simulation.

[0337] "Voice characteristics" are data that indicate the pitch, tone, intonation, etc. of a voice, and are used to reproduce the characteristics of the voices of family members, friends, and pets entered by the user.

[0338] "Classification" is the process of dividing information into multiple categories and finding similarities and differences based on each person's personality and voice characteristics.

[0339] "Training" is the process by which an AI model learns from data to produce more accurate and efficient outputs.

[0340] An "AI model" is a collection of algorithms that use artificial intelligence to make predictions or generate results, and operates based on trained data.

[0341] "Conversational simulation" refers to the process of using a trained AI model to simulate a conversation with a user.

[0342] "Emotion analysis" is a technology that infers emotions from a user's voice, text, facial expressions, etc., and is used by the system to generate more natural and empathetic responses.

[0343] "Tone" refers to the intonation and timbre of a voice, and is an element that determines the atmosphere in which the content of a conversation is conveyed.

[0344] "Tuning" is the process of changing the content and tone of the conversation based on the analyzed data to provide an appropriate and empathetic response to the user.

[0345] An "API endpoint" refers to a specific URL or URI through which a user device and a server communicate, and through which data is sent and received.

[0346] "Terminal" refers to a device used by a user to perform operations, including smartphones and tablets.

[0347] A "conversation log" is data that records the history of conversations between a user and an AI model, and can be referenced and reused later.

[0348] This invention is a system that uses AI technology to enable individuals who are feeling lonely, isolated, or worried to have everyday conversations and seek advice from a virtual family member, in order to achieve psychological stability.The system incorporates an emotion engine that recognizes the user's emotions, providing a more natural and empathetic conversation experience.

[0349] System Configuration

[0350] Server-side configuration

[0351] The server has the following features:

[0352] Data reception function: Receives information about family, friends, and pets sent from the user's device. Specifically, it receives JSON format data via HTTP requests.

[0353] Data analysis function: Analyzes received information and classifies personality and voice characteristics. Uses NLP libraries to perform text analysis and extract personality attributes.

[0354] Feature extraction function: Extracts important features from the analyzed data and stores them in a database, including personality traits, voice features, etc.

[0355] AI model training function: Trains generative AI models based on feature extraction data. Specific software used includes TensorFlow and PyTorch.

[0356] Conversation Simulation: A trained AI model is used to simulate conversations with users, with the virtual family generating prompts such as "How was your day?"

[0357] Emotion analysis function: Analyzes emotions from the user's voice, text tone, facial expressions, etc. It uses a voice recognition engine to identify positive, negative, and other emotions.

[0358] Conversation adjustment function: The content and tone of the conversation are adjusted based on the analyzed emotional data. For example, it may give gentle advice such as "You should take it easy today."

[0359] API endpoint setting function: Sets the trained AI model and emotion engine as an API endpoint accessible from the device. Sets a RESTful API and notifies the user device of the endpoint URL.

[0360] Terminal configuration

[0361] The terminal has the following features:

[0362] Information input screen: A screen is displayed where users can enter information about their family, friends, and pets. A smartphone application is used.

[0363] Data transmission function: Sends information entered by the user to the server. The entered text data is sent to the server via an HTTP POST request.

[0364] Emotion data collection function: Collects user voice, text, and facial expression data using the device's microphone and camera.

[0365] Response reception function: Receives responses from the server and displays them to the user. Retrieves JSON format responses from the server's API, parses them, and displays them in the UI.

[0366] Content display function: Displays the received response to the user. Not only in text format, but also in voice response using speech synthesis. Plays it aloud using a speech synthesis engine.

[0367] Conversation log storage function: Save the conversation logs between the user and the virtual family and send them to the server for later reuse. Save the conversation data in a local database and back it up to the server periodically.

[0368] Specific examples

[0369] Example 1: Everyday conversation

[0370] The user launches the application and enters information about their mother. For example, the "mother"'s personality could be "kind," her voice characteristic "high-pitched," and her relationship with the virtual mother "understanding." If the user talks about a difficult time at work, the virtual mother responds, "Really? What happened?" If the user continues, "Your boss scolded me," the virtual mother empathizes, saying, "That must have been tough. But you did a great job." At this point, the emotion engine analyzes the user's emotions, and the AI ​​adjusts the expressions of empathy and encouragement appropriately.

[0371] Example 2: Consultation

[0372] The user inputs information about a friend and sets the friend's personality to "energetic," the voice characteristics to "mid-tone," and the relationship to "good at encouraging others." When the user says, "I've been feeling down lately," the virtual friend will respond with an accepting attitude, saying, "What happened? Tell me." When the user says, "A lot of things have been happening...," the emotion engine analyzes the user's emotions, and based on that information, the virtual friend will dynamically adjust the words of encouragement and respond, saying, "That's tough, but I know you can get through it!"

[0373] Prompt Sentence Examples

[0374] "How was work today?"

[0375] "What's been going on lately?"

[0376] "Really? What happened?"

[0377] "That must have been tough. But you did a great job."

[0378] Through these functions and specific operation flow, this system can provide consistent psychological support to users and realize a natural and empathetic conversation experience.

[0379] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0380] Step 1: Display the information entry screen

[0381] The device displays a screen where the user can enter information about family, friends, and pets. Specifically, the device UI displays text input fields and various selection options. Through these fields, the user enters information such as name, personality, voice characteristics, and relationships. Input data includes the name "Mother," personality "gentle," and voice characteristics "high-pitched."

[0382] Step 2: Send user information to the server

[0383] The device sends the information entered by the user to the server. Specifically, the input data is converted to JSON format and sent to the server using an HTTP POST request. This request includes information such as name, personality, voice characteristics, and relationships. The data is formatted in a way that is easy for the server to receive.

[0384] Step 3: Receiving data

[0385] The server receives the information sent from the user's device. Specifically, the server receives an HTTP request and extracts information such as name, personality, voice characteristics, and relationships from the JSON-formatted data. The received data includes attributes such as "mother," "gentle," and "high-pitched voice."

[0386] Step 4: Data analysis

[0387] The server analyzes the received information and classifies each individual's personality and voice characteristics. Specifically, it uses an NLP library to perform text analysis and extract personality attributes. It also uses voice recognition technology to analyze voice features. The input data is the name, personality, voice characteristics, and relationships, and the output data is the analyzed personality attributes and voice features.

[0388] Step 5: Feature Extraction

[0389] The server extracts important features from the analyzed data and stores them in a database. Specifically, personality traits and voice features are extracted as important features. For example, a personality trait such as "gentle" or a voice feature such as "high-pitched" are extracted and registered in the database.

[0390] Step 6: Training the AI ​​model

[0391] The server then trains the AI ​​model based on the extracted feature data. Specifically, it uses TensorFlow and PyTorch to train the generative AI model while referring to past conversation data. During this process, the system processes a dataset containing personality traits and voice features to improve its ability to generate natural conversations with the user.

[0392] Step 7: Conversation simulation

[0393] The server uses the trained AI model to simulate a conversation with the user. Specifically, it receives a prompt from the user, such as "How was your day?", and the AI ​​model generates an appropriate response. For example, in response to the question, "You had a hard time at work," it generates the response, "Really? What happened?"

[0394] Step 8: Sentiment Analysis

[0395] The server collects the user's voice, text, and facial expression data and analyzes their emotions. Specifically, it uses a voice recognition engine and facial expression analysis software to identify the user's emotional state. For example, it can estimate emotions such as stress or relief from the user's voice data.

[0396] Step 9: Conversation Adjustment

[0397] The server adjusts the content and tone of the conversation based on the analyzed emotional data. Specifically, if the user is tired, it will generate a softer response based on the emotion analysis results, such as "You should get some rest today." This provides a customized response that is optimized for the user's emotions.

[0398] Step 10: Receive a response

[0399] The device receives the response from the server and displays it to the user. Specifically, it parses the JSON format response from the server and displays it as text on the UI. It also plays back the audio using a speech synthesis engine. For example, it provides the user with a response such as "That must have been tough, but you did well" both as text and audio.

[0400] Step 11: Save conversation log

[0401] The device saves conversation logs between the user and their virtual family and sends them to a server for later reuse. Specifically, each conversation episode is recorded in a local database and periodically backed up to the server. The saved data is used for future conversation simulations and user support.

[0402] (Application example 2)

[0403] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0404] In modern society, the number of individuals feeling lonely or isolated is increasing. While there is a need for methods to provide psychological stability to these individuals, real-life relationships alone are often insufficient. Furthermore, even in virtual shopping experiences, personalized support tailored to individual needs is lacking. Therefore, a system that uses a virtual companion to provide empathetic conversation and advice to individuals who feel lonely or isolated is needed.

[0405] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0406] In this invention, the server includes means for receiving person information input by a user, means for analyzing the received information and classifying the person's personality and voice characteristics, means for training an AI model based on the analyzed data, means for conducting a conversation simulation with the user using the trained AI model, means for setting an API endpoint accessible from a user terminal, and shopping assistance means for a person to provide product advice in a virtual store. This enables individuals who feel lonely or isolated to receive empathetic conversation and personalized shopping support in a virtual environment.

[0407] "User" refers to an individual who uses the system to have conversations with virtual people or receive shopping assistance.

[0408] "Input" refers to information provided by the user to the system, specifically data such as a person's personality and voice characteristics.

[0409] "People" refers to virtual beings such as family, friends, and companions that users recreate within the system.

[0410] "Information receiving means" refers to a function that allows the system to receive person information entered by the user.

[0411] "Analysis means" refers to a function for classifying a person's personality and voice characteristics based on the received information.

[0412] "Training means" refers to the function used to train an AI model based on analyzed data.

[0413] "Conversation simulation means" refers to a function that uses a trained AI model to conduct a conversation between a user and a virtual person.

[0414] "API endpoint setting means" refers to a function for setting an API endpoint that can be accessed from a user terminal.

[0415] "Shopping assistance means" refers to a function that enables a virtual person to provide product advice to a user in a virtual store.

[0416] System configuration

[0417] Server-side processing

[0418] 1. Means of receiving information

[0419] The server receives information about the person sent from the user's device, including their name, personality, voice characteristics, and relationships.

[0420] 2. Analysis method

[0421] The server analyzes the received information and classifies the person's personality and voice characteristics using natural language processing and voice analysis technology.

[0422] 3. Training methods

[0423] The server trains an AI model based on the analyzed data using a deep learning framework (e.g., TensorFlow, PyTorch).

[0424] 4. Conversation Simulation Methods

[0425] The server uses a trained AI model to simulate a conversation with the user, allowing the virtual persona to have a natural conversation.

[0426] 5. API endpoint configuration method

[0427] The server sets up an API endpoint that can be accessed from the user's device, allowing the user to interact with the virtual person at any time.

[0428] 6. Shopping assistance methods

[0429] The server allows a virtual person in the virtual store to provide product advice to the user, based on product information and user reviews.

[0430] Terminal side processing

[0431] 1. Information input screen

[0432] Users enter their personal information on the device screen, which is built using HTML, CSS, and JavaScript.

[0433] 2. Data transmission function

[0434] The terminal sends the input information to the server via a RESTful API.

[0435] 3. Emotion data collection function

[0436] The device uses WebRTC and TensorFlow.js to collect voice and facial expression data and send it to a server.

[0437] 4. Response Receiving Function

[0438] The device receives the response from the server and processes the received data using the JavaScript fetch API or similar.

[0439] 5. Content display function

[0440] The device displays the received response to the user, and can also play the received data aloud using a speech synthesis API (e.g., Google Text-to-Speech API).

[0441] User Actions

[0442] 1. Initial system setup

[0443] Users launch the application and enter their authentication information on the login screen. The first time they log in, they are required to create a new account.

[0444] 2. Enter your information

[0445] Users enter information about virtual characters such as family and friends on a screen where they can enter their names, personality traits, voice characteristics, relationships, etc.

[0446] 3. Start a conversation

[0447] Users can initiate everyday conversations such as "How was your day?" The virtual person will provide natural conversation.

[0448] Specific examples

[0449] During a virtual shopping trip, a user wearing smart glasses can choose new clothes and talk to their virtual family. When the user asks, "What do you think of this outfit?", the virtual mother responds, "It looks great, and the color suits you!"

[0450] Prompt Sentence Examples

[0451] User input: "I've been interested in this outfit lately. What do you think?"

[0452] Generative AI Model Input: The AI ​​model, the user is interested in a casual shirt in a solid blue color. Given that the user's mother has a "kind" personality, generate an empathetic response when the user asks about the shirt.

[0453] The system can provide empathetic conversations and personalized shopping support through a virtual companion to individuals who feel lonely or isolated.

[0454] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0455] Step 1:

[0456] The terminal collects the user's login information on the input screen and sends the authentication information to the server. The input includes the user name and password, and session information is generated as output. The server receives this information, performs authentication, and returns the session information.

[0457] Step 2:

[0458] The user enters information about virtual people, such as family and friends, into the input screen. The input includes name, personality, voice characteristics, and relationships, and is sent as output in JSON format to the server. The device then sends this to the server.

[0459] Step 3:

[0460] The server analyzes the received personal information and classifies personality and voice characteristics using Natural Language Processing (NLP) and voice analysis technology. It processes the JSON data received as input, converts personality and voice characteristics into numerical data, and outputs the classification results.

[0461] Step 4:

[0462] The server trains an AI model based on the classified data. It receives the data classified in the previous step as input, trains the model using a deep learning framework (e.g., TensorFlow, PyTorch), and generates a trained AI model as output.

[0463] Step 5:

[0464] The device uses its emotion data collection function to collect the user's voice and facial expression data. It uses data acquired from the microphone and camera as input, processes it using WebRTC and TensorFlow.js, and outputs analyzed emotion data.

[0465] Step 6:

[0466] The server uses a trained AI model to simulate a conversation between the user and a virtual character. It receives emotional data and the user's questions and comments as input, generates conversation content, and obtains response data as output.

[0467] Step 7:

[0468] The server sends the response data to the user device through the API endpoint, processes the HTTP request via the API using the generated response data as input, and sends the data to the device as output.

[0469] Step 8:

[0470] The response data received by the device is played back to the user as audio using a speech synthesis API. Response data from the server is received as input, converted into audio using a speech synthesis API (e.g., Google Text-to-Speech API), and an audio response is obtained as output.

[0471] Step 9:

[0472] The device saves the conversation log between the user and the virtual person and sends it to the server as reusable data later. It receives the conversation data as input, saves it in LocalStorage, and obtains the saved log data as output.

[0473] Step 10:

[0474] The server generates shopping assistance data for the virtual store, and the virtual person provides product advice to the user. It receives product data and user interests as input, generates advice using an AI model, and sends the advice data to the terminal as output.

[0475] Step 11:

[0476] The terminal presents the received advice data to the user. Using the advice data received from the server as input, the advice is displayed on the screen or via voice, and the advice is provided to the user as output.

[0477] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0478] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0479] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0480] [Second embodiment]

[0481] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0482] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0483] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0484] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0485] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0486] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0487] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0488] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0489] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0490] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0491] In the smart glasses 214, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0492] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0493] This invention is a system that uses AI technology to provide everyday conversations and consultations with virtual family members to help individuals who are feeling lonely, isolated, or worried to find psychological stability. The system is designed to receive input from users, train an AI model based on that information, and engage in natural conversations with the users.

[0494] System Configuration

[0495] Server-side configuration

[0496] The server has the following features:

[0497] Data reception function: Receives information about family, friends, and pets sent from the user's device.

[0498] Data analysis function: Analyzes received information and classifies personality and voice characteristics.

[0499] Feature extraction function: Extracts important features from the analyzed data.

[0500] AI model training function: Trains an AI model based on feature-extracted data.

[0501] Conversation simulation function: Conducts conversation simulations with users using a trained AI model.

[0502] API endpoint setting function: Set a trained AI model to an API endpoint that can be accessed from the device.

[0503] Terminal configuration

[0504] The terminal has the following features:

[0505] Information input screen: Displays a screen where users can enter information about family, friends, and pets.

[0506] Data transmission function: Sends the entered information to the server.

[0507] Response reception function: Receives a response from the server.

[0508] Content display function: Displays the received response to the user or provides an audio response using speech synthesis.

[0509] Conversation log saving function: Saves conversation logs between the user and their virtual family.

[0510] User Actions

[0511] 1. Log in

[0512] The user launches the application on their device and enters their authentication information on the login screen. The first time they log in, they are required to create a new account.

[0513] 2. Enter information

[0514] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[0515] 3. Start a conversation

[0516] After the information is sent to the server, the user can initiate everyday conversations such as "How was your day?"

[0517] Specific examples

[0518] Example 1: Everyday conversation

[0519] The user launches the application and enters information about their mother. For example, the "mother"'s personality can be "kind," her voice characteristic can be "high-pitched," and her relationship status can be "understanding." If the user says, "I had a hard time at work today," the virtual mother will respond, "Really? What happened?" If the user continues, "My boss scolded me," the virtual mother will empathize, saying, "That must have been tough. But you did a great job."

[0520] Example 2: Consultation

[0521] The user inputs information about a friend, and for example, sets the friend's personality to "energetic," the voice characteristics to "mid-tone," and the relationship to "good at encouraging others." If the user says, "I've been feeling down lately," the virtual friend will show an accepting attitude by saying, "What happened? Tell me." If the user says, "A lot of things have been happening at once...," the virtual friend will encourage them by saying, "That's tough, but I know you can get through it!"

[0522] In this way, the present invention makes full use of AI technology to provide users with an experience that makes them feel as if they are conversing with real family, friends, or pets, providing psychological support.

[0523] The processing flow will be explained below.

[0524] Program processing flow

[0525] Server-side processing

[0526] Step 1:

[0527] The server receives information about family, friends, and pets entered by the user, including their names, personalities, vocal characteristics, and relationships.

[0528] Step 2:

[0529] The server analyzes the received information and classifies the personality and voice characteristics into specific categories. For example, personality can be classified as "friendly," "strict," or "energetic," while voice characteristics can be classified as "gentle," "mid-range," or "high-pitched."

