system

Abstract

Provide a system. The method comprises: collecting data on the hobbies, preferences, and conversation styles of a particular individual; a means for analyzing the collected data to create a profile; means for generating a replica of a particular individual based on the profile; a means for interacting with a user using the generated replica; A system including:

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] In modern society, losing a loved one brings about a great deal of mental stress. In particular, when a close family member or friend passes away, the grief and sense of loss can have a long-lasting impact. Conventional methods have mainly used photographs and videos to preserve memories of the deceased, but these do not allow for interactive dialogue and are limited in their ability to provide emotional care. Therefore, a system is needed that can reproduce the hobbies, preferences, and conversation style of the deceased and enable interactive dialogue so that surviving family and friends can receive mental support. [Means for solving the problem]

[0005] The present invention provides a system including a means for collecting data on a specific individual's hobbies, preferences, and conversation style, a means for analyzing the collected data to create a profile, a means for generating a replica of the specific individual based on the profile, and a means for engaging in a dialogue with a user using the generated replica. This enables interactive dialogue with a replica of a deceased person, making it easier for surviving family and friends to receive emotional support. The system also includes a means for analyzing voice data to extract voice characteristics and a means for generating voice responses based on the extracted voice characteristics, providing a more realistic dialogue experience. The system also includes a means for analyzing collected text data to extract language patterns and a means for generating text responses based on the extracted language patterns, thereby reproducing the target individual's language characteristics and enabling natural conversation.

[0006] "Designated individual" refers to a specific person selected as the subject of data collection.

[0007] "Hobbies and preferences" refers to activities and tastes that a particular individual is particularly interested in and enjoys frequently.

[0008] "Conversational style" refers to the linguistic expressions, phrases, tone, and speaking characteristics that a particular individual uses on a daily basis.

[0009] "Data collection means" refers to a method or device for collecting text data, voice data, or other information about a specific individual.

[0010] "Profile creation means" refers to a method or device that analyzes collected data and generates a model that reflects the characteristics of a particular individual.

[0011] "Replica generation means" refers to a method or device that generates an imitation of a specific individual based on a created profile.

[0012] "Means for interacting with the user" refers to a method or device for interactively communicating with surviving family members or friends (users) using the generated replica.

[0013] The term "voice response generating means" refers to a method or device that synthesizes voice based on the voice characteristics of a specific individual and generates a response.

[0014] "Text data analysis" refers to the process of analyzing collected text data and extracting specific individual language patterns and frequently occurring topics.

[0015] "Voice data analysis" refers to the process of analyzing collected voice data to extract the voice characteristics and speaking patterns of a specific individual.

[0016] "Text response generator" refers to a method or device that generates natural-sounding text responses based on analyzed language patterns. [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] The present invention is a system that collects data on the hobbies, preferences, and conversational style of a specific individual, analyzes the data to create a profile, and generates a replica based on this profile. The generated replica can interact with the user and supports both voice and text. Specific embodiments are described below.

[0039] Data collection

[0040] 1. Terminal: Data entry

[0041] Users use devices such as smartphones and personal computers to input text data in various formats, such as diaries, text messages, emails, and posts on social media.

[0042] The user uses an application on the device to record their own voice, and this voice data contains the target individual's characteristic phrasing and tone of voice.

[0043] 2. Terminal: Data transmission

[0044] The device transmits the collected text and voice data to the server over a secure connection, after which the device notifies the user that the data was successfully transmitted.

[0045] Data analysis

[0046] 1. Server: Data storage

[0047] The server stores the received text data and voice data in a database, which is organized and managed for each user.

[0048] 2. Server: Text data analysis

[0049] The server uses morphological and dependency structure analysis to analyze the text data for language patterns, frequent topics, and specific response tendencies, and the analysis results are stored in a database as user profiles.

[0050] 3. Server: Voice data analysis

[0051] The server uses a voice recognition engine to convert the voice data into text, which is then used for analysis.

[0052] The target individual's vocal characteristics are modeled through spectral analysis, which extracts the frequency components and intonation of the voice.

[0053] Replication

[0054] 1. Server: Chatbot generation

[0055] Based on the results of analyzing the text data, the server generates a chatbot that mimics the conversational style and language patterns of the target individual, and is capable of generating natural responses in everyday conversations and on specific topics.

[0056] 2. Server: Building a speech synthesis model

[0057] The server trains the speech synthesis engine based on the analysis of the voice data and builds a speech synthesis model that reproduces the voice characteristics of a specific individual, allowing it to convert text responses into voice responses.

[0058] User Interaction

[0059] 1. User: Initiate conversation

[0060] Surviving family members and friends can then interact with the replica through a dedicated application, for example by typing, "What should I cook today?"

[0061] 2. Device: Send Message

[0062] The terminal sends this input message to the server.

[0063] 3. Server: Response Generation

[0064] The chatbot on the server generates an appropriate response based on the received message, for example, "How about making curry today?" in text and voice.

[0065] 4. Terminal: Receives response

[0066] The terminal displays and plays the text and voice responses received from the server, allowing the user to confirm and proceed with the next dialogue.

[0067] Specific examples

[0068] For example, if a particular individual frequently talks about movies they liked in the past, then based on that conversation log, when a user asks, "What movie have you seen recently?", the server's chatbot can generate a response such as, "You used to like action movies, so check out these new action movies!"

[0069] This system allows surviving family and friends to receive emotional support by interacting with a replica that faithfully reproduces the individual's hobbies, preferences, and conversation style.

[0070] The processing flow will be explained below.

[0071] Step 1:

[0072] Users use the device's applications to enter information about their diary, conversation logs, and hobbies and preferences, including text messages, emails, and social media posts.

[0073] Users can also use the device's recording function to record voice data, including everyday conversations and voice messages on specific topics.

[0074] Step 2:

[0075] The device sends the collected text and voice data to a server over a secure connection, and the data is encrypted before transmission.

[0076] After the transmission is complete, the terminal notifies the user that the data has been successfully transmitted.

[0077] Step 3:

[0078] The server stores the received text data and voice data in a database.

[0079] The database is divided by user, and each user's data is organized and managed.

[0080] Step 4:

[0081] The server analyzes the collected text data using morphological analysis and dependency structure analysis.

[0082] Specifically, it identifies the parts of speech of words in a sentence, analyzes the structure of the sentence, and extracts frequently occurring words, phrases, and frequently discussed topics, thereby identifying the user's language patterns.

[0083] Step 5:

[0084] The server converts the voice data into text data using a voice recognition engine.

[0085] The converted text data is also analyzed and stored in a database. The server also performs spectral analysis to extract the frequency components of the voice and the features of each phoneme.

[0086] Step 6:

[0087] The server then creates a user profile based on the analysis results, which includes the user's language patterns, interests, and vocal characteristics.

[0088] The profile is stored in a database for later processing.

[0089] Step 7:

[0090] The server generates a chatbot that learns the language patterns of the target user based on the user profile.

[0091] The chatbot is configured to respond to a user's specific phrases and topical trends.

[0092] Step 8:

[0093] Based on the results of the voice data analysis, the server trains the voice synthesis engine and builds a voice synthesis model that reproduces the user's voice characteristics.

[0094] This speech synthesis model is used to convert the generated text response into a speech response.

[0095] Step 9:

[0096] Users can initiate interactions with the created replicas through dedicated applications.

[0097] When the user initiates a conversation, the terminal sends the user's message to the server.

[0098] Step 10:

[0099] The chatbot on the server generates an appropriate response based on the received message.

[0100] For example, if a user asks, "What should I cook today?" the server generates a text response saying, "How about making curry?"

[0101] Step 11:

[0102] The server converts the generated text response into voice using a speech synthesis model.

[0103] This causes the text response to be output as a voice response.

[0104] Step 12:

[0105] The terminal displays and plays back the text and audio responses received from the server.

[0106] The user can then confirm the response and continue the dialogue.

[0107] Example 1

[0108] 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."

[0109] In modern society, after a specific individual passes away, there is a need to build an interactive dialogue system based on that individual's hobbies, preferences, and conversational style. However, there is still a lack of technology to efficiently collect available data, accurately analyze it, and conduct natural dialogue. The purpose of this invention is to solve this problem and provide a system that generates a replica based on the hobbies, preferences, and conversational style of a specific individual, enabling dialogue with the user.

[0110] 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.

[0111] In this invention, the server includes: means for a user to input text data and voice data using a terminal; means for transmitting the input text data and voice data to the server via a secure connection; means for the server to store the received text data and voice data in a database; means for the server to analyze the text data using morphological analysis and dependency structure analysis to extract language patterns and frequent topics; means for the server to convert the voice data into text using a voice recognition engine and model voice characteristics through spectral analysis; means for the server to generate a chatbot and a voice synthesis model that imitate the conversation style and language patterns of a specific individual based on the analysis results of the text data and voice data; and means for engaging in a dialogue with the user using the generated chatbot and voice synthesis model. This makes it possible to faithfully reproduce the hobbies, preferences, and conversation style of a specific individual and realize natural, interactive dialogue.

[0112] A "terminal" is an electronic device used by a user, such as a smartphone or a personal computer.

[0113] "Text data" refers to data of sentences or character strings that a user inputs using a terminal.

[0114] "Audio data" refers to data of audio recorded by a user using a terminal.

[0115] A "secure connection" is an encrypted communication method that protects data from being intercepted or tampered with by third parties during transmission.

[0116] A "server" is a computer system connected to the Internet that receives, analyzes, and stores text and audio data.

[0117] A "database" is a data structure for efficiently organizing and managing data within a server.

[0118] "Morphological analysis" is a natural language processing technique that divides text data into words and identifies their parts of speech in order to understand their meaning.

[0119] "Dependency structure analysis" is a natural language processing technique that analyzes the relationships between words in text data and reveals the structure of sentences.

[0120] A "speech recognition engine" is a software or hardware system for converting voice data into text data.