[0530] Step 3:

[0531] The server extracts important features from each piece of analyzed information, allowing the AI ​​to more accurately recreate the personalities and characteristics of family, friends, and pets.

[0532] Step 4:

[0533] The server trains an AI model based on the feature-extracted data using a machine learning algorithm (e.g., a natural language processing model).

[0534] Step 5:

[0535] The server uses the trained AI model to simulate a conversation with the user, using natural language processing based on the user's input.

[0536] Step 6:

[0537] The server deploys the trained AI model to an API endpoint, making it accessible from user devices.

[0538] Terminal side processing

[0539] Step 1:

[0540] The user launches the application on their device and enters their authentication information on the login screen. The first time they log in, they are required to create a new account.

[0541] Step 2:

[0542] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[0543] Step 3:

[0544] The device sends the entered information to the server using HTTPS, ensuring secure communication.

[0545] Step 4:

[0546] The device receives a response from the server. The response is the content of the conversation generated by the AI ​​model (text data, voice data, etc.).

[0547] Step 5:

[0548] The device displays the received response to the user. In addition to displaying it in text format, it can also respond in voice using speech synthesis. The device uses a speech synthesis library or service (e.g., Google Text-to-Speech).

[0549] Step 6:

[0550] The device will store conversation logs between the user and their virtual family, which will then be sent to a server for analysis and to improve the AI ​​model.

[0551] User Actions

[0552] Step 1:

[0553] The user launches the app and enters their credentials on the login screen. The first time they log in, they are required to create a new account.

[0554] Step 2:

[0555] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[0556] Step 3:

[0557] The user initiates a conversation, for example, "How was your day?", and the virtual family member responds, "You did a great job today. Did anything special happen?"

[0558] Step 4:

[0559] For example, if a user says, "Work has been tough lately," their virtual family member will empathize and respond, "That must be tough. Tell me more about what happened. I hope it makes you feel a little better."

[0560] In this way, the system can provide psychological support to the user through each step.

[0561] Example 1

[0562] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0563] In modern society, many individuals suffer from loneliness, isolation, and worries, and there is a need for a means to receive psychological support. However, due to the limited time and opportunity to consult or talk with family and friends, there is a lack of effective systems to provide such psychological support. Another issue is the lack of systems that can provide dialogue tailored to the individual needs of users.

[0564] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0565] In this invention, the server includes means for receiving information about family, friends, and pets input by the user, means for analyzing the received information and classifying the personalities and vocal characteristics of the family, friends, and pets, means for training an AI model based on the analyzed and feature-extracted data, means for conducting a conversation simulation with the user using the trained AI model, means for setting an API endpoint that enables access from the user terminal, means for the user to input authentication information into a login screen for authentication, means for displaying the information input by the user as text or voice or responding using voice synthesis, and means for saving a conversation log between the user and the virtual family. This enables the user to receive psychological support through natural conversations with their virtual family and friends.

[0566] A "user" is an individual who uses the system to enter information about family, friends, and pets and engage in virtual interactions.

[0567] A "server" is a computer system that receives information sent by users and performs analysis, feature extraction, AI model training, conversation simulation, etc.

[0568] A "terminal" is a device through which a user inputs information and engages in virtual interactions via communication with a server. Examples include smartphones and tablets.

[0569] The "data receiving means" is a function that allows the server to receive information about family, friends, and pets sent by the user.

[0570] "Data analysis means" is a function for analyzing received information and classifying the personalities and vocal characteristics of family, friends, and pets.

[0571] "Feature extraction means" is a function that extracts important features from the analyzed data and uses them to train the AI ​​model.

[0572] "AI model training means" is a function for training an artificial intelligence model based on feature-extracted data.

[0573] "Conversation simulation means" is a function for simulating a virtual conversation between a user and the AI ​​model using a trained AI model.

[0574] The "API endpoint setting means" is a function for setting a trained AI model as an API endpoint that can be accessed from a user terminal.

[0575] The "login authentication means" is a function in which a user inputs authentication information into a login screen and the server verifies that information.

[0576] "Information display means" is a function for displaying information entered by the user as text or voice, or for responding using voice synthesis.

[0577] The "conversation log storage means" is a function for storing conversation logs between the user and the virtual family and transmitting them to the server as reusable data later.

[0578] "Virtual family" refers to family, friends, and pets recreated by an AI model trained based on information entered by the user.

[0579] This invention is a system that uses artificial intelligence (AI) technology to provide everyday conversations and consultations with a virtual family, helping individuals experiencing loneliness, isolation, or anxiety to find psychological stability. The system is designed to receive user input, train an AI model based on that information, and have natural conversations with the user.

[0580] System Configuration

[0581] Server-side configuration

[0582] The server has the following features:

[0583] Data reception function: Receives information about family, friends, and pets sent by the user from the device. Specifically, it receives data such as names, personalities, voice characteristics, and relationships entered by the user.

[0584] Data analysis function: Analyzes the received information and classifies the personalities and vocal characteristics of family, friends, and pets. For example, it analyzes personality and vocal characteristics such as "gentle" or "high-pitched."

[0585] Feature extraction function: Extracts important features from the analyzed data, thereby extracting the elements necessary for training the AI ​​model.

[0586] AI model training function: The AI ​​model is trained based on the feature-extracted data, for example, using a deep learning algorithm to replicate the personality and voice of specific family or friends selected by the user.

[0587] Conversation simulation function: A trained AI model is used to simulate a conversation with the user, generating appropriate responses to prompts entered by the user.

[0588] API endpoint setting function: Set the trained AI model to an API endpoint accessible from the device, allowing users to easily access it at any time.

[0589] Login authentication function: A function that allows users to enter authentication information on the login screen and perform authentication.

[0590] Terminal configuration

[0591] The terminal has the following features:

[0592] Information entry screen: Displays a screen where users can enter information about family, friends, and pets, including details such as their names, personalities, vocal characteristics, and relationships.

[0593] Data transmission function: Sends the entered information to the server. For example, it sends data to the server using an HTTP POST request.

[0594] Response reception function: Receives responses from the server, analyzes the received responses, and displays them in an easy-to-understand manner for the user.

[0595] Content display function: A function that displays the received response to the user, and can display text or output audio using voice synthesis.

[0596] Conversation log saving function: The function to save the conversation log between the user and the virtual family, which can be used to review the conversation later.

[0597] Usage example

[0598] Example 1: Everyday conversation

[0599] The user launches the application and enters information about their mother. For example, the "mother"'s personality can be "kind," her voice characteristic can be "high-pitched," and her relationship status can be "understanding." If the user says, "I had a hard time at work today," the virtual mother will respond, "Really? What happened?" If the user continues, "My boss scolded me," the virtual mother will empathize, saying, "That must have been tough. But you did a great job."

[0600] Example 2: Consultation

[0601] The user enters information about a friend, setting the friend's personality to "energetic," voice characteristics to "mid-tone," and relationship to "good at encouraging others." If the user says, "I've been feeling down lately," the virtual friend will respond with an accepting attitude, "What happened? Tell me." If the user says, "A lot of things have been happening at once...," the virtual friend will encourage them, saying, "That's tough, but I know you can get through it!"

[0602] With the above configuration, the system provides users with an experience that makes them feel as if they are having a conversation with their real family, friends, or pets, and can provide psychological support.

[0603] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0604] Step 1:

[0605] User login: The user launches the application and enters authentication information on the login screen. When the user enters their email address and password and presses the "Login" button, the authentication information is sent from the device to the server. The server compares this with the database, and if authentication is successful, the user is taken to the home screen. The input is the user's authentication information, and the output is the authentication result.

[0606] Step 2:

[0607] Information input: An information input screen is displayed for the user to enter information about family, friends, and pets. Here, information such as "name," "personality," "voice characteristics," and "relationship" are entered in the input fields. When the user confirms the input and presses the "send" button, the data is sent from the device to the server. The input is the information about family members, etc. entered by the user, and the output is the sent data.

[0608] Step 3:

[0609] Data transmission: The device sends the information entered by the user to the server. Specifically, the device's data transmission function converts the input data into JSON format and sends it to the server as an HTTP POST request. The input is the user's input data, and the output is the data sent to the server.

[0610] Step 4:

[0611] Data reception: The server receives the data sent from the terminal. The data reception function analyzes the HTTP request, extracts the user data, and stores it in the database. The input is the data in the HTTP request format, and the output is the stored data.

[0612] Step 5:

[0613] Data analysis: The server analyzes the received data. The data analysis function analyzes information such as name, personality, vocal characteristics, and relationships, and performs classification. For example, characteristics such as "kind," "high-pitched voice," and "understanding" are analyzed. The input is the saved user data, and the output is the analyzed classified data.

[0614] Step 6:

[0615] Feature Extraction: The server extracts important features from the analyzed data. The feature extraction function quantifies the user data and generates parameters for use in training the AI ​​model. The input is the analyzed classified data, and the output is the feature-extracted data.

[0616] Step 7:

[0617] AI model training: The server trains the AI ​​model based on the feature-extracted data. The AI ​​model training function uses a deep learning algorithm to build a model that matches the user's specific data. The input is the feature-extracted data, and the output is the trained AI model.

[0618] Step 8:

[0619] Conversation simulation: The server uses a trained AI model to simulate a conversation with the user. The conversation simulation function receives the user's utterances and generates appropriate responses. The input is the user's utterance prompt, and the output is the generated response.

[0620] Step 9:

[0621] Displaying and responding to conversation: The device displays the response received from the server to the user. The response receiving function analyzes the received data and presents it to the user in text or audio format. The input is the response data from the server, and the output is the content displayed to the user.

[0622] Step 10:

[0623] Conversation log storage: The device stores conversation logs between the user and virtual family members. The conversation log storage function records the conversation content in a database so that the user can review the conversation later. The input is the conversation data, and the output is the saved conversation log.

[0624] The above is the specific processing flow of the program of this system.

[0625] (Application example 1)

[0626] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0627] In modern society, many people suffer from loneliness, isolation, and worries, and are seeking ways to achieve psychological stability. There is also a growing need for virtual assistance with shopping-related concerns and product selection. However, effective technology to solve these problems is not yet available. In particular, there is no system that combines AI technology that can hold natural conversations based on user information with a virtual assistant that provides product information and purchasing advice.

[0628] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0629] In this invention, the server includes means for receiving information about family, friends, and pets input by the user, means for analyzing the received information and classifying the personalities and vocal characteristics of the family, friends, and pets, means for training an AI model based on the analyzed data, means for conducting a conversation simulation with the user using the trained AI model, means for setting an API endpoint accessible from the user's terminal, means for the user to converse with the virtual assistant and receive product information and purchasing consultation support, and means for the virtual assistant to recommend appropriate products based on the received information. This allows the user to not only receive psychological support but also virtually receive shopping-related consultation and product selection support.

[0630] "User" refers to an individual or group that uses the system.

[0631] "Inputted Information" refers to data about the personalities and vocal characteristics of family members, friends, pets, etc. that a user provides to the system.

[0632] "Means for receiving" refers to the technical means by which the system receives information entered by the user.

[0633] "Means for analyzing" refers to technical means for analyzing received information and extracting specific patterns or attributes.

[0634] "Means for classifying personality and voice characteristics" refers to technical means for categorizing personality and voice characteristics based on the analysis results.

[0635] "Training means" refers to the technical means by which an AI model is trained using classified data.

[0636] "Means for conducting conversational simulation" means technical means for conducting a conversation with a user using a trained AI model.

[0637] "Means for configuring API endpoints" means the technical means for defining API endpoints for accessing a trained AI model.

[0638] A "virtual assistant" is a virtual support character that uses AI technology to interact with users and provide various types of assistance.

[0639] "Providing product information" refers to the act of a virtual assistant presenting information about a specific product to a user.

[0640] "Means for assisting with purchasing consultation" are technological means by which a virtual assistant can provide advice to a user when selecting a product.

[0641] "Means for making product recommendations" refers to the technical means by which the virtual assistant recommends appropriate products based on the user's input information.

[0642] "Means for storing conversation logs" means technical means for storing the history of conversations between a user and a virtual assistant.

[0643] "Server" is a remote data processing system that processes information entered by the user and assists in the operation of the virtual assistant.

[0644] This invention is a system that uses AI technology to interact with virtual family and virtual assistants. This system trains an AI model based on information about family, friends, and pets entered by the user, and provides natural conversations to the user. Furthermore, this system also has a shopping support function in a virtual store.

[0645] Server-side configuration

[0646] The server has the following features:

[0647] 1. Information reception function: Receives information about family, friends, and pets sent from the user's device. This information includes names, personalities, voice characteristics, relationships, etc.

[0648] 2. Information analysis function: Analyzes received data and classifies personality and voice characteristics. Specifically, it uses Python and TensorFlow to preprocess the data and then analyzes it using natural language processing technology.

[0649] 3. Feature extraction: Extracting important features from the analyzed data. This process uses machine learning algorithms.

[0650] 4. AI model training function: Trains an AI model based on the extracted features. Model training is performed using Python and TensorFlow.

[0651] 5. Conversation Simulation: Simulate conversations with users using a trained AI model. Build a RESTful API using Flask and generate responses based on user requests.

[0652] 6. API endpoint configuration function: Set the trained AI model as an API endpoint accessible from the device. Deploy the API using AWS or GCP.

[0653] Terminal configuration

[0654] The terminal has the following features:

[0655] 1. Information input screen: Provides a screen where users can enter information about family, friends, and pets. Build a cross-platform application using React Native.

[0656] 2. Data transmission function: Sends the input information to the server. Axios is used as the HTTP client library.

[0657] 3. Response reception function: A function that receives responses from the server. It displays them on the UI using React Native.

[0658] 4. Content display function: Displays the received response to the user and, if necessary, provides a voice response using speech synthesis. For speech synthesis, Google's Text-to-Speech API is used.

[0659] 5. Conversation log storage function: A function to save conversation logs between users and virtual assistants. Saved in cloud storage using Firebase.

[0660] Specific examples

[0661] As a concrete example, we present a scenario in which a user converses with a virtual assistant to shop.

[0662] The user opens the app and selects a "friend" character. For example, the "friend"'s personality can be set to "friendly," the voice characteristics to "mid-tone," and the relationship to "likes shopping." If the user says, "I want a new smartphone," the virtual friend will ask, "What features do you want?" If the user answers, "I want one with a good camera," the virtual friend will suggest a specific product, saying, "This model has an excellent camera."

[0663] Prompt Sentence Examples

[0664] The user says, "I want a new smartphone." The virtual assistant asks, "What features do you want?" If the user replies, "I want one with a good camera," the virtual assistant suggests a specific product, saying, "This model has an excellent camera."

[0665] In this way, we provide a system that allows users to receive not only psychological support but also shopping-related advice and product selection through a virtual assistant. This system is expected to alleviate loneliness and isolation in modern society and improve the shopping experience.

[0666] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0667] Step 1:

[0668] The user starts the application and enters the required data into the information input screen.

[0669] Specifically, you enter information in text format, such as the names, personalities, vocal characteristics, and relationships of family members, friends, and pets.

[0670] The input data is temporarily saved on the device. The user checks whether the input is correct and then presses the send button.

[0671] Step 2:

[0672] The terminal receives information entered by the user and sends it to the server.

[0673] Specifically, input data is sent as an HTTP request using Axios.

[0674] The input data is sent to the server, and the server completes receiving the data.

[0675] Step 3:

[0676] The server parses the received data.

[0677] Specifically, using Python and TensorFlow, text data is analyzed using natural language processing technology to classify names, personalities, voice characteristics, and relationships.

[0678] The classified data is temporarily stored in an internal database, and the analysis results are extracted as features such as personality type, voice characteristics, and relationships.

[0679] Step 4:

[0680] The server trains the AI ​​model based on the analyzed data.

[0681] Specifically, features are input into the model using TensorFlow, and the model parameters are updated.

[0682] The input is feature data, and the output is a trained AI model.

[0683] Step 5:

[0684] The server uses a trained AI model to simulate a conversation with the user.

[0685] Specifically, user requests are sent to a RESTful API built using Flask, and the AI ​​model generates a response.

[0686] The input is the user's question or comment, and the output is the generated response message.

[0687] Step 6:

[0688] The server sets the API endpoint to be accessible from the device.

[0689] Specifically, the trained AI model is placed on an API endpoint deployed using AWS or GCP.

[0690] The input is an AI model and API configuration information, and the output is an accessible API endpoint.

[0691] Step 7:

[0692] The terminal receives the response from the server and displays the content.

[0693] Specifically, the response data received as a response to an HTTP request is displayed on the screen using React Native.

[0694] It also uses a speech synthesis function to output responses in voice. The input is response data from the API, and the output is displayed on the screen and output as voice.

[0695] Step 8:

[0696] The device stores a log of the conversation between the user and the virtual assistant.

[0697] Specifically, the content of the conversation is saved in real time to cloud storage such as Firebase.

[0698] The input is the text data of the conversation, and the output is the saved conversation log.

[0699] In this way, the entire system processes each step in order, enabling natural conversations and shopping consultations with users.

[0700] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0701] This invention is a system that uses AI technology to enable individuals who are feeling lonely, isolated, or worried to have everyday conversations and seek advice from a virtual family member, in order to achieve psychological stability.The system incorporates an emotion engine that recognizes the user's emotions, providing a more natural and empathetic conversation experience.

[0702] System Configuration

[0703] Server-side configuration

[0704] The server has the following features:

[0705] Data reception function: Receives information about family, friends, and pets sent from the user's device.

[0706] Data analysis function: Analyzes received information and classifies personality and voice characteristics.

[0707] Feature extraction function: Extracts important features from the analyzed data.

[0708] AI model training function: Trains an AI model based on feature-extracted data.

[0709] Conversation simulation function: Conducts conversation simulations with users using a trained AI model.

[0710] Emotion analysis function: Analyzes emotions from the user's voice, text tone, facial expressions, etc.

[0711] Conversation adjustment function: Adjusts the content and tone of conversations based on analyzed emotional data.

[0712] API endpoint setting function: Set the trained AI model and emotion engine to an API endpoint accessible from the device.

[0713] Terminal configuration

[0714] The terminal has the following features:

[0715] Information input screen: Displays a screen where users can enter information about family, friends, and pets.