[0121] "Spectral analysis" is a technique for analyzing the frequency components and intonation of voice data.

[0122] "Voice characteristics" refer to the unique characteristics of individual voices, including, for example, tone, pitch, and intonation.

[0123] A "chatbot" is an automated response system that uses text data to interact with users.

[0124] A "voice synthesis model" is a model for converting text data into voice data based on analyzed voice characteristics.

[0125] A "profile" is information generated based on collected data that summarizes a specific individual's hobbies, preferences, and conversation style.

[0126] The present invention is a system that collects data on the hobbies, preferences, and conversational style of a specific individual, analyzes the data to create a profile, and generates a replica based on this profile. The system analyzes text data and voice data entered by a user using a terminal, and generates a chatbot and a voice synthesis model that mimics the conversational style and language patterns of the specific individual. The following describes how to specifically implement the present invention.

[0127] Data Collection and Transmission

[0128] Users use devices such as smartphones and PCs to input text data in various formats, such as diaries, text messages, emails, and social media posts. For example, a user might input "It was a nice day today" into a dedicated smartphone app. The user might also use an application on the device to record their own voice. This voice data contains the target individual's characteristic phrasing and tone of voice. For example, a user might record and save the voice saying "I slept well today."

[0129] The device sends the collected text data and voice data to the server via a secure connection (such as SSL / TLS). For example, the device sends text data such as "The weather was nice today" and voice data such as "I slept well today" to the server. After sending, the device notifies the user that the data has been sent successfully. For example, the device displays a message on the screen saying "The data has been sent successfully."

[0130] Data Storage and Analysis

[0131] The server stores the received text data and voice data in a database. This database is a data structure for organizing and managing data for each user. For example, the server stores the text data "The weather was nice today" and the voice file "I slept well today" in the "user123" folder.

[0132] The server then uses morphological analysis and dependency structure analysis to analyze the text data for language patterns, frequently occurring topics, and specific response trends. The software used includes the Python morphological analysis library "MeCab" and the dependency structure analysis engine "spaCy." For example, the server analyzes "The weather was nice today" and recognizes that "weather" is a frequently occurring topic and that it is a pleasant day. The analysis results are saved in a database as a user profile. For example, the server can add information to the profile such as "I have many good experiences with the weather."

[0133] The server also uses a speech recognition engine to convert the voice data into text. For this, it uses "Google (registered trademark) Cloud Speech-to-Text." For example, the voice data "I slept well today" is converted into text. Furthermore, it performs spectral analysis to extract the frequency components and intonation of the voice. For this, it uses "Praat." For example, the server analyzes and models the intonation and pitch of the voice "I slept well today."

[0134] Replication and User Interaction

[0135] Based on the results of analyzing the text data, the server generates a chatbot that mimics the target individual's conversation style and language patterns. The generative AI model used is "OpenAI (registered trademark) GPT-4 (registered trademark)." For example, based on the information that the server has "many positive experiences related to the weather," it generates a chatbot that prefers topics related to sunny days.

[0136] The server then trains a speech synthesis engine based on the results of the voice data analysis to build a speech synthesis model that can reproduce the voice characteristics of a specific individual. Engines used include Google Cloud Text-to-Speech and Amazon Polly. For example, the server builds a voice model that can reproduce the tone and tempo of the phrase "I slept well tonight."

[0137] Surviving family and friends can start a conversation with the created replica through a dedicated application. For example, a family member can type, "What should we cook today?" The device sends this input message to the server. The chatbot on the server generates an appropriate response based on the received message. For example, it might generate a response in text and voice, such as, "How about cooking curry today?" The device displays and plays back the text and voice response received from the server. The user can confirm this and carry out the next conversation.

[0138] For example, if a particular individual frequently talked about their favorite movies in the past, when a user asks "What movie did you see recently?" based on the conversation log, the chatbot on the server can generate a response such as "You used to like action movies, so check out the new action movies!" In this way, surviving family and friends can receive emotional support by interacting with a replica that faithfully reproduces the hobbies, preferences, and conversation style of the particular individual.

[0139] Example prompt sentence:

[0140] "What activities have you been interested in lately?"

[0141] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0142] Step 1:

[0143] A user inputs text data and voice data using a device. Using an application on a smartphone or computer, the user inputs text data in the form of a diary entry, text message, email, or social media post. The user also uses an application to record their own voice. For example, a user inputs text data such as "It was nice weather today" into a dedicated smartphone application and records voice such as "I slept well today." The input data is text data and voice data.

[0144] Step 2:

[0145] The device sends the collected text and voice data to the server through a secure connection. Specifically, the data is sent using encrypted communication such as SSL / TLS. When the transmission is complete, the device displays a notification to the user stating, "The data was sent successfully." The input data is the text and voice data entered by the user, and the output data is a notification indicating that the data has been sent.

[0146] Step 3:

[0147] The server stores the received text data and voice data in a database. Specifically, the server saves the text data "It was nice weather today" and the voice file "I slept well today" in the "user123" folder. The input data is the transmitted text data and voice data, and the output data is the data stored in the database.

[0148] Step 4:

[0149] The server analyzes the text data using morphological analysis and dependency structure analysis. Specifically, it uses the Python morphological analysis library "MeCab" and the dependency structure analysis engine "spaCy" to extract language patterns, frequently occurring topics, and specific reaction trends from the text data. For example, the server analyzes the text "It was nice weather today" and recognizes that "weather" is a frequently occurring topic. The input data is text data, and the output data is the analyzed language patterns and frequently occurring topics.

[0150] Step 5:

[0151] The server converts the voice data into text using a speech recognition engine. Specifically, it uses "Google Cloud Speech-to-Text" to convert the voice data "I slept well today" into text. It also performs spectral analysis to extract the frequency components and intonation of the voice. For example, using "Praat," the server analyzes and models the intonation and pitch of the voice "I slept well today." The input data is the voice data, and the output data is the analyzed voice characteristics and converted text data.

[0152] Step 6:

[0153] Based on the analysis results of the text and voice data, the server generates a chatbot and voice synthesis model that mimics the conversation style and language patterns of a specific individual. Specifically, it uses the generative AI model "OpenAI GPT-4" to generate a chatbot based on the text data, and builds a voice model using the voice synthesis engines "Google Cloud Text-to-Speech" and "Amazon Polly." The input data is the analysis results, and the output data is the chatbot and voice synthesis model.

[0154] Step 7:

[0155] The user initiates a conversation with the generated replica through a terminal. Specifically, the user inputs a message into a dedicated application, such as "What should I cook today?" The input message is sent to the server, and the chatbot on the server generates a response in text and voice, such as "How about making curry?" The input data is the message from the user, and the output data is the text and voice response from the chatbot.

[0156] Step 8:

[0157] The terminal displays and plays back the text and voice response received from the server. Specifically, it displays the chatbot's response text on the terminal's display and plays back the response aloud. For example, the terminal displays the text "How about making curry?" on the screen and plays back the voice. This allows the user to carry out the next dialogue. The input data is the text and voice response from the server, and the output data is the displayed text and played back voice.

[0158] An example of a prompt sentence: "What activities have you been interested in lately?"

[0159] (Application example 1)

[0160] 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."

[0161] In today's brick-and-mortar stores, it is difficult to provide personalized product recommendations and assistance to each individual consumer. Conventional services often provide general information and uniform customer service to all customers, failing to address individual preferences and needs. This can result in low user satisfaction and a loss of purchasing motivation. Furthermore, there is a lack of technology to effectively utilize profiles that utilize a specific individual's conversation style and preferences using a visual display device. Therefore, there is a need for the development of a system that can provide personalized product recommendations and assistance based on a specific individual's profile through a visual display device.

[0162] 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.

[0163] In this invention, the server includes means for collecting data on the hobbies, preferences, and conversation style of a specific individual, means for analyzing the collected data to create a profile, means for generating a replica of the specific individual based on the profile, means for interacting with the user using the generated replica, and means for providing customized product suggestions and assistance based on the analyzed profile using a visual display device attached to the user's communication device. This makes it possible to effectively provide personalized product suggestions and assistance based on the profile of a specific individual in a physical store.

[0164] "Hobbies" are preferences regarding the things and activities that a particular individual likes.

[0165] A "conversational style" is the linguistic expression or method of communication that a particular individual normally uses.

[0166] A "data collection instrument" is a device or method for collecting information about the tastes, preferences, and speaking styles of a particular individual.

[0167] A "profile" is a collection of information that describes the characteristics of a particular individual, created based on collected data.

[0168] A "replica" is an interactive virtual assistant generated based on a specific individual's profile.

[0169] The "interaction means" is a method or device for communicating with a user using the generated replica.

[0170] A "visual display device" is a communication device worn by a user that visually displays text and images.

[0171] "Product Recommendation" is a method or device aimed at recommending personalized products and services based on a user's profile.

[0172] "Assistance" refers to help or support provided based on a specific individual's preferences and conversational style.

[0173] This invention is a "personalized shopping assistant" system that provides personalized product recommendations and assistance in brick-and-mortar stores. It uses a visual display device attached to a user's communication device. Specifically, it is implemented using smart glasses. The system of this invention suggests products and services based on the user's tastes, preferences, and conversation style, providing an experience tailored to individual needs.

[0174] System Configuration and Operation

[0175] 1. Hardware and Software:

[0176] Hardware: Smart glasses (with voice recognition microphone and speaker), server, microphone

[0177] Software: GPT-2, speech recognition library (speech_recognition), speech synthesis engine (pyttsx3), Python

[0178] The smart glasses recognize the user's voice and send the data to the server, which analyzes the data and generates a response based on the user's profile, which is presented visually and audibly to the smart glasses.

[0179] 2. Data collection and analysis:

[0180] The user puts on the smart glasses and speaks their questions or requests. The speech recognition library, speech_recognition, converts the speech into text data.

[0181] The voice and text data sent to the server is used to generate appropriate responses based on the profile using GPT-2.

[0182] 3. Create a profile:

[0183] The server collects data about the user's interests and conversational style and creates a profile based on that data, including the types of products the user likes and conversational habits.

[0184] Through a speech recognition engine and speech synthesis engine, the user's language patterns and voice characteristics are analyzed and reflected in the profile.

[0185] 4. Generate and present the response:

[0186] When a user speaks to the smart glasses, the server creates a prompt based on the content and generates a response using the GPT-2 model.

[0187] The generated text response is converted to speech using the speech synthesis engine pyttsx3 and played through the smart glasses' speakers, and also displayed as text on the visual display.

[0188] Examples:

[0189] For example, if a user speaks to the smart glasses, "What kind of food would you like today?", the microphone in the smart glasses will capture the voice, and the speech recognition library, speech_recognition, will convert it into text. This text data will be sent to the server and processed as the following prompt sentence:

[0190] User: What would you like to eat today?

[0191] Assistant: The weather is nice today, why not enjoy a barbecue?

[0192] The responses are converted to speech and played back to the user, and are also displayed as text on a visual display, allowing for personalized product recommendations and assistance based on the user's preferences and conversational style.

[0193] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0194] Step 1:

[0195] User: Voice Input

[0196] Users wear smart glasses and input questions or requests by voice.

[0197] Input: User's voice data

[0198] Output: Audio data is captured

[0199] Specific operation: A microphone installed in the smart glasses collects the user's voice.

[0200] Step 2:

[0201] Device: Voice recognition

[0202] The device uses the speech recognition library speech_recognition to convert the voice data into text data.

[0203] Input: Captured audio data

[0204] Output: Text data

[0205] Specific operation: Audio data collected from the microphone is sent to the speech recognition library, and corresponding text data is generated.

[0206] Step 3:

[0207] Terminal: Data transmission

[0208] The terminal transmits the converted text data to the server over a secure connection.

[0209] Input: Text data

[0210] Output: Text data sent to the server

[0211] Specific operation: The converted text data is encrypted and sent securely to the server.

[0212] Step 4:

[0213] Server:Profile Creation

[0214] The server analyzes the received text data and creates a profile of the user's interests, preferences and conversation style.

[0215] Input: Text data

[0216] Output: User profile

[0217] Specific operation: Through morphological analysis and dependency structure analysis, language patterns and preferences are extracted from text data and a profile is generated.

[0218] Step 5:

[0219] Server: Response Generation

[0220] The server uses the generative AI model GPT-2 to create appropriate prompts and generate responses based on the user's profile.

[0221] Input: Profile and text data

[0222] Output: Text response

[0223] How it works: A prompt sentence is input into the GPT-2 model, which generates a text response based on the user's interests and preferences.

[0224] Example prompt sentence:

[0225] User: What would you like to eat today?

[0226] Assistant: The weather is nice today, why not enjoy a barbecue?

[0227] Step 6:

[0228] Server: Speech synthesis

[0229] The server converts the text response into speech using the speech synthesis engine pyttsx3.

[0230] Input: Text response

[0231] Output: Audio data

[0232] What it does: The generated text response is fed into a speech synthesis engine to generate a voice that sounds similar to a human voice.

[0233] Step 7:

[0234] Terminal: Response presentation

[0235] The terminal plays the generated audio data through the speaker of the smart glasses and displays the text data on a visual display.

[0236] Input: Audio and text data

[0237] Output: Presents the response to the user

[0238] Specific Actions: Provide a response to the user by playing audio data over a speaker and displaying text data on a visual display device.

[0239] 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.

[0240] The present invention is a system that collects data on a specific individual's hobbies, preferences, and conversation style, analyzes the data, and creates a profile. It then generates a replica based on the profile, and combines this with an emotion engine that recognizes the user's emotions. The generated replica can converse with the user and adjusts its responses according to the user's emotional state. Specific embodiments are described below.

[0241] Data collection

[0242] 1. Terminal: Data entry

[0243] Users use devices such as smartphones and personal computers to input text data in various formats, such as diaries, text messages, emails, and posts on social media.

[0244] The user uses an application on the device to record their own voice, and this voice data contains the target individual's characteristic phrasing and tone of voice.

[0245] 2. Terminal: Data transmission

[0246] The device transmits the collected text and voice data to the server over a secure connection, after which the device notifies the user that the data was successfully transmitted.

[0247] Data analysis

[0248] 1. Server: Data storage

[0249] The server stores the received text data and voice data in a database.

[0250] Within the database, data is organized and managed for each user.

[0251] 2. Server: Text data analysis

[0252] The server uses morphological and dependency structure analysis to analyze the text data for language patterns, frequent topics, and specific response tendencies, and the analysis results are stored in a database as user profiles.

[0253] 3. Server: Voice data analysis

[0254] The server uses a voice recognition engine to convert the voice data into text, and based on this, the voice data is also used for analysis.

[0255] The server performs spectral analysis to extract the frequency components and intonation of the voice, thereby modeling the voice characteristics.

[0256] Replication

[0257] 1. Server: Chatbot generation

[0258] Based on the results of analyzing the text data, the server generates a chatbot that has learned the target individual's language patterns and is capable of generating natural responses in everyday conversations and on specific topics.

[0259] 2. Server: Building a speech synthesis model

[0260] The server trains the speech synthesis engine based on the analysis of the voice data and builds a speech synthesis model that reproduces the voice characteristics of a specific individual, allowing it to convert text responses into voice responses.

[0261] Adding an Emotion Engine

[0262] 1. Server: User emotion recognition

[0263] The server uses an emotion engine to recognize emotions from the user's voice and text data, for example by analyzing voice characteristics such as tone, speed, and volume, and from text data by analyzing the context and emotional tendencies of selected words.

[0264] 2. Server: Emotion-based profile update

[0265] Based on the recognized emotions, the user profile is updated, which includes the user's emotional tendencies and current emotional state.

[0266] 3. Server: Emotion-based response generation

[0267] The generated response is tailored based on the emotional state recognized by the emotion engine. For example, if the user asks a question in a sad voice, the server will generate a more empathetic and comforting response.

[0268] User Interaction

[0269] 1. User: Initiate conversation

[0270] Family and friends can then interact with the replica through a dedicated application, which sends text and voice messages to the server.

[0271] 2. Device: Send Message

[0272] The terminal transmits the input text message and voice data to the server.

[0273] 3. Server: Response Generation

[0274] The server's chatbot generates an appropriate response based on the received message and emotional state. For example, if the user says "Today was tough" in a tired voice, the server generates a response like "Today was tough. Let's take a break."

[0275] 4. Server: Speech synthesis

[0276] The server converts the generated text response into voice using a speech synthesis model, which outputs the text response as a voice response.

[0277] 5. Terminal: Receives response

[0278] The terminal displays and plays back the text and voice responses received from the server, allowing the user to confirm the responses and continue the dialogue.

[0279] Specific examples

[0280] For example, if a specific individual has often talked about "movies" in the past, when the user asks "What movie did you see recently?", the chatbot on the server will generate a response such as "There was a new action movie recently that was good. You should definitely check it out" based on the user's preferences and emotional state. Also, by recognizing the user's emotional state, if the user is feeling down, for example, an encouraging response such as "Cheer up. Let's find a movie we can enjoy together" will be generated.

[0281] This system allows surviving family and friends to receive emotional support by interacting with a replica that faithfully reproduces the individual's hobbies, preferences, and conversation style. Furthermore, the emotional engine recognizes the user's emotional state and generates responses accordingly, enabling more empathetic and natural conversations.

[0282] The processing flow will be explained below.

[0283] Step 1:

[0284] Users can use the device's applications to enter information about their diary, conversation log, and hobbies and interests in text format, and also record voice messages using the device's recording function.

[0285] Step 2:

[0286] The device sends the collected text and voice data to the server over a secure connection, and the user is notified once the data has been sent.

[0287] Step 3:

[0288] The server stores the received text data in a database, and simultaneously stores the audio data in the database.

[0289] Step 4:

[0290] The server analyzes the received text data using morphological and dependency analysis to identify the user's language patterns and frequent topics and create a profile.

[0291] Step 5:

[0292] The server converts the voice data into text data using a speech recognition engine, and then performs spectral analysis to extract features for each frequency component and phoneme.

[0293] Step 6:

[0294] The server stores a user profile based on the analysis results in a database, which includes language patterns, interests, and voice characteristics.

[0295] Step 7:

[0296] The server generates a chatbot based on the profile, which is trained to replicate the language patterns of the specific individual and to engage in natural conversations.

[0297] Step 8:

[0298] The server uses the speech analysis results to train a speech synthesis engine and build a speech synthesis model that replicates the voice characteristics of a particular individual, which is then used to convert the text response into a speech response.

[0299] Step 9:

[0300] The user initiates a dialogue with the created replica through a dedicated application. At the beginning of the dialogue, the terminal sends the user's input to the server.

[0301] Step 10:

[0302] The server uses an emotion engine to recognize emotions from user input data, analyzing voice tone, speed, volume, etc. from voice data, and emotional trends from text data.

[0303] Step 11:

[0304] The server's chatbot generates more empathetic and appropriate responses based on the emotions recognized by the emotion engine. For example, if the user speaks in a sad voice, the server will generate a comforting response.

[0305] Step 12:

[0306] The server converts the generated text response into voice using a speech synthesis model, which outputs the text response as a voice response that reflects the appropriate emotion.

[0307] Step 13:

[0308] The device displays and plays the text and voice responses received from the server to the user, who can then confirm the responses and continue the dialogue.

[0309] Example 2

[0310] 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."