[0716] Data transmission function: Sends the entered information to the server.

[0717] Emotion data collection function: Collects user voice, text, and facial expression data.

[0718] Response reception function: Receives a response from the server.

[0719] Content display function: Displays the received response to the user or provides an audio response using speech synthesis.

[0720] Conversation log saving function: Saves conversation logs between the user and their virtual family.

[0721] User Actions

[0722] 1. Log in

[0723] The user launches the application on their device and enters their authentication information on the login screen. The first time they log in, they are required to create a new account.

[0724] 2. Enter information

[0725] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[0726] 3. Start a conversation

[0727] After the information is sent to the server, the user can initiate everyday conversations such as "How was your day?"

[0728] Specific examples

[0729] Example 1: Everyday conversation

[0730] The user launches the application and enters information about their mother. For example, the "mother"'s personality could be "kind," her voice characteristic "high-pitched," and her relationship with the virtual mother "understanding." If the user talks about a difficult time at work, the virtual mother responds, "Really? What happened?" If the user continues, "Your boss scolded me," the virtual mother empathizes, saying, "That must have been tough. But you did a great job." The user's emotions are then analyzed, and the AI ​​adjusts the expressions of empathy and encouragement appropriately.

[0731] Example 2: Consultation

[0732] The user inputs information about a friend and sets the "friend's" personality to "energetic," the voice characteristics to "mid-tone," and the relationship to "good at encouraging people." When the user says, "I've been feeling down lately," the virtual friend will respond with an accepting attitude, saying, "What happened? Tell me." When the user says, "A lot of things have been happening...," the emotion engine analyzes the user's emotional state, and based on that information, the virtual friend will dynamically adjust the words of encouragement and respond, saying, "That's tough, but I know you can get through it!"

[0733] Program processing

[0734] Server-side processing

[0735] The server receives and analyzes input information and emotion data from the user. It trains an AI model based on the analysis results and works with the emotion engine to simulate a conversation with the user. The results are returned to the user's device via an API endpoint.

[0736] Terminal side processing

[0737] The device collects input information and emotion data from the user and sends it to the server, receives responses from the server and displays them to the user in text or audio format, and saves the conversation log and sends it to the server for later reuse.

[0738] Through this operational flow and each function, the system provides consistent psychological support to users, realizing a natural and empathetic conversation experience.

[0739] The processing flow will be explained below.

[0740] Program processing flow

[0741] Server-side processing

[0742] Step 1:

[0743] The server receives information about family, friends, and pets entered by the user, including their names, personalities, vocal characteristics, and relationships.

[0744] Step 2:

[0745] The server analyzes the received information and classifies the personality and voice characteristics into specific categories. For example, personality can be classified as "friendly," "strict," or "energetic," and voice characteristics can be classified as "gentle," "mid-range," or "high-pitched."

[0746] Step 3:

[0747] The server extracts important features from each piece of analyzed information, allowing the AI ​​to more accurately recreate the personalities and characteristics of family, friends, and pets.

[0748] Step 4:

[0749] The server trains an AI model based on the feature-extracted data using a machine learning algorithm (e.g., a natural language processing model).

[0750] Step 5:

[0751] The server uses the trained AI model to simulate a conversation with the user, using natural language processing based on the user's input.

[0752] Step 6:

[0753] The server uses an emotion analysis engine to analyze the user's emotions from their voice, text, and facial expression data. For example, it can identify emotional states such as "angry," "sad," or "happy" from their voice.

[0754] Step 7:

[0755] The server adjusts the content and tone of the AI ​​model's response based on the results of the emotion analysis, generating an appropriate response that matches the emotions the user is feeling.

[0756] Step 8:

[0757] The server transmits data including the tailored conversation response to the user terminal.

[0758] Terminal side processing

[0759] Step 1:

[0760] The user launches the application on their device and enters their authentication information on the login screen. The first time they log in, they are required to create a new account.

[0761] Step 2:

[0762] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[0763] Step 3:

[0764] The device sends the entered information to the server using HTTPS, ensuring secure communication.

[0765] Step 4:

[0766] The device collects voice, text, and facial expression data and transmits it to a server in real time. Voice data is collected through the device's microphone, and facial expression data is collected using the device's camera.

[0767] Step 5:

[0768] The device receives a response from the server, which is the content of the conversation (such as text data or audio data) generated by sentiment analysis and the trained AI model.

[0769] Step 6:

[0770] The device displays the received response to the user. In addition to displaying it in text format, it can also respond verbally using speech synthesis. The device uses a speech synthesis library or service (e.g., Google Text-to-Speech).

[0771] Step 7:

[0772] The device will store conversation logs between the user and their virtual family, which will then be sent to a server for analysis and to improve the AI ​​model.

[0773] User Actions

[0774] Step 1:

[0775] The user launches the app and enters their credentials on the login screen. The first time they log in, they are required to create a new account.

[0776] Step 2:

[0777] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[0778] Step 3:

[0779] The user initiates a conversation, for example, "How was your day?" and their virtual family member responds, "You did a great job today, did anything special happen?"

[0780] Step 4:

[0781] The user expresses their emotions. For example, if the user says, "I've been feeling down lately," the emotion analysis engine analyzes the user's tone of voice and facial expressions, and the server generates an appropriate response. As a result, the virtual family member responds with empathy, saying, "That's tough. Can you tell me more about what happened? I hope it makes you feel a little better."

[0782] In this way, the system provides psychological support to users at each step, enabling a natural and empathetic conversational experience.

[0783] Example 2

[0784] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0785] In modern society, the number of individuals suffering from loneliness, isolation, and anxiety is increasing, and effective methods to achieve psychological stability are needed. However, existing technologies lack systems that provide customized conversations based on individual emotions and personalities, resulting in users being unable to receive sufficient empathy or support. In particular, there is a lack of systems that combine the functionality of emotion analysis and dynamically adjusting appropriate responses, resulting in an insufficient quality of psychological support for users.

[0786] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving information about family, friends, and pets input by the user, means for analyzing the received information and classifying the personalities and vocal characteristics of the family, friends, and pets, means for training an AI model based on the analyzed data, means for conducting a conversation simulation with the user using the trained AI model, means for collecting the user's voice, text, and facial expression data and analyzing emotions, means for adjusting the content and tone of the conversation based on the analyzed emotional data, and means for setting an API endpoint that allows access from the user terminal. This makes it possible to provide consistent psychological support to the user and realize a natural and empathetic conversation experience.

[0787] "Users" refer to individuals who use the system, and who often experience loneliness, isolation, and worries.

[0788] "Information" refers to data about family, friends, and pets, including names, personalities, vocal characteristics, and relationships.

[0789] "Receiving" refers to the process by which the server retrieves information sent from the user terminal.

[0790] "Analysis" refers to the process of analyzing received information and understanding and classifying its content.

[0791] "Personality" is data that indicates the behavioral characteristics and attitudes of humans and pets, and is used in this system for emotion analysis and conversation simulation.

[0792] "Voice characteristics" are data that indicate the pitch, tone, intonation, etc. of a voice, and are used to reproduce the characteristics of the voices of family members, friends, and pets entered by the user.

[0793] "Classification" is the process of dividing information into multiple categories and finding similarities and differences based on each person's personality and voice characteristics.

[0794] "Training" is the process by which an AI model learns from data to produce more accurate and efficient outputs.

[0795] An "AI model" is a collection of algorithms that use artificial intelligence to make predictions or generate results, and operates based on trained data.

[0796] "Conversational simulation" refers to the process of using a trained AI model to simulate a conversation with a user.

[0797] "Emotion analysis" is a technology that infers emotions from a user's voice, text, facial expressions, etc., and is used by the system to generate more natural and empathetic responses.

[0798] "Tone" refers to the intonation and timbre of a voice, and is an element that determines the atmosphere in which the content of a conversation is conveyed.

[0799] "Tuning" is the process of changing the content and tone of the conversation based on the analyzed data to provide an appropriate and empathetic response to the user.

[0800] An "API endpoint" refers to a specific URL or URI through which a user device and a server communicate, and through which data is sent and received.

[0801] "Terminal" refers to a device used by a user to perform operations, including smartphones and tablets.

[0802] A "conversation log" is data that records the history of conversations between a user and an AI model, and can be referenced and reused later.

[0803] This invention is a system that uses AI technology to enable individuals who are feeling lonely, isolated, or worried to have everyday conversations and seek advice from a virtual family member, in order to achieve psychological stability.The system incorporates an emotion engine that recognizes the user's emotions, providing a more natural and empathetic conversation experience.

[0804] System Configuration

[0805] Server-side configuration

[0806] The server has the following features:

[0807] Data reception function: Receives information about family, friends, and pets sent from the user's device. Specifically, it receives JSON format data via HTTP requests.

[0808] Data analysis function: Analyzes received information and classifies personality and voice characteristics. Uses NLP libraries to perform text analysis and extract personality attributes.

[0809] Feature extraction function: Extracts important features from the analyzed data and stores them in a database, including personality traits, voice features, etc.

[0810] AI model training function: Trains generative AI models based on feature extraction data. Specific software used includes TensorFlow and PyTorch.

[0811] Conversation Simulation: A trained AI model is used to simulate conversations with users, with the virtual family generating prompts such as "How was your day?"

[0812] Emotion analysis function: Analyzes emotions from the user's voice, text tone, facial expressions, etc. It uses a voice recognition engine to identify positive, negative, and other emotions.

[0813] Conversation adjustment function: The content and tone of the conversation are adjusted based on the analyzed emotional data. For example, it may give gentle advice such as "You should take it easy today."

[0814] API endpoint setting function: Sets the trained AI model and emotion engine as an API endpoint accessible from the device. Sets a RESTful API and notifies the user device of the endpoint URL.

[0815] Terminal configuration

[0816] The terminal has the following features:

[0817] Information input screen: A screen is displayed where users can enter information about their family, friends, and pets. A smartphone application is used.

[0818] Data transmission function: Sends information entered by the user to the server. The entered text data is sent to the server via an HTTP POST request.

[0819] Emotion data collection function: Collects user voice, text, and facial expression data using the device's microphone and camera.

[0820] Response reception function: Receives responses from the server and displays them to the user. Retrieves JSON format responses from the server's API, parses them, and displays them in the UI.

[0821] Content display function: Displays the received response to the user. Not only in text format, but also in voice response using speech synthesis. Plays it aloud using a speech synthesis engine.

[0822] Conversation log storage function: Save the conversation logs between the user and the virtual family and send them to the server for later reuse. Save the conversation data in a local database and back it up to the server periodically.

[0823] Specific examples

[0824] Example 1: Everyday conversation

[0825] The user launches the application and enters information about their mother. For example, the "mother"'s personality could be "kind," her voice characteristic "high-pitched," and her relationship with the virtual mother "understanding." If the user talks about a difficult time at work, the virtual mother responds, "Really? What happened?" If the user continues, "Your boss scolded me," the virtual mother empathizes, saying, "That must have been tough. But you did a great job." At this point, the emotion engine analyzes the user's emotions, and the AI ​​adjusts the expressions of empathy and encouragement appropriately.

[0826] Example 2: Consultation

[0827] The user inputs information about a friend and sets the friend's personality to "energetic," the voice characteristics to "mid-tone," and the relationship to "good at encouraging others." When the user says, "I've been feeling down lately," the virtual friend will respond with an accepting attitude, saying, "What happened? Tell me." When the user says, "A lot of things have been happening...," the emotion engine analyzes the user's emotions, and based on that information, the virtual friend will dynamically adjust the words of encouragement and respond, saying, "That's tough, but I know you can get through it!"

[0828] Prompt Sentence Examples

[0829] "How was work today?"

[0830] "What's been going on lately?"

[0831] "Really? What happened?"

[0832] "That must have been tough. But you did a great job."

[0833] Through these functions and specific operation flow, this system can provide consistent psychological support to users and realize a natural and empathetic conversation experience.

[0834] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0835] Step 1: Display the information entry screen

[0836] The device displays a screen where the user can enter information about family, friends, and pets. Specifically, the device UI displays text input fields and various selection options. Through these fields, the user enters information such as name, personality, voice characteristics, and relationships. Input data includes the name "Mother," personality "gentle," and voice characteristics "high-pitched."

[0837] Step 2: Send user information to the server

[0838] The device sends the information entered by the user to the server. Specifically, the input data is converted to JSON format and sent to the server using an HTTP POST request. This request includes information such as name, personality, voice characteristics, and relationships. The data is formatted in a way that is easy for the server to receive.

[0839] Step 3: Receiving data

[0840] The server receives the information sent from the user's device. Specifically, the server receives an HTTP request and extracts information such as name, personality, voice characteristics, and relationships from the JSON-formatted data. The received data includes attributes such as "mother," "gentle," and "high-pitched voice."

[0841] Step 4: Data analysis

[0842] The server analyzes the received information and classifies each individual's personality and voice characteristics. Specifically, it uses an NLP library to perform text analysis and extract personality attributes. It also uses voice recognition technology to analyze voice features. The input data is the name, personality, voice characteristics, and relationships, and the output data is the analyzed personality attributes and voice features.

[0843] Step 5: Feature Extraction

[0844] The server extracts important features from the analyzed data and stores them in a database. Specifically, personality traits and voice features are extracted as important features. For example, a personality trait such as "gentle" or a voice feature such as "high-pitched" are extracted and registered in the database.

[0845] Step 6: Training the AI ​​model

[0846] The server then trains the AI ​​model based on the extracted feature data. Specifically, it uses TensorFlow and PyTorch to train the generative AI model while referring to past conversation data. During this process, the system processes a dataset containing personality traits and voice features to improve its ability to generate natural conversations with the user.

[0847] Step 7: Conversation simulation

[0848] The server uses the trained AI model to simulate a conversation with the user. Specifically, it receives a prompt from the user, such as "How was your day?", and the AI ​​model generates an appropriate response. For example, in response to the question, "You had a hard time at work," it generates the response, "Really? What happened?"

[0849] Step 8: Sentiment Analysis

[0850] The server collects the user's voice, text, and facial expression data and analyzes their emotions. Specifically, it uses a voice recognition engine and facial expression analysis software to identify the user's emotional state. For example, it can estimate emotions such as stress or relief from the user's voice data.

[0851] Step 9: Conversation Adjustment

[0852] The server adjusts the content and tone of the conversation based on the analyzed emotional data. Specifically, if the user is tired, it will generate a softer response based on the emotion analysis results, such as "You should get some rest today." This provides a customized response that is optimized for the user's emotions.

[0853] Step 10: Receive a response

[0854] The device receives the response from the server and displays it to the user. Specifically, it parses the JSON format response from the server and displays it as text on the UI. It also plays back the audio using a speech synthesis engine. For example, it provides the user with a response such as "That must have been tough, but you did well" both as text and audio.

[0855] Step 11: Save conversation log

[0856] The device saves conversation logs between the user and their virtual family and sends them to a server for later reuse. Specifically, each conversation episode is recorded in a local database and periodically backed up to the server. The saved data is used for future conversation simulations and user support.

[0857] (Application example 2)

[0858] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0859] In modern society, the number of individuals feeling lonely or isolated is increasing. While there is a need for methods to provide psychological stability to these individuals, real-life relationships alone are often insufficient. Furthermore, even in virtual shopping experiences, personalized support tailored to individual needs is lacking. Therefore, a system that uses a virtual companion to provide empathetic conversation and advice to individuals who feel lonely or isolated is needed.

[0860] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0861] In this invention, the server includes means for receiving person information input by a user, means for analyzing the received information and classifying the person's personality and voice characteristics, means for training an AI model based on the analyzed data, means for conducting a conversation simulation with the user using the trained AI model, means for setting an API endpoint accessible from a user terminal, and shopping assistance means for a person to provide product advice in a virtual store. This enables individuals who feel lonely or isolated to receive empathetic conversation and personalized shopping support in a virtual environment.

[0862] "User" refers to an individual who uses the system to have conversations with virtual people or receive shopping assistance.

[0863] "Input" refers to information provided by the user to the system, specifically data such as a person's personality and voice characteristics.

[0864] "People" refers to virtual beings such as family, friends, and companions that users recreate within the system.

[0865] "Information receiving means" refers to a function that allows the system to receive person information entered by the user.

[0866] "Analysis means" refers to a function for classifying a person's personality and voice characteristics based on the received information.

[0867] "Training means" refers to the function used to train an AI model based on analyzed data.

[0868] "Conversation simulation means" refers to a function that uses a trained AI model to conduct a conversation between a user and a virtual person.

[0869] "API endpoint setting means" refers to a function for setting an API endpoint that can be accessed from a user terminal.

[0870] "Shopping assistance means" refers to a function that enables a virtual person to provide product advice to a user in a virtual store.

[0871] System configuration

[0872] Server-side processing

[0873] 1. Means of receiving information

[0874] The server receives information about the person sent from the user's device, including their name, personality, voice characteristics, and relationships.

[0875] 2. Analysis method

[0876] The server analyzes the received information and classifies the person's personality and voice characteristics using natural language processing and voice analysis technology.

[0877] 3. Training methods

[0878] The server trains an AI model based on the analyzed data using a deep learning framework (e.g., TensorFlow, PyTorch).

[0879] 4. Conversation Simulation Methods

[0880] The server uses a trained AI model to simulate a conversation with the user, allowing the virtual persona to have a natural conversation.

[0881] 5. API endpoint configuration method

[0882] The server sets up an API endpoint that can be accessed from the user's device, allowing the user to interact with the virtual person at any time.

[0883] 6. Shopping assistance methods

[0884] The server allows a virtual person in the virtual store to provide product advice to the user, based on product information and user reviews.

[0885] Terminal side processing

[0886] 1. Information input screen

[0887] Users enter their personal information on the device screen, which is built using HTML, CSS, and JavaScript.

[0888] 2. Data transmission function

[0889] The terminal sends the input information to the server via a RESTful API.

[0890] 3. Emotion data collection function

[0891] The device uses WebRTC and TensorFlow.js to collect voice and facial expression data and send it to a server.

[0892] 4. Response Receiving Function

[0893] The device receives the response from the server and processes the received data using the JavaScript fetch API or similar.