[0311] In modern society, emotional support gained through memories and conversations with specific individuals is highly valued. However, when a specific individual has passed away or is far away, means of sharing conversations and memories with that person are limited. Conventional technologies simply play back text or audio data, and do not realize natural conversations that reflect the user's emotions, hobbies, and preferences. Therefore, the present invention calls for a new system that faithfully reproduces the hobbies, preferences, and conversation style of a specific individual, enabling conversations that correspond to the user's emotional state.

[0312] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes: means for collecting data on the hobbies, preferences, and conversation style of a specific individual; means for encrypting the collected text data and voice data and transmitting it through a secure connection; means for analyzing the collected data to create a user profile; means for generating a chatbot and a voice synthesis model for the specific individual based on the user profile; means for recognizing the user's emotions using an emotion engine and updating the profile; and means for interacting with the user using the generated replica and adjusting responses based on the user's emotional state. This makes it possible to interactively interact with a replica that faithfully reproduces the characteristics of a specific individual and obtain responses that correspond to the user's emotional state.

[0313] 1. "Specified Individual" refers to a specific person about whom data is collected and analyzed by the system.

[0314] 2. "Preferences" refers to information about a particular individual's preferred activities, interests, and preferences.

[0315] 3. "Conversational style" refers to a particular individual's speaking style, phrasing, language patterns, and communication characteristics.

[0316] 4. "Data Collection Instrument" means the part of the system that captures text and voice data about a specific individual.

[0317] 5. "Data transmission means" refers to the function for encrypting collected data and transmitting it securely to the server.

[0318] 6. "Data analysis means" refers to the part of the system that analyzes collected data to extract and understand hobbies, preferences, and conversation styles.

[0319] 7. "User profile" refers to information that reproduces the characteristics of a specific individual, generated based on analyzed hobbies, preferences, and conversation style.

[0320] 8. "Chatbot generation means" refers to a function for generating a chatbot that reproduces the language patterns of a specific individual based on a user profile.

[0321] 9. "Voice synthesis model generation means" refers to a function for generating a voice synthesis model that reproduces the voice characteristics of a specific individual based on the results of voice data analysis.

[0322] 10. "Emotion engine" refers to software or algorithms that recognize and analyze user emotions.

[0323] 11. "Profile Updater" refers to functionality for updating a user profile based on the user's emotional state recognized by the Emotion Engine.

[0324] 12. "Response adjustment means" refers to a function that enables the generated replica to generate a response according to the emotional state during interaction with the user.

[0325] 13. "Replica" refers to a simulation that faithfully reproduces the tastes, preferences, and conversation style of a specific individual.

[0326] The present invention is a system that collects data on a specific individual's hobbies, preferences, and conversation style, analyzes the data, and creates a profile. A replica is generated based on this profile, and by combining it with an emotion engine that recognizes the user's emotions, a system is provided that can interact with the user. Specific embodiments are described below.

[0327] Data collection

[0328] 1. Terminal: Data entry

[0329] Users use devices such as smartphones and PCs to input various forms of text data, such as diaries, text messages, emails, and posts on social media. Users also use applications on their devices to record their own voice. This voice data includes the target individual's characteristic phrasing and tone of voice.

[0330] 2. Terminal: Data transmission

[0331] The device encrypts the collected text and voice data and sends it to the server over a secure connection, after which the device notifies the user that the data has been successfully sent.

[0332] Data analysis

[0333] 1. Server: Data storage

[0334] The server stores the received text and voice data in a database, where data is organized and managed for each user.

[0335] 2. Server: Text data analysis

[0336] The server uses morphological and dependency structure analysis to analyze the text data for language patterns, frequent topics, and specific response tendencies, and the analysis results are stored in a database as user profiles.

[0337] 3. Server: Voice data analysis

[0338] The server converts the voice data into text using a speech recognition engine. Based on this, the voice data is also analyzed. The server also performs spectral analysis to extract the frequency components and intonation of the voice and model the voice characteristics.

[0339] Replication

[0340] 1. Server: Chatbot generation

[0341] Based on the results of analyzing the text data, the server generates a chatbot that has learned the target individual's language patterns and is capable of generating natural responses in everyday conversations and on specific topics.

[0342] 2. Server: Building a speech synthesis model

[0343] The server trains the speech synthesis engine based on the analysis of the voice data and builds a speech synthesis model that reproduces the voice characteristics of a specific individual, allowing it to convert text responses into voice responses.

[0344] Adding an Emotion Engine

[0345] 1. Server: User emotion recognition

[0346] The server uses an emotion engine to recognize emotions from the user's voice and text data, for example by analyzing voice characteristics such as tone, speed, and volume, and from text data by analyzing the context and emotional tendencies of selected words.

[0347] 2. Server: Emotion-based profile update

[0348] Based on the recognized emotions, the user profile is updated, which includes the user's emotional tendencies and current emotional state.

[0349] 3. Server: Emotion-based response generation

[0350] The generated response is tailored based on the perceived emotional state: for example, if the user asks a question in a sad voice, the server will generate a more empathetic and comforting response.

[0351] User Interaction

[0352] 1. User: Initiate conversation

[0353] Users can start interacting with the replicas through a dedicated application, and when a user initiates an interaction, the device sends text and voice messages to the server.

[0354] 2. Device: Send Message

[0355] The terminal transmits the input text message and voice data to the server.

[0356] 3. Server: Response Generation

[0357] The server's chatbot generates an appropriate response based on the received message and emotional state. For example, if the user says "Today was tough" in a tired voice, the server generates a response like "Today was tough. Let's take a break."

[0358] 4. Server: Speech synthesis

[0359] The server converts the generated text response into voice using a speech synthesis model, which outputs the text response as a voice response.

[0360] 5. Terminal: Receives response

[0361] The terminal displays and plays back the text and voice responses received from the server, allowing the user to confirm the responses and continue the dialogue.

[0362] Specific examples and prompts for the generative AI model

[0363] Specific examples

[0364] For example, if a specific individual has often talked about "movies" in the past, when the user asks "What movie did you see recently?", the chatbot on the server will generate a response such as "There was a new action movie recently that was good. You should definitely check it out" based on the user's preferences and emotional state. Also, by recognizing the user's emotional state, if the user is feeling down, for example, an encouraging response such as "Cheer up. Let's find a movie we can enjoy together" will be generated.

[0365] Prompt Sentence Examples

[0366] 1. Prompt to analyze user text data

[0367] "Analyze text data from diaries, text messages, and social media posts using morphological and dependency analysis. Extract language patterns, frequently occurring topics, and specific reaction tendencies for each user to create a user profile."

[0368] 2. Prompt to request voice data analysis

[0369] "Use a speech recognition engine to convert the speech data into text, then perform spectral analysis to extract the frequency components and intonation of the speech. Based on the results, model the speech characteristics."

[0370] 3. Prompt to generate a chatbot

[0371] "Based on the results of analyzing the text data, create a chatbot that learns the user's language patterns. This chatbot should be able to generate natural responses in everyday conversations and on specific topics."

[0372] 4. Prompt to request emotion engine analysis

[0373] "Analyze voice data for characteristics such as tone, speed, and volume, and text data for context and emotional trends in chosen words. Based on the results, recognize the user's emotions and update the user profile."

[0374] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0375] System program processing flow

[0376] Step 1: Data entry

[0377] Users use smartphones or personal computers to input text data in various formats, such as diary entries, text messages, emails, and social media posts, into dedicated applications.

[0378] Input: Text data (e.g., diary entries, messages, posts)

[0379] Output: Collected text data

[0380] The user uses the same application to record their own voice, and the recorded voice data includes the user's characteristic phrasing and tone of voice.

[0381] Input: Audio data (e.g., recording of everyday conversation)

[0382] Output: Collected audio data

[0383] Step 2: Send data

[0384] The device encrypts the collected text and voice data and sends it to a server over a secure connection.

[0385] Input: Collected text and audio data

[0386] Output: Sending encrypted data to the server

[0387] After transmission, the terminal notifies the user that the data has been successfully transmitted.

[0388] Input: Send result

[0389] Output: User notification

[0390] Step 3: Data storage

[0391] The server stores the received text and voice data in a database, where data is organized and managed for each user.

[0392] Input: Received text and audio data

[0393] Output: Data stored in a database

[0394] Step 4: Text data analysis

[0395] The server performs morphological analysis and breaks down the text data into words.

[0396] Input: Text data

[0397] Output: A list of decomposed words

[0398] The server then uses dependency analysis to analyze grammatical and semantic relationships.

[0399] Input: A list of decomposed words

[0400] Output: Structural data showing grammatical and semantic relationships

[0401] Step 5: Audio data analysis

[0402] The server converts the voice data into text data using a voice recognition engine.

[0403] Input: Audio data

[0404] Output: Character data

[0405] In addition, the server performs spectral analysis to extract the frequency components and intonation of the voice.

[0406] Input: Audio data

[0407] Output: Frequency content and intonation data

[0408] Step 6: Create a user profile

[0409] The server creates a user profile based on the results of the analysis of the text and voice data.

[0410] Input: Analysis results (text data and audio data)

[0411] Output: User profile

[0412] Step 7: Chatbot generation

[0413] The server generates a chatbot that learns the user's language patterns based on the user profile.

[0414] Input: User Profile

[0415] Output: Chatbot

[0416] Step 8: Building a speech synthesis model

[0417] Based on the results of the voice data analysis, the server constructs a voice synthesis model that reproduces the voice characteristics of a specific individual.

[0418] Input: Audio data analysis results

[0419] Output: Speech synthesis model

[0420] Step 9: Emotion Recognition and Profile Update

[0421] The server uses an emotion engine to recognize the user's emotions and reflect them in the profile.

[0422] Input: Voice and text data

[0423] Output: Updated user profile

[0424] Step 10: Start a conversation

[0425] The user initiates a conversation with the created replica through a dedicated application. When the conversation begins, the terminal sends the user's text and voice messages to the server.

[0426] Input: User's message (text and voice)

[0427] Output: Send to server

[0428] Step 11: Send the message

[0429] The terminal transmits the input text message and voice data to the server.

[0430] Input: User message

[0431] Output: Data sent to the server

[0432] Step 12: Response Generation

[0433] The chatbot on the server generates an appropriate response based on the received message and emotional state.

[0434] Input: User message and emotion data

[0435] Output: Response message

[0436] Step 13: Speech synthesis

[0437] The server uses a speech synthesis model to convert the generated text response into a voice response.

[0438] Input: Text response

[0439] Output: Voice response

[0440] Step 14: Receive a response

[0441] The terminal displays and plays back the text and voice responses received from the server, allowing the user to confirm the responses and continue the dialogue.

[0442] Input: Response from the server (text and audio)

[0443] Output: Displayed and played response

[0444] (Application example 2)

[0445] 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."

[0446] Conventional food delivery services only suggest generic menus without considering the user's individual preferences or emotional state, resulting in a lack of personalized user experience. Even systems that reflect the user's preferences and conversational style lack the ability to generate appropriate responses based on the user's emotions, resulting in low user satisfaction.

[0447] 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.

[0448] In this invention, the server includes means for collecting data on the hobbies, preferences, and conversation style of a specific individual, means for analyzing the collected data to create a profile, means for generating a replica of the specific individual based on the profile, means for interacting with a user using the generated replica, means for analyzing the emotional state of the user, means for adjusting responses to the user based on the emotional state, means for recommending foods based on the individual's eating preferences and past ordering history, and means for using the responses in the food recommendation, thereby enabling a personalized food delivery experience that reflects the user's individual preferences and emotions.

[0449] "The hobbies and preferences of a particular individual" refers to the things that a particular individual likes or is interested in.

[0450] "Conversational style" refers to the language patterns and speaking characteristics that a particular individual uses in everyday interactions.

[0451] "Data collection means" refers to devices or methods for collecting information such as text, audio, or images.

[0452] "Means for analyzing data to create profiles" means processes or software that analyze collected data and generate profiles that reflect the characteristics of specific individuals.

[0453] "Means for generating a replica of a specific individual" refers to equipment or methods for creating a virtual entity (replica) that imitates a specific individual based on the results of analysis.

[0454] "Means for interacting with the user" refers to the interface and functions that allow the replica to communicate with the user.

[0455] "Means for analyzing emotional state" refers to analytical engines and algorithms for reading emotions from the user's voice or text.

[0456] "Means for adjusting responses" refers to techniques or systems for appropriately changing the content of responses generated based on the user's emotional state.

[0457] "Dietary preferences" refer to the types of foods and beverages that a particular individual prefers.

[0458] "Past Order History" refers to a record of food and beverage orders a user has placed.

[0459] "Food recommendation tools" refers to algorithms or systems that suggest appropriate foods based on a user's preferences and ordering history.

[0460] The present invention provides a system for enhancing user experience based on the tastes, preferences, and conversational styles of specific individuals. In particular, an example of application to a food delivery service will be described in detail.

[0461] System Overview

[0462] The system uses the following major hardware and software:

[0463] User's device (smartphone, tablet, etc.)

[0464] Server (performs data collection and analysis, profile generation, replica generation, and response generation)

[0465] Sentiment analysis engine (Hugging Face Transformers)

[0466] Response generation engine (OpenAI's GPT-3 (registered trademark))

[0467] Data collection

[0468] Users can use the device to input various types of text data, such as diary entries, text messages, emails, and social media posts, as well as record their own voice. This data is then sent to a server via a secure connection and stored in a database.

[0469] Profile Generation

[0470] The server analyzes the collected text and speech data. It uses morphological and dependency structure analysis to extract language patterns and frequently occurring topics from the text data, and creates a user profile based on these analysis results. It also converts speech data into text using a speech recognition engine and performs spectral analysis to model speech characteristics.

[0471] Replication and Interaction

[0472] The server then generates a chatbot that has learned the target individual's language patterns based on the profile. This chatbot is capable of generating natural responses based on everyday conversations and specific topics. It also trains a speech synthesis engine based on the results of voice data analysis, building a speech synthesis model that reproduces the specific individual's voice characteristics.

[0473] Sentiment Analysis and Response Adjustment

[0474] The server uses an emotion analysis engine to recognize emotions from the user's voice and text data. Based on the recognized emotional state, the profile is updated and the generated response is adjusted. For example, if the user says "Today was tough" in a tired voice, the server will generate a compassionate response such as "Today was tough. Let's take a break."

[0475] Food recommendation function

[0476] Furthermore, the system recommends foods based on an individual's dietary preferences and past ordering history. It combines the user's emotional state and profile information to make optimal food suggestions. For example, if a user says, "I'm feeling low today," the server will recommend foods along with a response like, "Maybe ice cream would be good. Something sweet will definitely cheer you up."

[0477] Specific examples

[0478] When a user types "What's your recommendation today?" into a food concierge app, the system analyzes the words for emotional nuances and suggests the best food options. An example prompt could be, "Generate a response as if the user were talking to a close friend: What's your recommendation today?"

[0479] This enables a personalized food delivery experience that reflects the user's individual preferences and emotions.

[0480] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0481] Step 1:

[0482] A user uses a device to input text data in various formats, such as diary entries, text messages, emails, and social media posts, and records audio data. This data is sent from the device to a server via a secure connection. The input data is text data and audio data, and is sent via secure communication. The output is a notification to the server that the data has been successfully received.

[0483] Step 2:

[0484] The server stores the received text and voice data in a database. The server organizes and manages this data, saving it in a different folder for each user. The input at this stage is the text and voice data sent by the user, and the output is storage in the database.

[0485] Step 3:

[0486] The server performs morphological and dependency structure analysis on the collected text data. Specifically, it uses a natural language processing engine to analyze the text data and extract language patterns, frequently occurring topics, and specific response trends. The input is the text data stored in the database, and the output is a user profile based on the analysis results.

[0487] Step 4:

[0488] The server uses a speech recognition engine to convert the speech data into text and performs spectral analysis, which extracts the frequency components and intonation of the speech and models the speech characteristics. The input is the speech data stored in the database, and the output is the speech characteristic model.

[0489] Step 5:

[0490] The server generates a chatbot that learns the target individual's language patterns based on the profile information. The generated chatbot has the ability to generate natural responses to everyday conversations and specific topics. The input is text data and the results of its analysis, and the output is the chatbot.

[0491] Step 6:

[0492] The server trains the speech synthesis engine based on the voice characteristic model and constructs a voice synthesis model that reproduces the voice characteristics of a specific individual. The input is the voice characteristic model and the output is the voice synthesis model.

[0493] Step 7:

[0494] The server uses an emotion analysis engine to recognize emotions from the user's text and voice. It obtains emotion labels (e.g., "positive" or "negative") and adds them to the profile. The input is the user's text and voice data, and the output is the emotion labels.

[0495] Step 8:

[0496] The server processes the generated response based on the recognized emotional state, modifying the tone and content of the response based on the emotional label. The input is the emotional label and the chatbot's base response, and the output is the optimized response.

[0497] Step 9:

[0498] The server recommends optimal foods based on the user's dietary preferences and past ordering history, suggesting foods that match the user's immediate emotional state and long-term preferences. The input is user profile information and emotional state, and the output is a list of recommended foods.

[0499] Step 10:

[0500] The user initiates a conversation with the chatbot from their device and receives food recommendations. They then check the response and list of recommended foods received from the server and reflect these in their next conversation or order. The input is the response from the server and the list of recommended foods, and the output is increased user satisfaction.

[0501] 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.

[0502] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (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.

[0503] 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.

[0504] [Second embodiment]

[0505] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0506] 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.

[0507] 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).

[0508] 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.

[0509] 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.

[0510] 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).

[0511] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for 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.

[0512] 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.

[0513] 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.

[0514] 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.

[0515] In the smart glasses 214, the 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.

[0516] 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."

[0517] The present invention is a system that collects data on the hobbies, preferences, and conversational style of a specific individual, analyzes the data to create a profile, and generates a replica based on this profile. The generated replica can interact with the user and supports both voice and text. Specific embodiments are described below.

[0518] Data collection

[0519] 1. Terminal: Data entry

[0520] Users use devices such as smartphones and personal computers to input text data in various formats, such as diaries, text messages, emails, and posts on social media.

[0521] The user uses an application on the device to record their own voice, and this voice data contains the target individual's characteristic phrasing and tone of voice.

[0522] 2. Terminal: Data transmission

[0523] The device transmits the collected text and voice data to the server over a secure connection, after which the device notifies the user that the data was successfully transmitted.

[0524] Data analysis

[0525] 1. Server: Data storage

[0526] The server stores the received text data and voice data in a database, which is organized and managed for each user.

[0527] 2. Server: Text data analysis

[0528] The server uses morphological and dependency structure analysis to analyze the text data for language patterns, frequent topics, and specific response tendencies, and the analysis results are stored in a database as user profiles.

[0529] 3. Server: Voice data analysis

[0530] The server uses a voice recognition engine to convert the voice data into text, which is then used for analysis.

[0531] The target individual's vocal characteristics are modeled through spectral analysis, which extracts the frequency components and intonation of the voice.

[0532] Replication

[0533] 1. Server: Chatbot generation

[0534] Based on the results of analyzing the text data, the server generates a chatbot that mimics the conversational style and language patterns of the target individual, and is capable of generating natural responses in everyday conversations and on specific topics.

[0535] 2. Server: Building a speech synthesis model

[0536] The server trains the speech synthesis engine based on the analysis of the voice data and builds a speech synthesis model that reproduces the voice characteristics of a specific individual, allowing it to convert text responses into voice responses.

[0537] User Interaction

[0538] 1. User: Initiate conversation

[0539] Surviving family members and friends can then interact with the replica through a dedicated application, for example by typing, "What should I cook today?"

[0540] 2. Device: Send Message

[0541] The terminal sends this input message to the server.