[0894] 5. Content display function

[0895] The device displays the received response to the user, and can also play the received data aloud using a speech synthesis API (e.g., Google Text-to-Speech API).

[0896] User Actions

[0897] 1. Initial system setup

[0898] Users launch the application and enter their authentication information on the login screen. The first time they log in, they are required to create a new account.

[0899] 2. Enter your information

[0900] Users enter information about virtual characters such as family and friends on a screen where they can enter their names, personality traits, voice characteristics, relationships, etc.

[0901] 3. Start a conversation

[0902] Users can initiate everyday conversations such as "How was your day?" The virtual person will provide natural conversation.

[0903] Specific examples

[0904] During a virtual shopping trip, a user wearing smart glasses can choose new clothes and talk to their virtual family. When the user asks, "What do you think of this outfit?", the virtual mother responds, "It looks great, and the color suits you!"

[0905] Prompt Sentence Examples

[0906] User input: "I've been interested in this outfit lately. What do you think?"

[0907] Generative AI Model Input: The AI ​​model, the user is interested in a casual shirt in a solid blue color. Given that the user's mother has a "kind" personality, generate an empathetic response when the user asks about the shirt.

[0908] The system can provide empathetic conversations and personalized shopping support through a virtual companion to individuals who feel lonely or isolated.

[0909] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0910] Step 1:

[0911] The terminal collects the user's login information on the input screen and sends the authentication information to the server. The input includes the user name and password, and session information is generated as output. The server receives this information, performs authentication, and returns the session information.

[0912] Step 2:

[0913] The user enters information about virtual people, such as family and friends, into the input screen. The input includes name, personality, voice characteristics, and relationships, and is sent as output in JSON format to the server. The device then sends this to the server.

[0914] Step 3:

[0915] The server analyzes the received personal information and classifies personality and voice characteristics using Natural Language Processing (NLP) and voice analysis technology. It processes the JSON data received as input, converts personality and voice characteristics into numerical data, and outputs the classification results.

[0916] Step 4:

[0917] The server trains an AI model based on the classified data. It receives the data classified in the previous step as input, trains the model using a deep learning framework (e.g., TensorFlow, PyTorch), and generates a trained AI model as output.

[0918] Step 5:

[0919] The device uses its emotion data collection function to collect the user's voice and facial expression data. It uses data acquired from the microphone and camera as input, processes it using WebRTC and TensorFlow.js, and outputs analyzed emotion data.

[0920] Step 6:

[0921] The server uses a trained AI model to simulate a conversation between the user and a virtual character. It receives emotional data and the user's questions and comments as input, generates conversation content, and obtains response data as output.

[0922] Step 7:

[0923] The server sends the response data to the user device through the API endpoint, processes the HTTP request via the API using the generated response data as input, and sends the data to the device as output.

[0924] Step 8:

[0925] The response data received by the device is played back to the user as audio using a speech synthesis API. Response data from the server is received as input, converted into audio using a speech synthesis API (e.g., Google Text-to-Speech API), and an audio response is obtained as output.

[0926] Step 9:

[0927] The device saves the conversation log between the user and the virtual person and sends it to the server as reusable data later. It receives the conversation data as input, saves it in LocalStorage, and obtains the saved log data as output.

[0928] Step 10:

[0929] The server generates shopping assistance data for the virtual store, and the virtual person provides product advice to the user. It receives product data and user interests as input, generates advice using an AI model, and sends the advice data to the terminal as output.

[0930] Step 11:

[0931] The terminal presents the received advice data to the user. Using the advice data received from the server as input, the advice is displayed on the screen or via voice, and the advice is provided to the user as output.

[0932] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0933] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0934] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0935] [Third embodiment]

[0936] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0937] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0938] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0939] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0940] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0941] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0942] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0943] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0944] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0945] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0946] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0947] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0948] This invention is a system that uses AI technology to provide everyday conversations and consultations with virtual family members to help individuals who are feeling lonely, isolated, or worried to find psychological stability. The system is designed to receive input from users, train an AI model based on that information, and engage in natural conversations with the users.

[0949] System Configuration

[0950] Server-side configuration

[0951] The server has the following features:

[0952] Data reception function: Receives information about family, friends, and pets sent from the user's device.

[0953] Data analysis function: Analyzes received information and classifies personality and voice characteristics.

[0954] Feature extraction function: Extracts important features from the analyzed data.

[0955] AI model training function: Trains an AI model based on feature-extracted data.

[0956] Conversation simulation function: Conducts conversation simulations with users using a trained AI model.

[0957] API endpoint setting function: Set a trained AI model to an API endpoint that can be accessed from the device.

[0958] Terminal configuration

[0959] The terminal has the following features:

[0960] Information input screen: Displays a screen where users can enter information about family, friends, and pets.

[0961] Data transmission function: Sends the entered information to the server.

[0962] Response reception function: Receives a response from the server.

[0963] Content display function: Displays the received response to the user or provides an audio response using speech synthesis.

[0964] Conversation log saving function: Saves conversation logs between the user and their virtual family.

[0965] User Actions

[0966] 1. Log in

[0967] The user launches the application on their device and enters their authentication information on the login screen. The first time they log in, they are required to create a new account.

[0968] 2. Enter information

[0969] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[0970] 3. Start a conversation

[0971] After the information is sent to the server, the user can initiate everyday conversations such as "How was your day?"

[0972] Specific examples

[0973] Example 1: Everyday conversation

[0974] The user launches the application and enters information about their mother. For example, the "mother"'s personality can be "kind," her voice characteristic can be "high-pitched," and her relationship status can be "understanding." If the user says, "I had a hard time at work today," the virtual mother will respond, "Really? What happened?" If the user continues, "My boss scolded me," the virtual mother will empathize, saying, "That must have been tough. But you did a great job."

[0975] Example 2: Consultation

[0976] The user inputs information about a friend, and for example, sets the friend's personality to "energetic," the voice characteristics to "mid-tone," and the relationship to "good at encouraging others." If the user says, "I've been feeling down lately," the virtual friend will show an accepting attitude by saying, "What happened? Tell me." If the user says, "A lot of things have been happening at once...," the virtual friend will encourage them by saying, "That's tough, but I know you can get through it!"

[0977] In this way, the present invention makes full use of AI technology to provide users with an experience that makes them feel as if they are conversing with real family, friends, or pets, providing psychological support.

[0978] The processing flow will be explained below.

[0979] Program processing flow

[0980] Server-side processing

[0981] Step 1:

[0982] The server receives information about family, friends, and pets entered by the user, including their names, personalities, vocal characteristics, and relationships.

[0983] Step 2:

[0984] The server analyzes the received information and classifies the personality and voice characteristics into specific categories. For example, personality can be classified as "friendly," "strict," or "energetic," while voice characteristics can be classified as "gentle," "mid-range," or "high-pitched."

[0985] Step 3:

[0986] The server extracts important features from each piece of analyzed information, allowing the AI ​​to more accurately recreate the personalities and characteristics of family, friends, and pets.

[0987] Step 4:

[0988] The server trains an AI model based on the feature-extracted data using machine learning algorithms (e.g., natural language processing models).

[0989] Step 5:

[0990] The server uses the trained AI model to simulate a conversation with the user, using natural language processing based on the user's input.

[0991] Step 6:

[0992] The server deploys the trained AI model to an API endpoint, making it accessible from user devices.

[0993] Terminal side processing

[0994] Step 1:

[0995] The user launches the application on their device and enters their authentication information on the login screen. The first time they log in, they are required to create a new account.

[0996] Step 2:

[0997] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[0998] Step 3:

[0999] The device sends the entered information to the server using HTTPS, ensuring secure communication.

[1000] Step 4:

[1001] The device receives a response from the server. The response is the content of the conversation generated by the AI ​​model (text data, voice data, etc.).

[1002] Step 5:

[1003] The device displays the received response to the user. In addition to displaying it in text format, it can also respond in voice using speech synthesis. The device uses a speech synthesis library or service (e.g., Google Text-to-Speech).

[1004] Step 6:

[1005] The device will store conversation logs between the user and their virtual family, which will then be sent to a server for analysis and to improve the AI ​​model.

[1006] User Actions

[1007] Step 1:

[1008] The user launches the app and enters their credentials on the login screen. The first time they log in, they are required to create a new account.

[1009] Step 2:

[1010] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[1011] Step 3:

[1012] The user initiates a conversation, for example, "How was your day?", and the virtual family member responds, "You did a great job today. Did anything special happen?"

[1013] Step 4:

[1014] For example, if a user says, "Work has been tough lately," their virtual family member will empathize and respond, "That must be tough. Tell me more about what happened. I hope it makes you feel a little better."

[1015] In this way, the system can provide psychological support to the user through each step.

[1016] Example 1

[1017] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1018] In modern society, many individuals suffer from loneliness, isolation, and worries, and there is a need for a means to receive psychological support. However, due to the limited time and opportunity to consult or talk with family and friends, there is a lack of effective systems to provide such psychological support. Another issue is the lack of systems that can provide dialogue tailored to the individual needs of users.

[1019] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1020] In this invention, the server includes means for receiving information about family, friends, and pets input by the user, means for analyzing the received information and classifying the personalities and vocal characteristics of the family, friends, and pets, means for training an AI model based on the analyzed and feature-extracted data, means for conducting a conversation simulation with the user using the trained AI model, means for setting an API endpoint that enables access from the user terminal, means for the user to input authentication information into a login screen for authentication, means for displaying the information input by the user as text or voice or responding using voice synthesis, and means for saving a conversation log between the user and the virtual family. This enables the user to receive psychological support through natural conversations with their virtual family and friends.

[1021] A "user" is an individual who uses the system to enter information about family, friends, and pets and engage in virtual interactions.

[1022] A "server" is a computer system that receives information sent by users and performs analysis, feature extraction, AI model training, conversation simulation, etc.

[1023] A "terminal" is a device through which a user inputs information and engages in virtual interactions via communication with a server. Examples include smartphones and tablets.

[1024] The "data receiving means" is a function that allows the server to receive information about family, friends, and pets sent by the user.

[1025] "Data analysis means" is a function for analyzing received information and classifying the personalities and vocal characteristics of family, friends, and pets.

[1026] "Feature extraction means" is a function that extracts important features from the analyzed data and uses them to train the AI ​​model.

[1027] "AI model training means" is a function for training an artificial intelligence model based on feature-extracted data.

[1028] "Conversation simulation means" is a function for simulating a virtual conversation between a user and the AI ​​model using a trained AI model.

[1029] The "API endpoint setting means" is a function for setting a trained AI model as an API endpoint that can be accessed from a user terminal.

[1030] The "login authentication means" is a function in which a user inputs authentication information into a login screen and the server verifies that information.

[1031] "Information display means" is a function for displaying information entered by the user as text or voice, or for responding using voice synthesis.

[1032] The "conversation log storage means" is a function for storing conversation logs between the user and the virtual family and transmitting them to the server as reusable data later.

[1033] "Virtual family" refers to family, friends, and pets recreated by an AI model trained based on information entered by the user.

[1034] This invention is a system that uses artificial intelligence (AI) technology to provide everyday conversations and consultations with a virtual family, helping individuals experiencing loneliness, isolation, or anxiety to find psychological stability. The system is designed to receive user input, train an AI model based on that information, and have natural conversations with the user.

[1035] System Configuration

[1036] Server-side configuration

[1037] The server has the following features:

[1038] Data reception function: Receives information about family, friends, and pets sent by the user from the device. Specifically, it receives data such as names, personalities, voice characteristics, and relationships entered by the user.

[1039] Data analysis function: Analyzes the received information and classifies the personalities and vocal characteristics of family, friends, and pets. For example, it analyzes personality and vocal characteristics such as "gentle" or "high-pitched."

[1040] Feature extraction function: Extracts important features from the analyzed data, thereby extracting the elements necessary for training the AI ​​model.

[1041] AI model training function: The AI ​​model is trained based on the feature-extracted data, for example, using a deep learning algorithm to replicate the personality and voice of specific family or friends selected by the user.

[1042] Conversation simulation function: A trained AI model is used to simulate a conversation with the user, generating appropriate responses to prompts entered by the user.

[1043] API endpoint setting function: Set the trained AI model to an API endpoint accessible from the device, allowing users to easily access it at any time.

[1044] Login authentication function: A function that allows users to enter authentication information on the login screen and perform authentication.

[1045] Terminal configuration

[1046] The terminal has the following features:

[1047] Information entry screen: Displays a screen where users can enter information about family, friends, and pets, including details such as their names, personalities, vocal characteristics, and relationships.

[1048] Data transmission function: Sends the input information to the server. For example, it sends data to the server using an HTTP POST request.

[1049] Response reception function: Receives responses from the server, analyzes the received responses, and displays them in an easy-to-understand manner for the user.

[1050] Content display function: A function that displays the received response to the user, and can display text or output audio using voice synthesis.

[1051] Conversation log saving function: The function to save the conversation log between the user and the virtual family, which can be used to review the conversation later.

[1052] Usage example

[1053] Example 1: Everyday conversation

[1054] The user launches the application and enters information about their mother. For example, the "mother"'s personality can be "kind," her voice characteristic can be "high-pitched," and her relationship status can be "understanding." If the user says, "I had a hard time at work today," the virtual mother will respond, "Really? What happened?" If the user continues, "My boss scolded me," the virtual mother will empathize, saying, "That must have been tough. But you did a great job."

[1055] Example 2: Consultation

[1056] The user enters information about a friend, setting the friend's personality to "energetic," voice characteristics to "mid-tone," and relationship to "good at encouraging others." If the user says, "I've been feeling down lately," the virtual friend will respond with an accepting attitude, "What happened? Tell me." If the user says, "A lot of things have been happening at once...," the virtual friend will encourage them, saying, "That's tough, but I know you can get through it!"

[1057] With the above configuration, the system provides users with an experience that makes them feel as if they are having a conversation with their real family, friends, or pets, and can provide psychological support.

[1058] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1059] Step 1:

[1060] User login: The user launches the application and enters authentication information on the login screen. When the user enters their email address and password and presses the "Login" button, the authentication information is sent from the device to the server. The server compares this with the database, and if authentication is successful, the user is taken to the home screen. The input is the user's authentication information, and the output is the authentication result.

[1061] Step 2:

[1062] Information input: An information input screen is displayed for the user to enter information about family, friends, and pets. Here, information such as "name," "personality," "voice characteristics," and "relationship" are entered in the input fields. When the user confirms the input and presses the "send" button, the data is sent from the device to the server. The input is the information about family members, etc. entered by the user, and the output is the sent data.

[1063] Step 3:

[1064] Data transmission: The device sends the information entered by the user to the server. Specifically, the device's data transmission function converts the input data into JSON format and sends it to the server as an HTTP POST request. The input is the user's input data, and the output is the data sent to the server.

[1065] Step 4:

[1066] Data reception: The server receives the data sent from the terminal. The data reception function analyzes the HTTP request, extracts the user data, and stores it in the database. The input is the data in the HTTP request format, and the output is the stored data.

[1067] Step 5:

[1068] Data analysis: The server analyzes the received data. The data analysis function analyzes information such as name, personality, vocal characteristics, and relationships, and performs classification. For example, characteristics such as "kind," "high-pitched voice," and "understanding" are analyzed. The input is the saved user data, and the output is the analyzed classified data.

[1069] Step 6:

[1070] Feature Extraction: The server extracts important features from the analyzed data. The feature extraction function quantifies the user data and generates parameters for use in training the AI ​​model. The input is the analyzed classified data, and the output is the feature-extracted data.

[1071] Step 7:

[1072] AI model training: The server trains the AI ​​model based on the feature-extracted data. The AI ​​model training function uses a deep learning algorithm to build a model that matches the user's specific data. The input is the feature-extracted data, and the output is the trained AI model.

[1073] Step 8:

[1074] Conversation simulation: The server uses a trained AI model to simulate a conversation with the user. The conversation simulation function receives the user's utterances and generates appropriate responses. The input is the user's utterance prompt, and the output is the generated response.

[1075] Step 9:

[1076] Displaying and responding to conversation: The device displays the response received from the server to the user. The response receiving function analyzes the received data and presents it to the user in text or audio format. The input is the response data from the server, and the output is the content displayed to the user.

[1077] Step 10:

[1078] Conversation log storage: The device stores conversation logs between the user and virtual family members. The conversation log storage function records the content of the conversation in a database so that the user can review the conversation later. The input is the conversation data, and the output is the saved conversation log.

[1079] The above is the specific processing flow of the program of this system.

[1080] (Application example 1)

[1081] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1082] In modern society, many people suffer from loneliness, isolation, and worries, and are seeking ways to achieve psychological stability. There is also a growing need for virtual assistance with shopping-related concerns and product selection. However, effective technology to solve these problems is not yet available. In particular, there is no system that combines AI technology that can hold natural conversations based on user information with a virtual assistant that provides product information and purchasing advice.

[1083] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1084] In this invention, the server includes means for receiving information about family, friends, and pets input by the user, means for analyzing the received information and classifying the personalities and vocal characteristics of the family, friends, and pets, means for training an AI model based on the analyzed data, means for conducting a conversation simulation with the user using the trained AI model, means for setting an API endpoint accessible from the user's terminal, means for the user to converse with the virtual assistant and receive product information and purchasing consultation support, and means for the virtual assistant to recommend appropriate products based on the received information. This allows the user to not only receive psychological support but also virtually receive shopping-related consultation and product selection support.

[1085] "User" refers to an individual or group that uses the system.

[1086] "Inputted Information" refers to data about the personalities and vocal characteristics of family members, friends, pets, etc. that a user provides to the system.

[1087] "Means for receiving" refers to the technical means by which the system receives information entered by the user.

[1088] "Means for analyzing" refers to technical means for analyzing received information and extracting specific patterns or attributes.

[1089] "Means for classifying personality and voice characteristics" refers to technical means for categorizing personality and voice characteristics based on the analysis results.

[1090] "Training means" refers to the technical means by which an AI model is trained using classified data.

[1091] "Means for conducting conversational simulation" means technical means for conducting a conversation with a user using a trained AI model.