[0542] 3. Server: Response Generation

[0543] The chatbot on the server generates an appropriate response based on the received message, for example, "How about making curry today?" in text and voice.

[0544] 4. Terminal: Receives response

[0545] The terminal displays and plays the text and voice responses received from the server, allowing the user to confirm and proceed with the next dialogue.

[0546] Specific examples

[0547] For example, if a particular individual frequently talks about movies they liked in the past, then based on that conversation log, when a user asks, "What movie have you seen recently?", the server's chatbot can generate a response such as, "You used to like action movies, so check out these new action movies!"

[0548] This system allows surviving family and friends to receive emotional support by interacting with a replica that faithfully reproduces the individual's hobbies, preferences, and conversation style.

[0549] The processing flow will be explained below.

[0550] Step 1:

[0551] Users use the device's applications to enter information about their diary, conversation logs, and hobbies and preferences, including text messages, emails, and social media posts.

[0552] Users can also use the device's recording function to record voice data, including everyday conversations and voice messages on specific topics.

[0553] Step 2:

[0554] The device sends the collected text and voice data to a server over a secure connection, and the data is encrypted before transmission.

[0555] After the transmission is complete, the terminal notifies the user that the data has been successfully transmitted.

[0556] Step 3:

[0557] The server stores the received text data and voice data in a database.

[0558] The database is divided by user, and each user's data is organized and managed.

[0559] Step 4:

[0560] The server analyzes the collected text data using morphological analysis and dependency structure analysis.

[0561] Specifically, it identifies the parts of speech of words in a sentence, analyzes the structure of the sentence, and extracts frequently occurring words, phrases, and frequently discussed topics, thereby identifying the user's language patterns.

[0562] Step 5:

[0563] The server converts the voice data into text data using a voice recognition engine.

[0564] The converted text data is also analyzed and stored in a database. The server also performs spectral analysis to extract the frequency components of the voice and the features of each phoneme.

[0565] Step 6:

[0566] The server then creates a user profile based on the analysis results, which includes the user's language patterns, interests, and vocal characteristics.

[0567] The profile is stored in a database for later processing.

[0568] Step 7:

[0569] The server generates a chatbot that learns the language patterns of the target user based on the user profile.

[0570] The chatbot is configured to respond to a user's specific phrases and topical trends.

[0571] Step 8:

[0572] Based on the results of the voice data analysis, the server trains the voice synthesis engine and builds a voice synthesis model that reproduces the user's voice characteristics.

[0573] This speech synthesis model is used to convert the generated text response into a speech response.

[0574] Step 9:

[0575] Users can initiate interactions with the created replicas through dedicated applications.

[0576] When the user initiates a conversation, the terminal sends the user's message to the server.

[0577] Step 10:

[0578] The chatbot on the server generates an appropriate response based on the received message.

[0579] For example, if a user asks, "What should I cook today?" the server generates a text response saying, "How about making curry?"

[0580] Step 11:

[0581] The server converts the generated text response into voice using a speech synthesis model.

[0582] This causes the text response to be output as a voice response.

[0583] Step 12:

[0584] The terminal displays and plays back the text and audio responses received from the server.

[0585] The user can then confirm the response and continue the dialogue.

[0586] Example 1

[0587] 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."

[0588] In modern society, after a specific individual passes away, there is a need to build an interactive dialogue system based on that individual's hobbies, preferences, and conversational style. However, there is still a lack of technology to efficiently collect available data, accurately analyze it, and conduct natural dialogue. The purpose of this invention is to solve this problem and provide a system that generates a replica based on the hobbies, preferences, and conversational style of a specific individual, enabling dialogue with the user.

[0589] 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.

[0590] In this invention, the server includes: means for a user to input text data and voice data using a terminal; means for transmitting the input text data and voice data to the server via a secure connection; means for the server to store the received text data and voice data in a database; means for the server to analyze the text data using morphological analysis and dependency structure analysis to extract language patterns and frequent topics; means for the server to convert the voice data into text using a voice recognition engine and model voice characteristics through spectral analysis; means for the server to generate a chatbot and a voice synthesis model that imitate the conversation style and language patterns of a specific individual based on the analysis results of the text data and voice data; and means for engaging in a dialogue with the user using the generated chatbot and voice synthesis model. This makes it possible to faithfully reproduce the hobbies, preferences, and conversation style of a specific individual and realize natural, interactive dialogue.

[0591] A "terminal" is an electronic device used by a user, such as a smartphone or a personal computer.

[0592] "Text data" refers to data of sentences or character strings that a user inputs using a terminal.

[0593] "Audio data" refers to data of audio recorded by a user using a terminal.

[0594] A "secure connection" is an encrypted communication method that protects data from being intercepted or tampered with by third parties during transmission.

[0595] A "server" is a computer system connected to the Internet that receives, analyzes, and stores text and audio data.

[0596] A "database" is a data structure for efficiently organizing and managing data within a server.

[0597] "Morphological analysis" is a natural language processing technique that divides text data into words and identifies their parts of speech in order to understand their meaning.

[0598] "Dependency structure analysis" is a natural language processing technique that analyzes the relationships between words in text data and reveals the structure of sentences.

[0599] A "speech recognition engine" is a software or hardware system for converting voice data into text data.

[0600] "Spectral analysis" is a technique for analyzing the frequency components and intonation of voice data.

[0601] "Voice characteristics" refer to the unique characteristics of individual voices, including, for example, tone, pitch, and intonation.

[0602] A "chatbot" is an automated response system that uses text data to interact with users.

[0603] A "voice synthesis model" is a model for converting text data into voice data based on analyzed voice characteristics.

[0604] A "profile" is information generated based on collected data that summarizes a specific individual's hobbies, preferences, and conversation style.

[0605] The present invention is a system that collects data on the hobbies, preferences, and conversational style of a specific individual, analyzes the data to create a profile, and generates a replica based on this profile. The system analyzes text data and voice data entered by a user using a terminal, and generates a chatbot and a voice synthesis model that mimics the conversational style and language patterns of the specific individual. The following describes how to specifically implement the present invention.

[0606] Data Collection and Transmission

[0607] Users use devices such as smartphones and PCs to input text data in various formats, such as diaries, text messages, emails, and social media posts. For example, a user might input "It was a nice day today" into a dedicated smartphone app. The user might also use an application on the device to record their own voice. This voice data contains the target individual's characteristic phrasing and tone of voice. For example, a user might record and save the voice saying "I slept well today."

[0608] The device sends the collected text data and voice data to the server via a secure connection (such as SSL / TLS). For example, the device sends text data such as "The weather was nice today" and voice data such as "I slept well today" to the server. After sending, the device notifies the user that the data has been sent successfully. For example, the device displays a message on the screen saying "The data has been sent successfully."

[0609] Data Storage and Analysis

[0610] The server stores the received text data and voice data in a database. This database is a data structure for organizing and managing data for each user. For example, the server stores the text data "The weather was nice today" and the voice file "I slept well today" in the "user123" folder.

[0611] The server then uses morphological analysis and dependency structure analysis to analyze the text data for language patterns, frequently occurring topics, and specific response trends. The software used includes the Python morphological analysis library "MeCab" and the dependency structure analysis engine "spaCy." For example, the server analyzes "The weather was nice today" and recognizes that "weather" is a frequently occurring topic and that it is a pleasant day. The analysis results are saved in a database as a user profile. For example, the server can add information to the profile such as "I have many good experiences with the weather."

[0612] The server also uses a speech recognition engine to convert the voice data into text. For this, it uses "Google Cloud Speech-to-Text." For example, it converts the voice data "I slept well today" into text. It then performs spectral analysis to extract the frequency components and intonation of the voice. For this, it uses "Praat." For example, the server analyzes and models the intonation and pitch of the voice "I slept well today."

[0613] Replication and User Interaction

[0614] The server generates a chatbot that mimics the target individual's conversation style and language patterns based on the results of analyzing the text data. The generative AI model used is "OpenAI GPT-4." For example, based on information that the server has "many positive experiences related to the weather," it generates a chatbot that prefers topics related to sunny days.

[0615] The server then trains a speech synthesis engine based on the results of the voice data analysis to build a speech synthesis model that can reproduce the voice characteristics of a specific individual. Engines used include Google Cloud Text-to-Speech and Amazon Polly. For example, the server builds a voice model that can reproduce the tone and tempo of the phrase "I slept well tonight."

[0616] Surviving family and friends can start a conversation with the created replica through a dedicated application. For example, a family member can type, "What should we cook today?" The device sends this input message to the server. The chatbot on the server generates an appropriate response based on the received message. For example, it might generate a response in text and voice, such as, "How about cooking curry today?" The device displays and plays back the text and voice response received from the server. The user can confirm this and carry out the next conversation.

[0617] For example, if a particular individual frequently talked about their favorite movies in the past, when a user asks "What movie did you see recently?" based on the conversation log, the chatbot on the server can generate a response such as "You used to like action movies, so check out the new action movies!" In this way, surviving family and friends can receive emotional support by interacting with a replica that faithfully reproduces the hobbies, preferences, and conversation style of the particular individual.

[0618] Example prompt sentence:

[0619] "What activities have you been interested in lately?"

[0620] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0621] Step 1:

[0622] A user inputs text data and voice data using a device. Using an application on a smartphone or computer, the user inputs text data in the form of a diary entry, text message, email, or social media post. The user also uses an application to record their own voice. For example, a user inputs text data such as "It was nice weather today" into a dedicated smartphone application and records voice such as "I slept well today." The input data is text data and voice data.

[0623] Step 2:

[0624] The device sends the collected text and voice data to the server through a secure connection. Specifically, the data is sent using encrypted communication such as SSL / TLS. When the transmission is complete, the device displays a notification to the user stating, "The data was sent successfully." The input data is the text and voice data entered by the user, and the output data is a notification indicating that the data has been sent.

[0625] Step 3:

[0626] The server stores the received text data and voice data in a database. Specifically, the server saves the text data "It was nice weather today" and the voice file "I slept well today" in the "user123" folder. The input data is the transmitted text data and voice data, and the output data is the data stored in the database.

[0627] Step 4:

[0628] The server analyzes the text data using morphological analysis and dependency structure analysis. Specifically, it uses the Python morphological analysis library "MeCab" and the dependency structure analysis engine "spaCy" to extract language patterns, frequently occurring topics, and specific reaction trends from the text data. For example, the server analyzes the text "It was nice weather today" and recognizes that "weather" is a frequently occurring topic. The input data is text data, and the output data is the analyzed language patterns and frequently occurring topics.

[0629] Step 5:

[0630] The server converts the voice data into text using a speech recognition engine. Specifically, it uses "Google Cloud Speech-to-Text" to convert the voice data "I slept well today" into text. It also performs spectral analysis to extract the frequency components and intonation of the voice. For example, using "Praat," the server analyzes and models the intonation and pitch of the voice "I slept well today." The input data is the voice data, and the output data is the analyzed voice characteristics and converted text data.

[0631] Step 6:

[0632] Based on the analysis results of the text and voice data, the server generates a chatbot and voice synthesis model that mimics the conversation style and language patterns of a specific individual. Specifically, it uses the generative AI model "OpenAI GPT-4" to generate a chatbot based on the text data, and builds a voice model using the voice synthesis engines "Google Cloud Text-to-Speech" and "Amazon Polly." The input data is the analysis results, and the output data is the chatbot and voice synthesis model.

[0633] Step 7:

[0634] The user initiates a conversation with the generated replica through a terminal. Specifically, the user inputs a message into a dedicated application, such as "What should I cook today?" The input message is sent to the server, and the chatbot on the server generates a response in text and voice, such as "How about making curry?" The input data is the message from the user, and the output data is the text and voice response from the chatbot.

[0635] Step 8:

[0636] The terminal displays and plays back the text and voice response received from the server. Specifically, it displays the chatbot's response text on the terminal's display and plays back the response aloud. For example, the terminal displays the text "How about making curry?" on the screen and plays back the voice. This allows the user to carry out the next dialogue. The input data is the text and voice response from the server, and the output data is the displayed text and played back voice.

[0637] An example of a prompt sentence: "What activities have you been interested in lately?"

[0638] (Application example 1)

[0639] 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."

[0640] In today's brick-and-mortar stores, it is difficult to provide personalized product recommendations and assistance to each individual consumer. Conventional services often provide general information and uniform customer service to all customers, failing to address individual preferences and needs. This can result in low user satisfaction and a loss of purchasing motivation. Furthermore, there is a lack of technology to effectively utilize profiles that utilize a specific individual's conversation style and preferences using a visual display device. Therefore, there is a need for the development of a system that can provide personalized product recommendations and assistance based on a specific individual's profile through a visual display device.

[0641] 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.

[0642] In this invention, the server includes means for collecting data on the hobbies, preferences, and conversation style of a specific individual, means for analyzing the collected data to create a profile, means for generating a replica of the specific individual based on the profile, means for interacting with the user using the generated replica, and means for providing customized product suggestions and assistance based on the analyzed profile using a visual display device attached to the user's communication device. This makes it possible to effectively provide personalized product suggestions and assistance based on the profile of a specific individual in a physical store.

[0643] "Hobbies" are preferences regarding the things and activities that a particular individual likes.

[0644] A "conversational style" is the linguistic expression or method of communication that a particular individual normally uses.

[0645] A "data collection instrument" is a device or method for collecting information about the tastes, preferences, and speaking styles of a particular individual.

[0646] A "profile" is a collection of information that describes the characteristics of a particular individual, created based on collected data.

[0647] A "replica" is an interactive virtual assistant generated based on a specific individual's profile.

[0648] The "interaction means" is a method or device for communicating with a user using the generated replica.

[0649] A "visual display device" is a communication device worn by a user that visually displays text and images.

[0650] "Product Recommendation" is a method or device aimed at recommending personalized products and services based on a user's profile.

[0651] "Assistance" refers to help or support provided based on a specific individual's preferences and conversational style.

[0652] This invention is a "personalized shopping assistant" system that provides personalized product recommendations and assistance in brick-and-mortar stores. It uses a visual display device attached to a user's communication device. Specifically, it is implemented using smart glasses. The system of this invention suggests products and services based on the user's tastes, preferences, and conversation style, providing an experience tailored to individual needs.

[0653] System Configuration and Operation

[0654] 1. Hardware and Software:

[0655] Hardware: Smart glasses (with voice recognition microphone and speaker), server, microphone

[0656] Software: GPT-2, speech recognition library (speech_recognition), speech synthesis engine (pyttsx3), Python

[0657] The smart glasses recognize the user's voice and send the data to the server, which analyzes the data and generates a response based on the user's profile, which is presented visually and audibly to the smart glasses.

[0658] 2. Data collection and analysis:

[0659] The user puts on the smart glasses and speaks their questions or requests. The speech recognition library, speech_recognition, converts the speech into text data.

[0660] The voice and text data sent to the server is used to generate appropriate responses based on the profile using GPT-2.

[0661] 3. Create a profile:

[0662] The server collects data about the user's interests and conversational style and creates a profile based on that data, including the types of products the user likes and conversational habits.

[0663] Through a speech recognition engine and speech synthesis engine, the user's language patterns and voice characteristics are analyzed and reflected in the profile.

[0664] 4. Generate and present the response:

[0665] When a user speaks to the smart glasses, the server creates a prompt based on the content and generates a response using the GPT-2 model.

[0666] The generated text response is converted to speech using the speech synthesis engine pyttsx3 and played through the smart glasses' speakers, and also displayed as text on the visual display.

[0667] Examples:

[0668] For example, if a user speaks to the smart glasses, "What kind of food would you like today?", the microphone in the smart glasses will capture the voice, and the speech recognition library, speech_recognition, will convert it into text. This text data will be sent to the server and processed as the following prompt sentence:

[0669] User: What would you like to eat today?

[0670] Assistant: The weather is nice today, why not enjoy a barbecue?

[0671] The responses are converted to speech and played back to the user, and are also displayed as text on a visual display, allowing for personalized product recommendations and assistance based on the user's preferences and conversational style.

[0672] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0673] Step 1:

[0674] User: Voice Input

[0675] Users wear smart glasses and input questions or requests by voice.

[0676] Input: User's voice data

[0677] Output: Audio data is captured

[0678] Specific operation: A microphone installed in the smart glasses collects the user's voice.

[0679] Step 2:

[0680] Device: Voice recognition

[0681] The device uses the speech recognition library speech_recognition to convert the voice data into text data.

[0682] Input: Captured audio data

[0683] Output: Text data

[0684] Specific operation: Audio data collected from the microphone is sent to the speech recognition library, and corresponding text data is generated.

[0685] Step 3:

[0686] Terminal: Data transmission

[0687] The terminal transmits the converted text data to the server over a secure connection.

[0688] Input: Text data

[0689] Output: Text data sent to the server

[0690] Specific operation: The converted text data is encrypted and sent securely to the server.

[0691] Step 4:

[0692] Server:Profile Creation

[0693] The server analyzes the received text data and creates a profile of the user's interests, preferences and conversation style.

[0694] Input: Text data

[0695] Output: User profile

[0696] Specific operation: Through morphological analysis and dependency structure analysis, language patterns and preferences are extracted from text data and a profile is generated.

[0697] Step 5:

[0698] Server: Response Generation

[0699] The server uses the generative AI model GPT-2 to create appropriate prompts and generate responses based on the user's profile.

[0700] Input: Profile and text data

[0701] Output: Text response

[0702] How it works: A prompt sentence is input into the GPT-2 model, which generates a text response based on the user's interests and preferences.

[0703] Example prompt sentence:

[0704] User: What would you like to eat today?

[0705] Assistant: The weather is nice today, why not enjoy a barbecue?

[0706] Step 6:

[0707] Server: Speech synthesis

[0708] The server converts the text response into speech using the speech synthesis engine pyttsx3.

[0709] Input: Text response

[0710] Output: Audio data

[0711] What it does: The generated text response is fed into a speech synthesis engine to generate a voice that sounds similar to a human voice.

[0712] Step 7:

[0713] Terminal: Response presentation

[0714] The terminal plays the generated audio data through the speaker of the smart glasses and displays the text data on a visual display.

[0715] Input: Audio and text data

[0716] Output: Presents the response to the user

[0717] Specific Actions: Provide a response to the user by playing audio data over a speaker and displaying text data on a visual display device.

[0718] 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.

[0719] The present invention is a system that collects data on a specific individual's hobbies, preferences, and conversation style, analyzes the data, and creates a profile. It then generates a replica based on the profile, and combines this with an emotion engine that recognizes the user's emotions. The generated replica can converse with the user and adjusts its responses according to the user's emotional state. Specific embodiments are described below.

[0720] Data collection

[0721] 1. Terminal: Data entry

[0722] Users use devices such as smartphones and personal computers to input text data in various formats, such as diaries, text messages, emails, and posts on social media.

[0723] The user uses an application on the device to record their own voice, and this voice data contains the target individual's characteristic phrasing and tone of voice.

[0724] 2. Terminal: Data transmission

[0725] The device transmits the collected text and voice data to the server over a secure connection, after which the device notifies the user that the data was successfully transmitted.

[0726] Data analysis

[0727] 1. Server: Data storage

[0728] The server stores the received text data and voice data in a database.

[0729] Within the database, data is organized and managed for each user.

[0730] 2. Server: Text data analysis

[0731] The server uses morphological and dependency structure analysis to analyze the text data for language patterns, frequent topics, and specific response tendencies, and the analysis results are stored in a database as user profiles.