[1092] "Means for configuring API endpoints" means the technical means for defining API endpoints for accessing a trained AI model.

[1093] A "virtual assistant" is a virtual support character that uses AI technology to interact with users and provide various types of assistance.

[1094] "Providing product information" refers to the act of a virtual assistant presenting information about a specific product to a user.

[1095] "Means for assisting with purchasing consultation" are technological means by which a virtual assistant can provide advice to a user when selecting a product.

[1096] "Means for making product recommendations" refers to the technical means by which the virtual assistant recommends appropriate products based on the user's input information.

[1097] "Means for storing conversation logs" means technical means for storing the history of conversations between a user and a virtual assistant.

[1098] "Server" is a remote data processing system that processes information entered by the user and assists in the operation of the virtual assistant.

[1099] This invention is a system that uses AI technology to interact with virtual family and virtual assistants. This system trains an AI model based on information about family, friends, and pets entered by the user, and provides natural conversations to the user. Furthermore, this system also has a shopping support function in a virtual store.

[1100] Server-side configuration

[1101] The server has the following features:

[1102] 1. Information reception function: Receives information about family, friends, and pets sent from the user's device. This information includes names, personalities, voice characteristics, relationships, etc.

[1103] 2. Information analysis function: Analyzes received data and classifies personality and voice characteristics. Specifically, it uses Python and TensorFlow to preprocess the data and then analyzes it using natural language processing technology.

[1104] 3. Feature extraction: Extracting important features from the analyzed data. This process uses machine learning algorithms.

[1105] 4. AI model training function: Trains an AI model based on the extracted features. Model training is performed using Python and TensorFlow.

[1106] 5. Conversation Simulation: Simulate conversations with users using a trained AI model. Build a RESTful API using Flask and generate responses based on user requests.

[1107] 6. API endpoint configuration function: Set the trained AI model as an API endpoint accessible from the device. Deploy the API using AWS or GCP.

[1108] Terminal configuration

[1109] The terminal has the following features:

[1110] 1. Information input screen: Provides a screen where users can enter information about family, friends, and pets. Build a cross-platform application using React Native.

[1111] 2. Data transmission function: Sends the input information to the server. Axios is used as the HTTP client library.

[1112] 3. Response reception function: A function that receives responses from the server. It displays them on the UI using React Native.

[1113] 4. Content display function: Displays the received response to the user and, if necessary, provides a voice response using speech synthesis. For speech synthesis, Google's Text-to-Speech API is used.

[1114] 5. Conversation log storage function: A function to save conversation logs between users and virtual assistants. Saved in cloud storage using Firebase.

[1115] Specific examples

[1116] As a concrete example, we present a scenario in which a user converses with a virtual assistant to shop.

[1117] The user opens the app and selects a "friend" character. For example, the "friend"'s personality can be set to "friendly," the voice characteristics to "mid-tone," and the relationship to "likes shopping." If the user says, "I want a new smartphone," the virtual friend will ask, "What features do you want?" If the user answers, "I want one with a good camera," the virtual friend will suggest a specific product, saying, "This model has an excellent camera."

[1118] Prompt Sentence Examples

[1119] The user says, "I want a new smartphone." The virtual assistant asks, "What features do you want?" If the user replies, "I want one with a good camera," the virtual assistant suggests a specific product, saying, "This model has an excellent camera."

[1120] In this way, we provide a system that allows users to receive not only psychological support but also shopping-related advice and product selection through a virtual assistant. This system is expected to alleviate loneliness and isolation in modern society and improve the shopping experience.

[1121] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1122] Step 1:

[1123] The user starts the application and enters the required data into the information input screen.

[1124] Specifically, you enter information in text format, such as the names, personalities, vocal characteristics, and relationships of family members, friends, and pets.

[1125] The input data is temporarily saved on the device. The user checks whether the input is correct and then presses the send button.

[1126] Step 2:

[1127] The terminal receives information entered by the user and sends it to the server.

[1128] Specifically, input data is sent as an HTTP request using Axios.

[1129] The input data is sent to the server, and the server completes receiving the data.

[1130] Step 3:

[1131] The server parses the received data.

[1132] Specifically, using Python and TensorFlow, text data is analyzed using natural language processing technology to classify names, personalities, voice characteristics, and relationships.

[1133] The classified data is temporarily stored in an internal database, and the analysis results are extracted as features such as personality type, voice characteristics, and relationships.

[1134] Step 4:

[1135] The server trains the AI ​​model based on the analyzed data.

[1136] Specifically, features are input into the model using TensorFlow, and the model parameters are updated.

[1137] The input is feature data, and the output is a trained AI model.

[1138] Step 5:

[1139] The server uses a trained AI model to simulate a conversation with the user.

[1140] Specifically, user requests are sent to a RESTful API built using Flask, and the AI ​​model generates a response.

[1141] The input is the user's question or comment, and the output is the generated response message.

[1142] Step 6:

[1143] The server sets the API endpoint to be accessible from the device.

[1144] Specifically, the trained AI model is placed on an API endpoint deployed using AWS or GCP.

[1145] The input is an AI model and API configuration information, and the output is an accessible API endpoint.

[1146] Step 7:

[1147] The terminal receives the response from the server and displays the content.

[1148] Specifically, the response data received as a response to an HTTP request is displayed on the screen using React Native.

[1149] It also uses a speech synthesis function to output responses in voice. The input is response data from the API, and the output is displayed on the screen and output as voice.

[1150] Step 8:

[1151] The device stores a log of the conversation between the user and the virtual assistant.

[1152] Specifically, the content of the conversation is saved in real time to cloud storage such as Firebase.

[1153] The input is the text data of the conversation, and the output is the saved conversation log.

[1154] In this way, the entire system processes each step in order, enabling natural conversations and shopping consultations with users.

[1155] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1156] This invention is a system that uses AI technology to enable individuals who are feeling lonely, isolated, or worried to have everyday conversations and seek advice from a virtual family member, in order to achieve psychological stability.The system incorporates an emotion engine that recognizes the user's emotions, providing a more natural and empathetic conversation experience.

[1157] System Configuration

[1158] Server-side configuration

[1159] The server has the following features:

[1160] Data reception function: Receives information about family, friends, and pets sent from the user's device.

[1161] Data analysis function: Analyzes received information and classifies personality and voice characteristics.

[1162] Feature extraction function: Extracts important features from the analyzed data.

[1163] AI model training function: Trains an AI model based on feature-extracted data.

[1164] Conversation simulation function: Conducts conversation simulations with users using a trained AI model.

[1165] Emotion analysis function: Analyzes emotions from the user's voice, text tone, facial expressions, etc.

[1166] Conversation adjustment function: Adjusts the content and tone of conversations based on analyzed emotional data.

[1167] API endpoint setting function: Set the trained AI model and emotion engine to an API endpoint accessible from the device.

[1168] Terminal configuration

[1169] The terminal has the following features:

[1170] Information input screen: Displays a screen where users can enter information about family, friends, and pets.

[1171] Data transmission function: Sends the entered information to the server.

[1172] Emotion data collection function: Collects user voice, text, and facial expression data.

[1173] Response reception function: Receives a response from the server.

[1174] Content display function: Displays the received response to the user or provides an audio response using speech synthesis.

[1175] Conversation log saving function: Saves conversation logs between the user and their virtual family.

[1176] User Actions

[1177] 1. Log in

[1178] The user launches the application on their device and enters their authentication information on the login screen. The first time they log in, they are required to create a new account.

[1179] 2. Enter information

[1180] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[1181] 3. Start a conversation

[1182] After the information is sent to the server, the user can initiate everyday conversations such as "How was your day?"

[1183] Specific examples

[1184] Example 1: Everyday conversation

[1185] The user launches the application and enters information about their mother. For example, the "mother"'s personality could be "kind," her voice characteristic "high-pitched," and her relationship with the virtual mother "understanding." If the user talks about a difficult time at work, the virtual mother responds, "Really? What happened?" If the user continues, "Your boss scolded me," the virtual mother empathizes, saying, "That must have been tough. But you did a great job." The user's emotions are then analyzed, and the AI ​​adjusts the expressions of empathy and encouragement appropriately.

[1186] Example 2: Consultation

[1187] The user inputs information about a friend and sets the "friend's" personality to "energetic," the voice characteristics to "mid-tone," and the relationship to "good at encouraging people." When the user says, "I've been feeling down lately," the virtual friend will respond with an accepting attitude, saying, "What happened? Tell me." When the user says, "A lot of things have been happening...," the emotion engine analyzes the user's emotional state, and based on that information, the virtual friend will dynamically adjust the words of encouragement and respond, saying, "That's tough, but I know you can get through it!"

[1188] Program processing

[1189] Server-side processing

[1190] The server receives and analyzes input information and emotion data from the user. It trains an AI model based on the analysis results and works with the emotion engine to simulate a conversation with the user. The results are returned to the user's device via an API endpoint.

[1191] Terminal side processing

[1192] The device collects input information and emotion data from the user and sends it to the server, receives responses from the server and displays them to the user in text or audio format, and saves the conversation log and sends it to the server for later reuse.

[1193] Through this operational flow and each function, the system provides consistent psychological support to users, realizing a natural and empathetic conversation experience.

[1194] The processing flow will be explained below.

[1195] Program processing flow

[1196] Server-side processing

[1197] Step 1:

[1198] The server receives information about family, friends, and pets entered by the user, including their names, personalities, vocal characteristics, and relationships.

[1199] Step 2:

[1200] The server analyzes the received information and classifies the personality and voice characteristics into specific categories. For example, personality can be classified as "friendly," "strict," or "energetic," and voice characteristics can be classified as "gentle," "mid-range," or "high-pitched."

[1201] Step 3:

[1202] The server extracts important features from each piece of analyzed information, allowing the AI ​​to more accurately recreate the personalities and characteristics of family, friends, and pets.

[1203] Step 4:

[1204] The server trains an AI model based on the feature-extracted data using machine learning algorithms (e.g., natural language processing models).

[1205] Step 5:

[1206] The server uses the trained AI model to simulate a conversation with the user, using natural language processing based on the user's input.

[1207] Step 6:

[1208] The server uses an emotion analysis engine to analyze the user's emotions from their voice, text, and facial expression data. For example, it can identify emotional states such as "angry," "sad," or "happy" from their voice.

[1209] Step 7:

[1210] The server adjusts the content and tone of the AI ​​model's response based on the results of the emotion analysis, generating an appropriate response that matches the emotions the user is feeling.

[1211] Step 8:

[1212] The server transmits data including the tailored conversation response to the user terminal.

[1213] Terminal side processing

[1214] Step 1:

[1215] The user launches the application on their device and enters their authentication information on the login screen. The first time they log in, they are required to create a new account.

[1216] Step 2:

[1217] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[1218] Step 3:

[1219] The device sends the entered information to the server using HTTPS, ensuring secure communication.

[1220] Step 4:

[1221] The device collects voice, text, and facial expression data and transmits it to a server in real time. Voice data is collected through the device's microphone, and facial expression data is collected using the device's camera.

[1222] Step 5:

[1223] The device receives a response from the server, which is the content of the conversation (such as text data or voice data) generated by sentiment analysis and the trained AI model.

[1224] Step 6:

[1225] The device displays the received response to the user. In addition to displaying it in text format, it can also respond verbally using speech synthesis. The device uses a speech synthesis library or service (e.g., Google Text-to-Speech).

[1226] Step 7:

[1227] The device will store conversation logs between the user and their virtual family, which will then be sent to a server for analysis and to improve the AI ​​model.

[1228] User Actions

[1229] Step 1:

[1230] The user launches the app and enters their credentials on the login screen. The first time they log in, they are required to create a new account.

[1231] Step 2:

[1232] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[1233] Step 3:

[1234] The user initiates a conversation, for example, "How was your day?" and their virtual family member responds, "You did a great job today, did anything special happen?"

[1235] Step 4:

[1236] The user expresses their emotions. For example, if the user says, "I've been feeling down lately," the emotion analysis engine analyzes the user's tone of voice and facial expressions, and the server generates an appropriate response. As a result, the virtual family member responds with empathy, saying, "That's tough. Can you tell me more about what happened? I hope it makes you feel a little better."

[1237] In this way, the system provides psychological support to users at each step, enabling a natural and empathetic conversational experience.

[1238] Example 2

[1239] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1240] In modern society, the number of individuals suffering from loneliness, isolation, and anxiety is increasing, and effective methods to achieve psychological stability are needed. However, existing technologies lack systems that provide customized conversations based on individual emotions and personalities, resulting in users being unable to receive sufficient empathy or support. In particular, there is a lack of systems that combine the functionality of emotion analysis and dynamically adjusting appropriate responses, resulting in an insufficient quality of psychological support for users.

[1241] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving information about family, friends, and pets input by the user, means for analyzing the received information and classifying the personalities and vocal characteristics of the family, friends, and pets, means for training an AI model based on the analyzed data, means for conducting a conversation simulation with the user using the trained AI model, means for collecting the user's voice, text, and facial expression data and analyzing emotions, means for adjusting the content and tone of the conversation based on the analyzed emotional data, and means for setting an API endpoint that allows access from the user terminal. This makes it possible to provide consistent psychological support to the user and realize a natural and empathetic conversation experience.

[1242] "Users" refer to individuals who use the system, and who often experience loneliness, isolation, and worries.

[1243] "Information" refers to data about family, friends, and pets, including names, personalities, vocal characteristics, and relationships.

[1244] "Receiving" refers to the process by which the server retrieves information sent from the user terminal.

[1245] "Analysis" refers to the process of analyzing received information and understanding and classifying its content.

[1246] "Personality" is data that indicates the behavioral characteristics and attitudes of humans and pets, and is used in this system for emotion analysis and conversation simulation.

[1247] "Voice characteristics" are data that indicate the pitch, tone, intonation, etc. of a voice, and are used to reproduce the characteristics of the voices of family members, friends, and pets entered by the user.

[1248] "Classification" is the process of dividing information into multiple categories and finding similarities and differences based on each person's personality and voice characteristics.

[1249] "Training" is the process by which an AI model learns from data to produce more accurate and efficient outputs.

[1250] An "AI model" is a collection of algorithms that use artificial intelligence to make predictions or generate results, and operates based on trained data.

[1251] "Conversational simulation" refers to the process of using a trained AI model to simulate a conversation with a user.

[1252] "Emotion analysis" is a technology that infers emotions from a user's voice, text, facial expressions, etc., and is used by the system to generate more natural and empathetic responses.

[1253] "Tone" refers to the intonation and timbre of a voice, and is an element that determines the atmosphere in which the content of a conversation is conveyed.

[1254] "Tuning" is the process of changing the content and tone of the conversation based on the analyzed data to provide an appropriate and empathetic response to the user.

[1255] An "API endpoint" refers to a specific URL or URI through which a user device and a server communicate, and through which data is sent and received.

[1256] "Terminal" refers to a device used by a user to perform operations, including smartphones and tablets.

[1257] A "conversation log" is data that records the history of conversations between a user and an AI model, and can be referenced and reused later.

[1258] This invention is a system that uses AI technology to enable individuals who are feeling lonely, isolated, or worried to have everyday conversations and seek advice from a virtual family member, in order to achieve psychological stability.The system incorporates an emotion engine that recognizes the user's emotions, providing a more natural and empathetic conversation experience.

[1259] System Configuration

[1260] Server-side configuration

[1261] The server has the following features:

[1262] Data reception function: Receives information about family, friends, and pets sent from the user's device. Specifically, it receives JSON format data via HTTP requests.

[1263] Data analysis function: Analyzes received information and classifies personality and voice characteristics. Uses NLP libraries to perform text analysis and extract personality attributes.

[1264] Feature extraction function: Extracts important features from the analyzed data and stores them in a database. These include personality traits, voice features, etc.

[1265] AI model training function: Trains generative AI models based on feature extraction data. Specific software used includes TensorFlow and PyTorch.

[1266] Conversation Simulation: A trained AI model is used to simulate conversations with users, with the virtual family generating prompts such as "How was your day?"

[1267] Emotion analysis function: Analyzes emotions from the user's voice, text tone, facial expressions, etc. It uses a voice recognition engine to identify positive, negative, and other emotions.

[1268] Conversation adjustment function: The content and tone of the conversation are adjusted based on the analyzed emotional data. For example, it may give gentle advice such as "You should take it easy today."

[1269] API endpoint setting function: Sets the trained AI model and emotion engine as an API endpoint accessible from the device. Sets a RESTful API and notifies the user device of the endpoint URL.

[1270] Terminal configuration

[1271] The terminal has the following features:

[1272] Information input screen: A screen is displayed where users can enter information about their family, friends, and pets. A smartphone application is used.

[1273] Data transmission function: Sends information entered by the user to the server. The entered text data is sent to the server via an HTTP POST request.

[1274] Emotion data collection function: Collects user voice, text, and facial expression data using the device's microphone and camera.

[1275] Response reception function: Receives responses from the server and displays them to the user. Retrieves JSON format responses from the server's API, parses them, and displays them in the UI.

[1276] Content display function: Displays the received response to the user. Not only in text format, but also in voice response using speech synthesis. Plays it aloud using a speech synthesis engine.

[1277] Conversation log storage function: Save the conversation logs between the user and the virtual family and send them to the server for later reuse. Save the conversation data in a local database and back it up to the server periodically.

[1278] Specific examples

[1279] Example 1: Everyday conversation

[1280] The user launches the application and enters information about their mother. For example, the "mother"'s personality could be "kind," her voice characteristic "high-pitched," and her relationship with the virtual mother "understanding." If the user talks about a difficult time at work, the virtual mother responds, "Really? What happened?" If the user continues, "Your boss scolded me," the virtual mother empathizes, saying, "That must have been tough. But you did a great job." At this point, the emotion engine analyzes the user's emotions, and the AI ​​adjusts the expressions of empathy and encouragement appropriately.

[1281] Example 2: Consultation

[1282] The user inputs information about a friend and sets the friend's personality to "energetic," the voice characteristics to "mid-tone," and the relationship to "good at encouraging others." When the user says, "I've been feeling down lately," the virtual friend will respond with an accepting attitude, saying, "What happened? Tell me." When the user says, "A lot of things have been happening...," the emotion engine analyzes the user's emotions, and based on that information, the virtual friend will dynamically adjust the words of encouragement and respond, saying, "That's tough, but I know you can get through it!"

[1283] Prompt Sentence Examples

[1284] "How was work today?"

[1285] "What's been going on lately?"

[1286] "Really? What happened?"

[1287] "That must have been tough. But you did a great job."

[1288] Through these functions and specific operation flow, this system can provide consistent psychological support to users and realize a natural and empathetic conversation experience.

[1289] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1290] Step 1: Display the information entry screen

[1291] The device displays a screen where the user can enter information about family, friends, and pets. Specifically, the device UI displays text input fields and various selection options. Through these fields, the user enters information such as name, personality, voice characteristics, and relationships. Input data includes the name "Mother," personality "gentle," and voice characteristics "high-pitched."

[1292] Step 2: Send user information to the server

[1293] The device sends the information entered by the user to the server. Specifically, the input data is converted to JSON format and sent to the server using an HTTP POST request. This request includes information such as name, personality, voice characteristics, and relationships. The data is formatted in a way that is easy for the server to receive.

[1294] Step 3: Receiving data

[1295] The server receives the information sent from the user's device. Specifically, the server receives an HTTP request and extracts information such as name, personality, voice characteristics, and relationships from the JSON-formatted data. The received data includes attributes such as "mother," "gentle," and "high-pitched voice."

[1296] Step 4: Data analysis

[1297] The server analyzes the received information and classifies each individual's personality and voice characteristics. Specifically, it uses an NLP library to perform text analysis and extract personality attributes. It also uses voice recognition technology to analyze voice features. The input data is the name, personality, voice characteristics, and relationships, and the output data is the analyzed personality attributes and voice features.

[1298] Step 5: Feature Extraction

[1299] The server extracts important features from the analyzed data and stores them in a database. Specifically, personality traits and voice features are extracted as important features. For example, a personality trait such as "gentle" or a voice feature such as "high-pitched" are extracted and registered in the database.

[1300] Step 6: Training the AI ​​model

[1301] The server then trains the AI ​​model based on the extracted feature data. Specifically, it uses TensorFlow and PyTorch to train the generative AI model while referring to past conversation data. During this process, the system processes a dataset containing personality traits and voice features to improve its ability to generate natural conversations with the user.

[1302] Step 7: Conversation simulation

[1303] The server uses the trained AI model to simulate a conversation with the user. Specifically, it receives a prompt from the user, such as "How was your day?", and the AI ​​model generates an appropriate response. For example, in response to the question, "You had a hard time at work," it generates the response, "Really? What happened?"

[1304] Step 8: Sentiment Analysis

[1305] The server collects the user's voice, text, and facial expression data and analyzes their emotions. Specifically, it uses a voice recognition engine and facial expression analysis software to identify the user's emotional state. For example, it can estimate emotions such as stress or relief from the user's voice data.

[1306] Step 9: Conversation Adjustment

[1307] The server adjusts the content and tone of the conversation based on the analyzed emotional data. Specifically, if the user is tired, it will generate a softer response based on the emotion analysis results, such as "You should get some rest today." This provides a customized response that is optimized for the user's emotions.

[1308] Step 10: Receive a response

[1309] The device receives the response from the server and displays it to the user. Specifically, it parses the JSON format response from the server and displays it as text on the UI. It also plays back the audio using a speech synthesis engine. For example, it provides the user with a response such as "That must have been tough, but you did well" both as text and audio.

[1310] Step 11: Save conversation logs

[1311] The device saves conversation logs between the user and their virtual family and sends them to a server for later reuse. Specifically, each conversation episode is recorded in a local database and periodically backed up to the server. The saved data is used for future conversation simulations and user support.

[1312] (Application example 2)

[1313] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1314] In modern society, the number of individuals feeling lonely or isolated is increasing. While there is a need for methods to provide psychological stability to these individuals, real-life relationships alone are often insufficient. Furthermore, even in virtual shopping experiences, personalized support tailored to individual needs is lacking. Therefore, a system that uses a virtual companion to provide empathetic conversation and advice to individuals who feel lonely or isolated is needed.

[1315] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1316] In this invention, the server includes means for receiving person information input by a user, means for analyzing the received information and classifying the person's personality and voice characteristics, means for training an AI model based on the analyzed data, means for conducting a conversation simulation with the user using the trained AI model, means for setting an API endpoint accessible from a user terminal, and shopping assistance means for a person to provide product advice in a virtual store. This enables individuals who feel lonely or isolated to receive empathetic conversation and personalized shopping support in a virtual environment.

[1317] "User" refers to an individual who uses the system to have conversations with virtual people or receive shopping assistance.

[1318] "Input" refers to information provided by the user to the system, specifically data such as a person's personality and voice characteristics.

[1319] "People" refers to virtual beings such as family, friends, and companions that users recreate within the system.

[1320] "Information receiving means" refers to a function that allows the system to receive person information entered by the user.

[1321] "Analysis means" refers to a function for classifying a person's personality and voice characteristics based on the received information.

[1322] "Training means" refers to the function used to train an AI model based on analyzed data.

[1323] "Conversation simulation means" refers to a function that uses a trained AI model to conduct a conversation between a user and a virtual person.

[1324] "API endpoint setting means" refers to a function for setting an API endpoint that can be accessed from a user terminal.

[1325] "Shopping assistance means" refers to a function that enables a virtual person to provide product advice to a user in a virtual store.

[1326] System configuration

[1327] Server-side processing

[1328] 1. Means of receiving information

[1329] The server receives information about the person sent from the user's device, including their name, personality, voice characteristics, and relationships.

[1330] 2. Analysis method

[1331] The server analyzes the received information and classifies the person's personality and voice characteristics using natural language processing and voice analysis technology.

[1332] 3. Training methods

[1333] The server trains an AI model based on the analyzed data using a deep learning framework (e.g., TensorFlow, PyTorch).

[1334] 4. Conversation Simulation Methods

[1335] The server uses a trained AI model to simulate a conversation with the user, allowing the virtual persona to have a natural conversation.

[1336] 5. API endpoint configuration method

[1337] The server sets up an API endpoint that can be accessed from the user's device, allowing the user to interact with the virtual person at any time.

[1338] 6. Shopping assistance methods

[1339] The server allows a virtual person in the virtual store to provide product advice to the user, based on product information and user reviews.

[1340] Terminal side processing

[1341] 1. Information input screen

[1342] Users enter their personal information on the device screen, which is built using HTML, CSS, and JavaScript.

[1343] 2. Data transmission function

[1344] The terminal sends the input information to the server via a RESTful API.

[1345] 3. Emotion data collection function

[1346] The device uses WebRTC and TensorFlow.js to collect voice and facial expression data and send it to a server.

[1347] 4. Response Receiving Function

[1348] The device receives the response from the server and processes the received data using the JavaScript fetch API or similar.

[1349] 5. Content display function

[1350] The device displays the received response to the user, and can also play the received data aloud using a speech synthesis API (e.g., Google Text-to-Speech API).

[1351] User Actions

[1352] 1. Initial system setup

[1353] Users launch the application and enter their authentication information on the login screen. The first time they log in, they are required to create a new account.

[1354] 2. Enter your information

[1355] Users enter information about virtual characters such as family and friends on a screen where they can enter their names, personality traits, voice characteristics, relationships, etc.

[1356] 3. Start a conversation

[1357] Users can initiate everyday conversations such as "How was your day?" The virtual person will provide natural conversation.

[1358] Specific examples

[1359] During a virtual shopping trip, a user wearing smart glasses can choose new clothes and talk to their virtual family. When the user asks, "What do you think of this outfit?", the virtual mother responds, "It looks great, and the color suits you!"

[1360] Prompt Sentence Examples

[1361] User input: "I've been interested in this outfit lately. What do you think?"

[1362] Generative AI Model Input: The AI ​​model, the user is interested in a casual shirt in a solid blue color. Given that the user's mother has a "kind" personality, generate an empathetic response when the user asks about the shirt.

[1363] The system can provide empathetic conversations and personalized shopping support through a virtual companion to individuals who feel lonely or isolated.

[1364] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1365] Step 1:

[1366] The terminal collects the user's login information on the input screen and sends the authentication information to the server. The input includes the user name and password, and session information is generated as output. The server receives this information, performs authentication, and returns the session information.

[1367] Step 2:

[1368] The user enters information about virtual people, such as family and friends, into the input screen. The input includes name, personality, voice characteristics, and relationships, and is sent as output in JSON format to the server. The device then sends this to the server.

[1369] Step 3:

[1370] The server analyzes the received personal information and classifies personality and voice characteristics using Natural Language Processing (NLP) and voice analysis technology. It processes the JSON data received as input, converts personality and voice characteristics into numerical data, and outputs the classification results.

[1371] Step 4:

[1372] The server trains an AI model based on the classified data. It receives the data classified in the previous step as input, trains the model using a deep learning framework (e.g., TensorFlow, PyTorch), and generates a trained AI model as output.

[1373] Step 5:

[1374] The device uses its emotion data collection function to collect the user's voice and facial expression data. It uses data acquired from the microphone and camera as input, processes it using WebRTC and TensorFlow.js, and outputs analyzed emotion data.

[1375] Step 6:

[1376] The server uses a trained AI model to simulate a conversation between the user and a virtual character. It receives emotional data and the user's questions and comments as input, generates conversation content, and obtains response data as output.

[1377] Step 7:

[1378] The server sends the response data to the user device through the API endpoint, processes the HTTP request via the API using the generated response data as input, and sends the data to the device as output.

[1379] Step 8:

[1380] The response data received by the device is played back to the user as audio using a speech synthesis API. Response data from the server is received as input, converted into audio using a speech synthesis API (e.g., Google Text-to-Speech API), and an audio response is obtained as output.

[1381] Step 9:

[1382] The device saves the conversation log between the user and the virtual person and sends it to the server as reusable data later. It receives the conversation data as input, saves it in LocalStorage, and obtains the saved log data as output.

[1383] Step 10:

[1384] The server generates shopping assistance data for the virtual store, and the virtual person provides product advice to the user. It receives product data and user interests as input, generates advice using an AI model, and sends the advice data to the terminal as output.

[1385] Step 11:

[1386] The terminal presents the received advice data to the user. Using the advice data received from the server as input, the advice is displayed on the screen or via voice, and the advice is provided to the user as output.

[1387] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1388] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1389] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1390] [Fourth embodiment]

[1391] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1392] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1393] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1394] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1395] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1396] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1397] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1398] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1399] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1400] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1401] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1402] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1403] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1404] This invention is a system that uses AI technology to provide everyday conversations and consultations with virtual family members to help individuals who are feeling lonely, isolated, or worried to find psychological stability. The system is designed to receive input from users, train an AI model based on that information, and engage in natural conversations with the users.

[1405] System Configuration

[1406] Server-side configuration

[1407] The server has the following features:

[1408] Data reception function: Receives information about family, friends, and pets sent from the user's device.

[1409] Data analysis function: Analyzes received information and classifies personality and voice characteristics.

[1410] Feature extraction function: Extracts important features from the analyzed data.

[1411] AI model training function: Trains an AI model based on feature-extracted data.

[1412] Conversation simulation function: Conducts conversation simulations with users using a trained AI model.

[1413] API endpoint setting function: Set a trained AI model to an API endpoint that can be accessed from the device.

[1414] Terminal configuration

[1415] The terminal has the following features:

[1416] Information input screen: Displays a screen where users can enter information about family, friends, and pets.

[1417] Data transmission function: Sends the entered information to the server.

[1418] Response reception function: Receives a response from the server.

[1419] Content display function: Displays the received response to the user or provides an audio response using speech synthesis.

[1420] Conversation log saving function: Saves conversation logs between the user and their virtual family.

[1421] User Actions

[1422] 1. Log in

[1423] The user launches the application on their device and enters their authentication information on the login screen. The first time they log in, they are required to create a new account.

[1424] 2. Enter information

[1425] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[1426] 3. Start a conversation

[1427] After the information is sent to the server, the user can initiate everyday conversations such as "How was your day?"

[1428] Specific examples

[1429] Example 1: Everyday conversation

[1430] The user launches the application and enters information about their mother. For example, the "mother"'s personality can be "kind," her voice characteristic can be "high-pitched," and her relationship status can be "understanding." If the user says, "I had a hard time at work today," the virtual mother will respond, "Really? What happened?" If the user continues, "My boss scolded me," the virtual mother will empathize, saying, "That must have been tough. But you did a great job."

[1431] Example 2: Consultation

[1432] The user inputs information about a friend, and for example, sets the friend's personality to "energetic," the voice characteristics to "mid-tone," and the relationship to "good at encouraging others." If the user says, "I've been feeling down lately," the virtual friend will show an accepting attitude by saying, "What happened? Tell me." If the user says, "A lot of things have been happening at once...," the virtual friend will encourage them by saying, "That's tough, but I know you can get through it!"

[1433] In this way, the present invention makes full use of AI technology to provide users with an experience that makes them feel as if they are conversing with real family, friends, or pets, providing psychological support.

[1434] The processing flow will be explained below.

[1435] Program processing flow

[1436] Server-side processing

[1437] Step 1:

[1438] The server receives information about family, friends, and pets entered by the user, including their names, personalities, vocal characteristics, and relationships.

[1439] Step 2:

[1440] The server analyzes the received information and classifies the personality and voice characteristics into specific categories. For example, personality can be classified as "friendly," "strict," or "energetic," while voice characteristics can be classified as "gentle," "mid-range," or "high-pitched."

[1441] Step 3:

[1442] The server extracts important features from each piece of analyzed information, allowing the AI ​​to more accurately recreate the personalities and characteristics of family, friends, and pets.

[1443] Step 4:

[1444] The server trains an AI model based on the feature-extracted data using machine learning algorithms (e.g., natural language processing models).

[1445] Step 5:

[1446] The server uses the trained AI model to simulate a conversation with the user, using natural language processing based on the user's input.

[1447] Step 6:

[1448] The server deploys the trained AI model to an API endpoint, making it accessible from user devices.

[1449] Terminal side processing

[1450] Step 1:

[1451] The user launches the application on their device and enters their authentication information on the login screen. The first time they log in, they are required to create a new account.

[1452] Step 2:

[1453] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[1454] Step 3:

[1455] The device sends the entered information to the server using HTTPS, ensuring secure communication.

[1456] Step 4:

[1457] The device receives a response from the server. The response is the content of the conversation generated by the AI ​​model (text data, voice data, etc.).

[1458] Step 5:

[1459] The device displays the received response to the user. In addition to displaying it in text format, it can also respond in voice using speech synthesis. The device uses a speech synthesis library or service (e.g., Google Text-to-Speech).

[1460] Step 6:

[1461] The device will store conversation logs between the user and their virtual family, which will then be sent to a server for analysis and to improve the AI ​​model.

[1462] User Actions

[1463] Step 1:

[1464] The user launches the app and enters their credentials on the login screen. The first time they log in, they are required to create a new account.

[1465] Step 2:

[1466] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[1467] Step 3:

[1468] The user initiates a conversation, for example, "How was your day?", and the virtual family member responds, "You did a great job today. Did anything special happen?"

[1469] Step 4:

[1470] For example, if a user says, "Work has been tough lately," their virtual family member will empathize and respond, "That must be tough. Tell me more about what happened. I hope it makes you feel a little better."

[1471] In this way, the system can provide psychological support to the user through each step.

[1472] Example 1

[1473] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1474] In modern society, many individuals suffer from loneliness, isolation, and worries, and there is a need for a means to receive psychological support. However, due to the limited time and opportunity to consult or talk with family and friends, there is a lack of effective systems to provide such psychological support. Another issue is the lack of systems that can provide dialogue tailored to the individual needs of users.

[1475] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1476] In this invention, the server includes means for receiving information about family, friends, and pets input by the user, means for analyzing the received information and classifying the personalities and vocal characteristics of the family, friends, and pets, means for training an AI model based on the analyzed and feature-extracted data, means for conducting a conversation simulation with the user using the trained AI model, means for setting an API endpoint that enables access from the user terminal, means for the user to input authentication information into a login screen for authentication, means for displaying the information input by the user as text or voice or responding using voice synthesis, and means for saving a conversation log between the user and the virtual family. This enables the user to receive psychological support through natural conversations with their virtual family and friends.

[1477] A "user" is an individual who uses the system to enter information about family, friends, and pets and engage in virtual interactions.

[1478] A "server" is a computer system that receives information sent by users and performs analysis, feature extraction, AI model training, conversation simulation, etc.

[1479] A "terminal" is a device through which a user inputs information and engages in virtual interactions via communication with a server. Examples include smartphones and tablets.

[1480] The "data receiving means" is a function that allows the server to receive information about family, friends, and pets sent by the user.

[1481] "Data analysis means" is a function for analyzing received information and classifying the personalities and vocal characteristics of family, friends, and pets.

[1482] "Feature extraction means" is a function that extracts important features from the analyzed data and uses them to train the AI ​​model.

[1483] "AI model training means" is a function for training an artificial intelligence model based on feature-extracted data.

[1484] "Conversation simulation means" is a function for simulating a virtual conversation between a user and the AI ​​model using a trained AI model.

[1485] The "API endpoint setting means" is a function for setting a trained AI model as an API endpoint that can be accessed from a user terminal.

[1486] The "login authentication means" is a function in which a user inputs authentication information into a login screen and the server verifies that information.

[1487] "Information display means" is a function for displaying information entered by the user as text or voice, or for responding using voice synthesis.

[1488] The "conversation log storage means" is a function for storing conversation logs between the user and the virtual family and transmitting them to the server as reusable data later.

[1489] "Virtual family" refers to family, friends, and pets recreated by an AI model trained based on information entered by the user.

[1490] This invention is a system that uses artificial intelligence (AI) technology to provide everyday conversations and consultations with a virtual family, helping individuals experiencing loneliness, isolation, or anxiety to find psychological stability. The system is designed to receive user input, train an AI model based on that information, and have natural conversations with the user.

[1491] System Configuration

[1492] Server-side configuration

[1493] The server has the following features:

[1494] Data reception function: Receives information about family, friends, and pets sent by the user from the device. Specifically, it receives data such as names, personalities, voice characteristics, and relationships entered by the user.

[1495] Data analysis function: Analyzes the received information and classifies the personalities and vocal characteristics of family, friends, and pets. For example, it analyzes personality and vocal characteristics such as "gentle" or "high-pitched."

[1496] Feature extraction function: Extracts important features from the analyzed data, thereby extracting the elements necessary for training the AI ​​model.

[1497] AI model training function: The AI ​​model is trained based on the feature-extracted data, for example, using a deep learning algorithm to replicate the personality and voice of specific family or friends selected by the user.

[1498] Conversation simulation function: A trained AI model is used to simulate a conversation with the user, generating appropriate responses to prompts entered by the user.

[1499] API endpoint setting function: Set the trained AI model to an API endpoint accessible from the device, allowing users to easily access it at any time.

[1500] Login authentication function: A function that allows users to enter authentication information on the login screen and perform authentication.

[1501] Terminal configuration

[1502] The terminal has the following features:

[1503] Information entry screen: Displays a screen where users can enter information about family, friends, and pets, including details such as their names, personalities, vocal characteristics, and relationships.

[1504] Data transmission function: Sends the input information to the server. For example, it sends data to the server using an HTTP POST request.

[1505] Response reception function: Receives responses from the server, analyzes the received responses, and displays them in an easy-to-understand manner for the user.

[1506] Content display function: A function that displays the received response to the user, and can display text or output audio using voice synthesis.

[1507] Conversation log saving function: The function to save the conversation log between the user and the virtual family, which can be used to review the conversation later.

[1508] Usage example

[1509] Example 1: Everyday conversation

[1510] The user launches the application and enters information about their mother. For example, the "mother"'s personality can be "kind," her voice characteristic can be "high-pitched," and her relationship status can be "understanding." If the user says, "I had a hard time at work today," the virtual mother will respond, "Really? What happened?" If the user continues, "My boss scolded me," the virtual mother will empathize, saying, "That must have been tough. But you did a great job."

[1511] Example 2: Consultation

[1512] The user enters information about a friend, setting the friend's personality to "energetic," voice characteristics to "mid-tone," and relationship to "good at encouraging others." If the user says, "I've been feeling down lately," the virtual friend will respond with an accepting attitude, "What happened? Tell me." If the user says, "A lot of things have been happening at once...," the virtual friend will encourage them, saying, "That's tough, but I know you can get through it!"

[1513] With the above configuration, the system provides users with an experience that makes them feel as if they are having a conversation with their real family, friends, or pets, and can provide psychological support.

[1514] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1515] Step 1:

[1516] User login: The user launches the application and enters authentication information on the login screen. When the user enters their email address and password and presses the "Login" button, the authentication information is sent from the device to the server. The server compares this with the database, and if authentication is successful, the user is taken to the home screen. The input is the user's authentication information, and the output is the authentication result.

[1517] Step 2:

[1518] Information input: An information input screen is displayed for the user to enter information about family, friends, and pets. Here, information such as "name," "personality," "voice characteristics," and "relationship" are entered in the input fields. When the user confirms the input and presses the "send" button, the data is sent from the device to the server. The input is the information about family members, etc. entered by the user, and the output is the sent data.

[1519] Step 3:

[1520] Data transmission: The device sends the information entered by the user to the server. Specifically, the device's data transmission function converts the input data into JSON format and sends it to the server as an HTTP POST request. The input is the user's input data, and the output is the data sent to the server.

[1521] Step 4:

[1522] Data reception: The server receives the data sent from the terminal. The data reception function analyzes the HTTP request, extracts the user data, and stores it in the database. The input is the data in the HTTP request format, and the output is the stored data.

[1523] Step 5:

[1524] Data analysis: The server analyzes the received data. The data analysis function analyzes information such as name, personality, vocal characteristics, and relationships, and performs classification. For example, characteristics such as "kind," "high-pitched voice," and "understanding" are analyzed. The input is the saved user data, and the output is the analyzed classified data.

[1525] Step 6:

[1526] Feature Extraction: The server extracts important features from the analyzed data. The feature extraction function quantifies the user data and generates parameters for use in training the AI ​​model. The input is the analyzed classified data, and the output is the feature-extracted data.

[1527] Step 7:

[1528] AI model training: The server trains the AI ​​model based on the feature-extracted data. The AI ​​model training function uses a deep learning algorithm to build a model that matches the user's specific data. The input is the feature-extracted data, and the output is the trained AI model.

[1529] Step 8:

[1530] Conversation simulation: The server uses a trained AI model to simulate a conversation with the user. The conversation simulation function receives the user's utterances and generates appropriate responses. The input is the user's utterance prompt, and the output is the generated response.

[1531] Step 9:

[1532] Displaying and responding to conversation: The device displays the response received from the server to the user. The response receiving function analyzes the received data and presents it to the user in text or audio format. The input is the response data from the server, and the output is the content displayed to the user.

[1533] Step 10:

[1534] Conversation log storage: The device stores conversation logs between the user and virtual family members. The conversation log storage function records the content of the conversation in a database so that the user can review the conversation later. The input is the conversation data, and the output is the saved conversation log.

[1535] The above is the specific processing flow of the program of this system.

[1536] (Application example 1)

[1537] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1538] In modern society, many people suffer from loneliness, isolation, and worries, and are seeking ways to achieve psychological stability. There is also a growing need for virtual assistance with shopping-related concerns and product selection. However, effective technology to solve these problems is not yet available. In particular, there is no system that combines AI technology that can hold natural conversations based on user information with a virtual assistant that provides product information and purchasing advice.

[1539] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1540] In this invention, the server includes means for receiving information about family, friends, and pets input by the user, means for analyzing the received information and classifying the personalities and vocal characteristics of the family, friends, and pets, means for training an AI model based on the analyzed data, means for conducting a conversation simulation with the user using the trained AI model, means for setting an API endpoint accessible from the user's terminal, means for the user to converse with the virtual assistant and receive product information and purchasing consultation support, and means for the virtual assistant to recommend appropriate products based on the received information. This allows the user to not only receive psychological support but also virtually receive shopping-related consultation and product selection support.

[1541] "User" refers to an individual or group that uses the system.

[1542] "Inputted Information" refers to data about the personalities and vocal characteristics of family members, friends, pets, etc. that a user provides to the system.

[1543] "Means for receiving" refers to the technical means by which the system receives information entered by the user.

[1544] "Means for analyzing" refers to technical means for analyzing received information and extracting specific patterns or attributes.

[1545] "Means for classifying personality and voice characteristics" refers to technical means for categorizing personality and voice characteristics based on the analysis results.

[1546] "Training means" refers to the technical means by which an AI model is trained using classified data.

[1547] "Means for conducting conversational simulation" means technical means for conducting a conversation with a user using a trained AI model.

[1548] "Means for configuring API endpoints" means the technical means for defining API endpoints for accessing a trained AI model.

[1549] A "virtual assistant" is a virtual support character that uses AI technology to interact with users and provide various types of assistance.

[1550] "Providing product information" refers to the act of a virtual assistant presenting information about a specific product to a user.

[1551] "Means for assisting with purchasing consultation" are technological means by which a virtual assistant can provide advice to a user when selecting a product.

[1552] "Means for making product recommendations" refers to the technical means by which the virtual assistant recommends appropriate products based on the user's input information.

[1553] "Means for storing conversation logs" means technical means for storing the history of conversations between a user and a virtual assistant.

[1554] "Server" is a remote data processing system that processes information entered by the user and assists in the operation of the virtual assistant.

[1555] This invention is a system that uses AI technology to interact with virtual family and virtual assistants. This system trains an AI model based on information about family, friends, and pets entered by the user, and provides natural conversations to the user. Furthermore, this system also has a shopping support function in a virtual store.

[1556] Server-side configuration

[1557] The server has the following features:

[1558] 1. Information reception function: Receives information about family, friends, and pets sent from the user's device. This information includes names, personalities, voice characteristics, relationships, etc.

[1559] 2. Information analysis function: Analyzes received data and classifies personality and voice characteristics. Specifically, it uses Python and TensorFlow to preprocess the data and then analyzes it using natural language processing technology.

[1560] 3. Feature extraction: Extracting important features from the analyzed data. This process uses machine learning algorithms.

[1561] 4. AI model training function: Trains an AI model based on the extracted features. Model training is performed using Python and TensorFlow.

[1562] 5. Conversation Simulation: Simulate conversations with users using a trained AI model. Build a RESTful API using Flask and generate responses based on user requests.

[1563] 6. API endpoint configuration function: Set the trained AI model as an API endpoint accessible from the device. Deploy the API using AWS or GCP.

[1564] Terminal configuration

[1565] The terminal has the following features:

[1566] 1. Information input screen: Provides a screen where users can enter information about family, friends, and pets. Build a cross-platform application using React Native.

[1567] 2. Data transmission function: Sends the input information to the server. Axios is used as the HTTP client library.

[1568] 3. Response reception function: A function that receives responses from the server. It displays them on the UI using React Native.

[1569] 4. Content display function: Displays the received response to the user and, if necessary, provides a voice response using speech synthesis. For speech synthesis, Google's Text-to-Speech API is used.

[1570] 5. Conversation log storage function: A function to save conversation logs between users and virtual assistants. Saved in cloud storage using Firebase.

[1571] Specific examples

[1572] As a concrete example, we present a scenario in which a user converses with a virtual assistant to shop.

[1573] The user opens the app and selects a "friend" character. For example, the "friend"'s personality can be set to "friendly," the voice characteristics to "mid-tone," and the relationship to "likes shopping." If the user says, "I want a new smartphone," the virtual friend will ask, "What features do you want?" If the user answers, "I want one with a good camera," the virtual friend will suggest a specific product, saying, "This model has an excellent camera."

[1574] Prompt Sentence Examples

[1575] The user says, "I want a new smartphone." The virtual assistant asks, "What features do you want?" If the user replies, "I want one with a good camera," the virtual assistant suggests a specific product, saying, "This model has an excellent camera."

[1576] In this way, we provide a system that allows users to receive not only psychological support but also shopping-related advice and product selection through a virtual assistant. This system is expected to alleviate loneliness and isolation in modern society and improve the shopping experience.

[1577] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1578] Step 1:

[1579] The user starts the application and enters the required data into the information input screen.

[1580] Specifically, you enter information in text format, such as the names, personalities, vocal characteristics, and relationships of family members, friends, and pets.

[1581] The input data is temporarily saved on the device. The user checks whether the input is correct and then presses the send button.

[1582] Step 2:

[1583] The terminal receives information entered by the user and sends it to the server.

[1584] Specifically, input data is sent as an HTTP request using Axios.

[1585] The input data is sent to the server, and the server completes receiving the data.

[1586] Step 3:

[1587] The server parses the received data.

[1588] Specifically, using Python and TensorFlow, text data is analyzed using natural language processing technology to classify names, personalities, voice characteristics, and relationships.

[1589] The classified data is temporarily stored in an internal database, and the analysis results are extracted as features such as personality type, voice characteristics, and relationships.

[1590] Step 4:

[1591] The server trains the AI ​​model based on the analyzed data.

[1592] Specifically, features are input into the model using TensorFlow, and the model parameters are updated.

[1593] The input is feature data, and the output is a trained AI model.

[1594] Step 5:

[1595] The server uses a trained AI model to simulate a conversation with the user.

[1596] Specifically, user requests are sent to a RESTful API built using Flask, and the AI ​​model generates a response.

[1597] The input is the user's question or comment, and the output is the generated response message.

[1598] Step 6:

[1599] The server sets the API endpoint to be accessible from the device.

[1600] Specifically, the trained AI model is placed on an API endpoint deployed using AWS or GCP.

[1601] The input is an AI model and API configuration information, and the output is an accessible API endpoint.

[1602] Step 7:

[1603] The terminal receives the response from the server and displays the content.

[1604] Specifically, the response data received as a response to an HTTP request is displayed on the screen using React Native.

[1605] It also uses a speech synthesis function to output responses in voice. The input is response data from the API, and the output is displayed on the screen and output as voice.

[1606] Step 8:

[1607] The device stores a log of the conversation between the user and the virtual assistant.

[1608] Specifically, the content of the conversation is saved in real time to cloud storage such as Firebase.

[1609] The input is the text data of the conversation, and the output is the saved conversation log.

[1610] In this way, the entire system processes each step in order, enabling natural conversations and shopping consultations with users.

[1611] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1612] This invention is a system that uses AI technology to enable individuals who are feeling lonely, isolated, or worried to have everyday conversations and seek advice from a virtual family member, in order to achieve psychological stability.The system incorporates an emotion engine that recognizes the user's emotions, providing a more natural and empathetic conversation experience.

[1613] System Configuration

[1614] Server-side configuration

[1615] The server has the following features:

[1616] Data reception function: Receives information about family, friends, and pets sent from the user's device.

[1617] Data analysis function: Analyzes received information and classifies personality and voice characteristics.

[1618] Feature extraction function: Extracts important features from the analyzed data.

[1619] AI model training function: Trains an AI model based on feature-extracted data.

[1620] Conversation simulation function: Conducts conversation simulations with users using a trained AI model.

[1621] Emotion analysis function: Analyzes emotions from the user's voice, text tone, facial expressions, etc.

[1622] Conversation adjustment function: Adjusts the content and tone of conversations based on analyzed emotional data.

[1623] API endpoint setting function: Set the trained AI model and emotion engine to an API endpoint accessible from the device.

[1624] Terminal configuration

[1625] The terminal has the following features:

[1626] Information input screen: Displays a screen where users can enter information about family, friends, and pets.

[1627] Data transmission function: Sends the entered information to the server.

[1628] Emotion data collection function: Collects user voice, text, and facial expression data.

[1629] Response reception function: Receives a response from the server.

[1630] Content display function: Displays the received response to the user or provides an audio response using speech synthesis.

[1631] Conversation log saving function: Saves conversation logs between the user and their virtual family.

[1632] User Actions

[1633] 1. Log in

[1634] The user launches the application on their device and enters their authentication information on the login screen. The first time they log in, they are required to create a new account.

[1635] 2. Enter information

[1636] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[1637] 3. Start a conversation

[1638] After the information is sent to the server, the user can initiate everyday conversations such as "How was your day?"

[1639] Specific examples

[1640] Example 1: Everyday conversation

[1641] The user launches the application and enters information about their mother. For example, the "mother"'s personality could be "kind," her voice characteristic "high-pitched," and her relationship with the virtual mother "understanding." If the user talks about a difficult time at work, the virtual mother responds, "Really? What happened?" If the user continues, "Your boss scolded me," the virtual mother empathizes, saying, "That must have been tough. But you did a great job." The user's emotions are then analyzed, and the AI ​​adjusts the expressions of empathy and encouragement appropriately.

[1642] Example 2: Consultation

[1643] The user inputs information about a friend and sets the "friend's" personality to "energetic," the voice characteristics to "mid-tone," and the relationship to "good at encouraging people." When the user says, "I've been feeling down lately," the virtual friend will respond with an accepting attitude, saying, "What happened? Tell me." When the user says, "A lot of things have been happening...," the emotion engine analyzes the user's emotional state, and based on that information, the virtual friend will dynamically adjust the words of encouragement and respond, saying, "That's tough, but I know you can get through it!"

[1644] Program processing

[1645] Server-side processing

[1646] The server receives and analyzes input information and emotion data from the user. It trains an AI model based on the analysis results and works with the emotion engine to simulate a conversation with the user. The results are returned to the user's device via an API endpoint.

[1647] Terminal side processing

[1648] The device collects input information and emotion data from the user and sends it to the server, receives responses from the server and displays them to the user in text or audio format, and saves the conversation log and sends it to the server for later reuse.

[1649] Through this operational flow and each function, the system provides consistent psychological support to users, realizing a natural and empathetic conversation experience.

[1650] The processing flow will be explained below.

[1651] Program processing flow

[1652] Server-side processing

[1653] Step 1:

[1654] The server receives information about family, friends, and pets entered by the user, including their names, personalities, vocal characteristics, and relationships.

[1655] Step 2:

[1656] The server analyzes the received information and classifies the personality and voice characteristics into specific categories. For example, personality can be classified as "friendly," "strict," or "energetic," and voice characteristics can be classified as "gentle," "mid-range," or "high-pitched."

[1657] Step 3:

[1658] The server extracts important features from each piece of analyzed information, allowing the AI ​​to more accurately recreate the personalities and characteristics of family, friends, and pets.

[1659] Step 4:

[1660] The server trains an AI model based on the feature-extracted data using machine learning algorithms (e.g., natural language processing models).

[1661] Step 5:

[1662] The server uses the trained AI model to simulate a conversation with the user, using natural language processing based on the user's input.

[1663] Step 6:

[1664] The server uses an emotion analysis engine to analyze the user's emotions from their voice, text, and facial expression data. For example, it can identify emotional states such as "angry," "sad," or "happy" from their voice.

[1665] Step 7:

[1666] The server adjusts the content and tone of the AI ​​model's response based on the results of the emotion analysis, generating an appropriate response that matches the emotions the user is feeling.

[1667] Step 8:

[1668] The server transmits data including the tailored conversation response to the user terminal.

[1669] Terminal side processing

[1670] Step 1:

[1671] The user launches the application on their device and enters their authentication information on the login screen. The first time they log in, they are required to create a new account.

[1672] Step 2:

[1673] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[1674] Step 3:

[1675] The device sends the entered information to the server using HTTPS, ensuring secure communication.

[1676] Step 4:

[1677] The device collects voice, text, and facial expression data and transmits it to a server in real time. Voice data is collected through the device's microphone, and facial expression data is collected using the device's camera.

[1678] Step 5:

[1679] The device receives a response from the server, which is the content of the conversation (such as text data or voice data) generated by sentiment analysis and the trained AI model.

[1680] Step 6:

[1681] The device displays the received response to the user. In addition to displaying it in text format, it can also respond verbally using speech synthesis. The device uses a speech synthesis library or service (e.g., Google Text-to-Speech).

[1682] Step 7:

[1683] The device will store conversation logs between the user and their virtual family, which will then be sent to a server for analysis and to improve the AI ​​model.

[1684] User Actions

[1685] Step 1:

[1686] The user launches the app and enters their credentials on the login screen. The first time they log in, they are required to create a new account.

[1687] Step 2:

[1688] Users are then presented with a screen where they can enter information about their family, friends, and pets, including details such as their names, personality traits, vocal characteristics, and relationships.

[1689] Step 3:

[1690] The user initiates a conversation, for example, "How was your day?" and their virtual family member responds, "You did a great job today, did anything special happen?"

[1691] Step 4:

[1692] The user expresses their emotions. For example, if the user says, "I've been feeling down lately," the emotion analysis engine analyzes the user's tone of voice and facial expressions, and the server generates an appropriate response. As a result, the virtual family member responds with empathy, saying, "That's tough. Can you tell me more about what happened? I hope it makes you feel a little better."

[1693] In this way, the system provides psychological support to users at each step, enabling a natural and empathetic conversational experience.

[1694] Example 2

[1695] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1696] In modern society, the number of individuals suffering from loneliness, isolation, and anxiety is increasing, and effective methods to achieve psychological stability are needed. However, existing technologies lack systems that provide customized conversations based on individual emotions and personalities, resulting in users being unable to receive sufficient empathy or support. In particular, there is a lack of systems that combine the functionality of emotion analysis and dynamically adjusting appropriate responses, resulting in an insufficient quality of psychological support for users.

[1697] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving information about family, friends, and pets input by the user, means for analyzing the received information and classifying the personalities and vocal characteristics of the family, friends, and pets, means for training an AI model based on the analyzed data, means for conducting a conversation simulation with the user using the trained AI model, means for collecting the user's voice, text, and facial expression data and analyzing emotions, means for adjusting the content and tone of the conversation based on the analyzed emotional data, and means for setting an API endpoint that allows access from the user terminal. This makes it possible to provide consistent psychological support to the user and realize a natural and empathetic conversation experience.

[1698] "Users" refer to individuals who use the system, and who often experience loneliness, isolation, and worries.

[1699] "Information" refers to data about family, friends, and pets, including names, personalities, vocal characteristics, and relationships.

[1700] "Receiving" refers to the process by which the server retrieves information sent from the user terminal.

[1701] "Analysis" refers to the process of analyzing received information and understanding and classifying its content.

[1702] "Personality" is data that indicates the behavioral characteristics and attitudes of humans and pets, and is used in this system for emotion analysis and conversation simulation.

[1703] "Voice characteristics" are data that indicate the pitch, tone, intonation, etc. of a voice, and are used to reproduce the characteristics of the voices of family members, friends, and pets entered by the user.

[1704] "Classification" is the process of dividing information into multiple categories and finding similarities and differences based on each person's personality and voice characteristics.

[1705] "Training" is the process by which an AI model learns from data to produce more accurate and efficient outputs.

[1706] An "AI model" is a collection of algorithms that use artificial intelligence to make predictions or generate results, and operates based on trained data.

[1707] "Conversational simulation" refers to the process of using a trained AI model to simulate a conversation with a user.

[1708] "Emotion analysis" is a technology that infers emotions from a user's voice, text, facial expressions, etc., and is used by the system to generate more natural and empathetic responses.

[1709] "Tone" refers to the intonation and timbre of a voice, and is an element that determines the atmosphere in which the content of a conversation is conveyed.

[1710] "Tuning" is the process of changing the content and tone of the conversation based on the analyzed data to provide an appropriate and empathetic response to the user.

[1711] An "API endpoint" refers to a specific URL or URI through which a user device and a server communicate, and through which data is sent and received.

[1712] "Terminal" refers to a device used by a user to perform operations, including smartphones and tablets.

[1713] A "conversation log" is data that records the history of conversations between a user and an AI model, and can be referenced and reused later.

[1714] This invention is a system that uses AI technology to enable individuals who are feeling lonely, isolated, or worried to have everyday conversations and seek advice from a virtual family member, in order to achieve psychological stability.The system incorporates an emotion engine that recognizes the user's emotions, providing a more natural and empathetic conversation experience.

[1715] System Configuration

[1716] Server-side configuration

[1717] The server has the following features:

[1718] Data reception function: Receives information about family, friends, and pets sent from the user's device. Specifically, it receives JSON format data via HTTP requests.

[1719] Data analysis function: Analyzes received information and classifies personality and voice characteristics. Uses NLP libraries to perform text analysis and extract personality attributes.

[1720] Feature extraction function: Extracts important features from the analyzed data and stores them in a database. These include personality traits, voice features, etc.

[1721] AI model training function: Trains generative AI models based on feature extraction data. Specific software used includes TensorFlow and PyTorch.

[1722] Conversation Simulation: A trained AI model is used to simulate conversations with users, with the virtual family generating prompts such as "How was your day?"

[1723] Emotion analysis function: Analyzes emotions from the user's voice, text tone, facial expressions, etc. It uses a voice recognition engine to identify positive, negative, and other emotions.

[1724] Conversation adjustment function: The content and tone of the conversation are adjusted based on the analyzed emotional data. For example, it may give gentle advice such as "You should take it easy today."

[1725] API endpoint setting function: Sets the trained AI model and emotion engine as an API endpoint accessible from the device. Sets a RESTful API and notifies the user device of the endpoint URL.

[1726] Terminal configuration

[1727] The terminal has the following features:

[1728] Information input screen: A screen is displayed where users can enter information about their family, friends, and pets. A smartphone application is used.

[1729] Data transmission function: Sends information entered by the user to the server. The entered text data is sent to the server via an HTTP POST request.

[1730] Emotion data collection function: Collects user voice, text, and facial expression data using the device's microphone and camera.

[1731] Response reception function: Receives responses from the server and displays them to the user. Retrieves JSON format responses from the server's API, parses them, and displays them in the UI.

[1732] Content display function: Displays the received response to the user. Not only in text format, but also in voice response using speech synthesis. Plays it aloud using a speech synthesis engine.

[1733] Conversation log storage function: Save the conversation logs between the user and the virtual family and send them to the server for later reuse. Save the conversation data in a local database and back it up to the server periodically.

[1734] Specific examples

[1735] Example 1: Everyday conversation

[1736] The user launches the application and enters information about their mother. For example, the "mother"'s personality could be "kind," her voice characteristic "high-pitched," and her relationship with the virtual mother "understanding." If the user talks about a difficult time at work, the virtual mother responds, "Really? What happened?" If the user continues, "Your boss scolded me," the virtual mother empathizes, saying, "That must have been tough. But you did a great job." At this point, the emotion engine analyzes the user's emotions, and the AI ​​adjusts the expressions of empathy and encouragement appropriately.

[1737] Example 2: Consultation

[1738] The user inputs information about a friend and sets the friend's personality to "energetic," the voice characteristics to "mid-tone," and the relationship to "good at encouraging others." When the user says, "I've been feeling down lately," the virtual friend will respond with an accepting attitude, saying, "What happened? Tell me." When the user says, "A lot of things have been happening...," the emotion engine analyzes the user's emotions, and based on that information, the virtual friend will dynamically adjust the words of encouragement and respond, saying, "That's tough, but I know you can get through it!"

[1739] Prompt Sentence Examples

[1740] "How was work today?"

[1741] "What's been going on lately?"

[1742] "Really? What happened?"

[1743] "That must have been tough. But you did a great job."

[1744] Through these functions and specific operation flow, this system can provide consistent psychological support to users and realize a natural and empathetic conversation experience.

[1745] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1746] Step 1: Display the information entry screen

[1747] The device displays a screen where the user can enter information about family, friends, and pets. Specifically, the device UI displays text input fields and various selection options. Through these fields, the user enters information such as name, personality, voice characteristics, and relationships. Input data includes the name "Mother," personality "gentle," and voice characteristics "high-pitched."

[1748] Step 2: Send user information to the server

[1749] The device sends the information entered by the user to the server. Specifically, the input data is converted to JSON format and sent to the server using an HTTP POST request. This request includes information such as name, personality, voice characteristics, and relationships. The data is formatted in a way that is easy for the server to receive.

[1750] Step 3: Receiving data

[1751] The server receives the information sent from the user's device. Specifically, the server receives an HTTP request and extracts information such as name, personality, voice characteristics, and relationships from the JSON-formatted data. The received data includes attributes such as "mother," "gentle," and "high-pitched voice."

[1752] Step 4: Data analysis

[1753] The server analyzes the received information and classifies each individual's personality and voice characteristics. Specifically, it uses an NLP library to perform text analysis and extract personality attributes. It also uses voice recognition technology to analyze voice features. The input data is the name, personality, voice characteristics, and relationships, and the output data is the analyzed personality attributes and voice features.

[1754] Step 5: Feature Extraction

[1755] The server extracts important features from the analyzed data and stores them in a database. Specifically, personality traits and voice features are extracted as important features. For example, a personality trait such as "gentle" or a voice feature such as "high-pitched" are extracted and registered in the database.

[1756] Step 6: Training the AI ​​model

[1757] The server then trains the AI ​​model based on the extracted feature data. Specifically, it uses TensorFlow and PyTorch to train the generative AI model while referring to past conversation data. During this process, the system processes a dataset containing personality traits and voice features to improve its ability to generate natural conversations with the user.

[1758] Step 7: Conversation simulation

[1759] The server uses the trained AI model to simulate a conversation with the user. Specifically, it receives a prompt from the user, such as "How was your day?", and the AI ​​model generates an appropriate response. For example, in response to the question, "You had a hard time at work," it generates the response, "Really? What happened?"

[1760] Step 8: Sentiment Analysis

[1761] The server collects the user's voice, text, and facial expression data and analyzes their emotions. Specifically, it uses a voice recognition engine and facial expression analysis software to identify the user's emotional state. For example, it can estimate emotions such as stress or relief from the user's voice data.

[1762] Step 9: Conversation Adjustment

[1763] The server adjusts the content and tone of the conversation based on the analyzed emotional data. Specifically, if the user is tired, it will generate a softer response based on the emotion analysis results, such as "You should get some rest today." This provides a customized response that is optimized for the user's emotions.

[1764] Step 10: Receive a response

[1765] The device receives the response from the server and displays it to the user. Specifically, it parses the JSON format response from the server and displays it as text on the UI. It also plays back the audio using a speech synthesis engine. For example, it provides the user with a response such as "That must have been tough, but you did well" both as text and audio.

[1766] Step 11: Save conversation logs

[1767] The device saves conversation logs between the user and their virtual family and sends them to a server for later reuse. Specifically, each conversation episode is recorded in a local database and periodically backed up to the server. The saved data is used for future conversation simulations and user support.

[1768] (Application example 2)

[1769] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1770] In modern society, the number of individuals feeling lonely or isolated is increasing. While there is a need for methods to provide psychological stability to these individuals, real-life relationships alone are often insufficient. Furthermore, even in virtual shopping experiences, personalized support tailored to individual needs is lacking. Therefore, a system that uses a virtual companion to provide empathetic conversation and advice to individuals who feel lonely or isolated is needed.

[1771] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1772] In this invention, the server includes means for receiving person information input by a user, means for analyzing the received information and classifying the person's personality and voice characteristics, means for training an AI model based on the analyzed data, means for conducting a conversation simulation with the user using the trained AI model, means for setting an API endpoint accessible from a user terminal, and shopping assistance means for a person to provide product advice in a virtual store. This enables individuals who feel lonely or isolated to receive empathetic conversation and personalized shopping support in a virtual environment.

[1773] "User" refers to an individual who uses the system to have conversations with virtual people or receive shopping assistance.

[1774] "Input" refers to information provided by the user to the system, specifically data such as a person's personality and voice characteristics.

[1775] "People" refers to virtual beings such as family, friends, and companions that users recreate within the system.

[1776] "Information receiving means" refers to a function that allows the system to receive person information entered by the user.

[1777] "Analysis means" refers to a function for classifying a person's personality and voice characteristics based on the received information.

[1778] "Training means" refers to the function used to train an AI model based on analyzed data.

[1779] "Conversation simulation means" refers to a function that uses a trained AI model to conduct a conversation between a user and a virtual person.

[1780] "API endpoint setting means" refers to a function for setting an API endpoint that can be accessed from a user terminal.

[1781] "Shopping assistance means" refers to a function that enables a virtual person to provide product advice to a user in a virtual store.

[1782] System configuration

[1783] Server-side processing

[1784] 1. Means of receiving information

[1785] The server receives information about the person sent from the user's device, including their name, personality, voice characteristics, and relationships.

[1786] 2. Analysis method

[1787] The server analyzes the received information and classifies the person's personality and voice characteristics using natural language processing and voice analysis technology.

[1788] 3. Training methods

[1789] The server trains an AI model based on the analyzed data using a deep learning framework (e.g., TensorFlow, PyTorch).

[1790] 4. Conversation Simulation Methods

[1791] The server uses a trained AI model to simulate a conversation with the user, allowing the virtual persona to have a natural conversation.

[1792] 5. API endpoint configuration method

[1793] The server sets up an API endpoint that can be accessed from the user's device, allowing the user to interact with the virtual person at any time.

[1794] 6. Shopping assistance methods

[1795] The server allows a virtual person in the virtual store to provide product advice to the user, based on product information and user reviews.

[1796] Terminal side processing

[1797] 1. Information input screen

[1798] Users enter their personal information on the device screen, which is built using HTML, CSS, and JavaScript.

[1799] 2. Data transmission function

[1800] The terminal sends the input information to the server via a RESTful API.

[1801] 3. Emotion data collection function

[1802] The device uses WebRTC and TensorFlow.js to collect voice and facial expression data and send it to a server.

[1803] 4. Response Receiving Function

[1804] The device receives the response from the server and processes the received data using the JavaScript fetch API or similar.

[1805] 5. Content display function

[1806] The device displays the received response to the user, and can also play the received data aloud using a speech synthesis API (e.g., Google Text-to-Speech API).

[1807] User Actions

[1808] 1. Initial system setup

[1809] Users launch the application and enter their authentication information on the login screen. The first time they log in, they are required to create a new account.

[1810] 2. Enter your information

[1811] Users enter information about virtual characters such as family and friends on a screen where they can enter their names, personality traits, voice characteristics, relationships, etc.

[1812] 3. Start a conversation

[1813] Users can initiate everyday conversations such as "How was your day?" The virtual person will provide natural conversation.

[1814] Specific examples

[1815] During a virtual shopping trip, a user wearing smart glasses can choose new clothes and talk to their virtual family. When the user asks, "What do you think of this outfit?", the virtual mother responds, "It looks great, and the color suits you!"

[1816] Prompt Sentence Examples

[1817] User input: "I've been interested in this outfit lately. What do you think?"

[1818] Generative AI Model Input: The AI ​​model, the user is interested in a casual shirt in a solid blue color. Given that the user's mother has a "kind" personality, generate an empathetic response when the user asks about the shirt.

[1819] The system can provide empathetic conversations and personalized shopping support through a virtual companion to individuals who feel lonely or isolated.

[1820] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1821] Step 1:

[1822] The terminal collects the user's login information on the input screen and sends the authentication information to the server. The input includes the user name and password, and session information is generated as output. The server receives this information, performs authentication, and returns the session information.

[1823] Step 2:

[1824] The user enters information about virtual people, such as family and friends, into the input screen. The input includes name, personality, voice characteristics, and relationships, and is sent as output in JSON format to the server. The device then sends this to the server.

[1825] Step 3:

[1826] The server analyzes the received personal information and classifies personality and voice characteristics using Natural Language Processing (NLP) and voice analysis technology. It processes the JSON data received as input, converts personality and voice characteristics into numerical data, and outputs the classification results.

[1827] Step 4:

[1828] The server trains an AI model based on the classified data. It receives the data classified in the previous step as input, trains the model using a deep learning framework (e.g., TensorFlow, PyTorch), and generates a trained AI model as output.

[1829] Step 5:

[1830] The device uses its emotion data collection function to collect the user's voice and facial expression data. It uses data acquired from the microphone and camera as input, processes it using WebRTC and TensorFlow.js, and outputs analyzed emotion data.

[1831] Step 6:

[1832] The server uses a trained AI model to simulate a conversation between the user and a virtual character. It receives emotional data and the user's questions and comments as input, generates conversation content, and obtains response data as output.

[1833] Step 7:

[1834] The server sends the response data to the user device through the API endpoint, processes the HTTP request via the API using the generated response data as input, and sends the data to the device as output.

[1835] Step 8:

[1836] The response data received by the device is played back to the user as audio using a speech synthesis API. Response data from the server is received as input, converted into audio using a speech synthesis API (e.g., Google Text-to-Speech API), and an audio response is obtained as output.

[1837] Step 9:

[1838] The device saves the conversation log between the user and the virtual person and sends it to the server as reusable data later. It receives the conversation data as input, saves it in LocalStorage, and obtains the saved log data as output.

[1839] Step 10:

[1840] The server generates shopping assistance data for the virtual store, and the virtual person provides product advice to the user. It receives product data and user interests as input, generates advice using an AI model, and sends the advice data to the terminal as output.

[1841] Step 11:

[1842] The terminal presents the received advice data to the user. Using the advice data received from the server as input, the advice is displayed on the screen or via voice, and the advice is provided to the user as output.

[1843] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1844] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1845] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1846] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1847] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1848] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1849] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1850] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1851] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1852] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1853] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1854] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1855] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1856] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 i...

Claims

1. A means for receiving information about family, friends, and pets entered by the user; A means of analyzing the received information and classifying the personalities and vocal characteristics of family, friends, and pets; A means to train AI models based on the analyzed data; and A means to conduct conversation simulations with users using trained AI models; A means of setting up an API endpoint that can be accessed from a user device; A system including:

2. The system of claim 1, which trains an AI model to reproduce the personality and voice of a specific family member, friend, or pet based on information entered by the user.

3. 2. The system according to claim 1, further comprising means for saving a conversation log between the user and the virtual family and transmitting it to the server as reusable data later.

Images