[0732] 3. Server: Voice data analysis

[0733] The server uses a voice recognition engine to convert the voice data into text, and based on this, the voice data is also used for analysis.

[0734] The server performs spectral analysis to extract the frequency components and intonation of the voice, thereby modeling the voice characteristics.

[0735] Replication

[0736] 1. Server: Chatbot generation

[0737] Based on the results of analyzing the text data, the server generates a chatbot that has learned the target individual's language patterns and is capable of generating natural responses in everyday conversations and on specific topics.

[0738] 2. Server: Building a speech synthesis model

[0739] The server trains the speech synthesis engine based on the analysis of the voice data and builds a speech synthesis model that reproduces the voice characteristics of a specific individual, allowing it to convert text responses into voice responses.

[0740] Adding an Emotion Engine

[0741] 1. Server: User emotion recognition

[0742] The server uses an emotion engine to recognize emotions from the user's voice and text data, for example by analyzing voice characteristics such as tone, speed, and volume, and from text data by analyzing the context and emotional tendencies of selected words.

[0743] 2. Server: Emotion-based profile update

[0744] Based on the recognized emotions, the user profile is updated, which includes the user's emotional tendencies and current emotional state.

[0745] 3. Server: Emotion-based response generation

[0746] The generated response is tailored based on the emotional state recognized by the emotion engine. For example, if the user asks a question in a sad voice, the server will generate a more empathetic and comforting response.

[0747] User Interaction

[0748] 1. User: Initiate conversation

[0749] Family and friends can then interact with the replica through a dedicated application, which sends text and voice messages to the server.

[0750] 2. Device: Send Message

[0751] The terminal transmits the input text message and voice data to the server.

[0752] 3. Server: Response Generation

[0753] The server's chatbot generates an appropriate response based on the received message and emotional state. For example, if the user says "Today was tough" in a tired voice, the server generates a response like "Today was tough. Let's take a break."

[0754] 4. Server: Speech synthesis

[0755] The server converts the generated text response into voice using a speech synthesis model, which outputs the text response as a voice response.

[0756] 5. Terminal: Receives response

[0757] The terminal displays and plays back the text and voice responses received from the server, allowing the user to confirm the responses and continue the dialogue.

[0758] Specific examples

[0759] For example, if a specific individual has often talked about "movies" in the past, when the user asks "What movie did you see recently?", the chatbot on the server will generate a response such as "There was a new action movie recently that was good. You should definitely check it out" based on the user's preferences and emotional state. Also, by recognizing the user's emotional state, if the user is feeling down, for example, an encouraging response such as "Cheer up. Let's find a movie we can enjoy together" will be generated.

[0760] This system allows surviving family and friends to receive emotional support by interacting with a replica that faithfully reproduces the individual's hobbies, preferences, and conversation style. Furthermore, the emotional engine recognizes the user's emotional state and generates responses accordingly, enabling more empathetic and natural conversations.

[0761] The processing flow will be explained below.

[0762] Step 1:

[0763] Users can use the device's applications to enter information about their diary, conversation log, and hobbies and interests in text format, and also record voice messages using the device's recording function.

[0764] Step 2:

[0765] The device sends the collected text and voice data to the server over a secure connection, and the user is notified once the data has been sent.

[0766] Step 3:

[0767] The server stores the received text data in a database, and simultaneously stores the audio data in the database.

[0768] Step 4:

[0769] The server analyzes the received text data using morphological and dependency analysis to identify the user's language patterns and frequent topics and create a profile.

[0770] Step 5:

[0771] The server converts the voice data into text data using a speech recognition engine, and then performs spectral analysis to extract features for each frequency component and phoneme.

[0772] Step 6:

[0773] The server stores a user profile based on the analysis results in a database, which includes language patterns, interests, and voice characteristics.

[0774] Step 7:

[0775] The server generates a chatbot based on the profile, which is trained to replicate the language patterns of the specific individual and to engage in natural conversations.

[0776] Step 8:

[0777] The server uses the speech analysis results to train a speech synthesis engine and build a speech synthesis model that replicates the voice characteristics of a particular individual, which is then used to convert the text response into a speech response.

[0778] Step 9:

[0779] The user initiates a dialogue with the created replica through a dedicated application. At the beginning of the dialogue, the terminal sends the user's input to the server.

[0780] Step 10:

[0781] The server uses an emotion engine to recognize emotions from user input data, analyzing voice tone, speed, volume, etc. from voice data, and emotional trends from text data.

[0782] Step 11:

[0783] The server's chatbot generates more empathetic and appropriate responses based on the emotions recognized by the emotion engine. For example, if the user speaks in a sad voice, the server will generate a comforting response.

[0784] Step 12:

[0785] The server converts the generated text response into voice using a speech synthesis model, which outputs the text response as a voice response that reflects the appropriate emotion.

[0786] Step 13:

[0787] The device displays and plays the text and voice responses received from the server to the user, who can then confirm the responses and continue the dialogue.

[0788] Example 2

[0789] 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."

[0790] In modern society, emotional support gained through memories and conversations with specific individuals is highly valued. However, when a specific individual has passed away or is far away, means of sharing conversations and memories with that person are limited. Conventional technologies simply play back text or audio data, and do not realize natural conversations that reflect the user's emotions, hobbies, and preferences. Therefore, the present invention calls for a new system that faithfully reproduces the hobbies, preferences, and conversation style of a specific individual, enabling conversations that correspond to the user's emotional state.

[0791] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes: means for collecting data on the hobbies, preferences, and conversation style of a specific individual; means for encrypting the collected text data and voice data and transmitting it through a secure connection; means for analyzing the collected data to create a user profile; means for generating a chatbot and a voice synthesis model for the specific individual based on the user profile; means for recognizing the user's emotions using an emotion engine and updating the profile; and means for interacting with the user using the generated replica and adjusting responses based on the user's emotional state. This makes it possible to interactively interact with a replica that faithfully reproduces the characteristics of a specific individual and obtain responses that correspond to the user's emotional state.

[0792] 1. "Specified Individual" refers to a specific person about whom data is collected and analyzed by the system.

[0793] 2. "Preferences" refers to information about a particular individual's preferred activities, interests, and preferences.

[0794] 3. "Conversational style" refers to a particular individual's speaking style, phrasing, language patterns, and communication characteristics.

[0795] 4. "Data Collection Instrument" means the part of the system that captures text and voice data about a specific individual.

[0796] 5. "Data transmission means" refers to the function for encrypting collected data and transmitting it securely to the server.

[0797] 6. "Data analysis means" refers to the part of the system that analyzes collected data to extract and understand hobbies, preferences, and conversation styles.

[0798] 7. "User profile" refers to information that reproduces the characteristics of a specific individual, generated based on analyzed hobbies, preferences, and conversation style.

[0799] 8. "Chatbot generation means" refers to a function for generating a chatbot that reproduces the language patterns of a specific individual based on a user profile.

[0800] 9. "Voice synthesis model generation means" refers to a function for generating a voice synthesis model that reproduces the voice characteristics of a specific individual based on the results of voice data analysis.

[0801] 10. "Emotion engine" refers to software or algorithms that recognize and analyze user emotions.

[0802] 11. "Profile Updater" refers to functionality for updating a user profile based on the user's emotional state recognized by the Emotion Engine.

[0803] 12. "Response adjustment means" refers to a function that enables the generated replica to generate a response according to the emotional state during interaction with the user.

[0804] 13. "Replica" refers to a simulation that faithfully reproduces the tastes, preferences, and conversation style of a specific individual.

[0805] The present invention is a system that collects data on a specific individual's hobbies, preferences, and conversation style, analyzes the data, and creates a profile. A replica is generated based on this profile, and by combining it with an emotion engine that recognizes the user's emotions, a system is provided that can interact with the user. Specific embodiments are described below.

[0806] Data collection

[0807] 1. Terminal: Data entry

[0808] Users use devices such as smartphones and PCs to input various forms of text data, such as diaries, text messages, emails, and posts on social media. Users also use applications on their devices to record their own voice. This voice data includes the target individual's characteristic phrasing and tone of voice.

[0809] 2. Terminal: Data transmission

[0810] The device encrypts the collected text and voice data and sends it to the server over a secure connection, after which the device notifies the user that the data has been successfully sent.

[0811] Data analysis

[0812] 1. Server: Data storage

[0813] The server stores the received text and voice data in a database, where data is organized and managed for each user.

[0814] 2. Server: Text data analysis

[0815] The server uses morphological and dependency structure analysis to analyze the text data for language patterns, frequent topics, and specific response tendencies, and the analysis results are stored in a database as user profiles.

[0816] 3. Server: Voice data analysis

[0817] The server converts the voice data into text using a speech recognition engine. Based on this, the voice data is also analyzed. The server also performs spectral analysis to extract the frequency components and intonation of the voice and model the voice characteristics.

[0818] Replication

[0819] 1. Server: Chatbot generation

[0820] Based on the results of analyzing the text data, the server generates a chatbot that has learned the target individual's language patterns and is capable of generating natural responses in everyday conversations and on specific topics.

[0821] 2. Server: Building a speech synthesis model

[0822] The server trains the speech synthesis engine based on the analysis of the voice data and builds a speech synthesis model that reproduces the voice characteristics of a specific individual, allowing it to convert text responses into voice responses.

[0823] Adding an Emotion Engine

[0824] 1. Server: User emotion recognition

[0825] The server uses an emotion engine to recognize emotions from the user's voice and text data, for example by analyzing voice characteristics such as tone, speed, and volume, and from text data by analyzing the context and emotional tendencies of selected words.

[0826] 2. Server: Emotion-based profile update

[0827] Based on the recognized emotions, the user profile is updated, which includes the user's emotional tendencies and current emotional state.

[0828] 3. Server: Emotion-based response generation

[0829] The generated response is tailored based on the perceived emotional state: for example, if the user asks a question in a sad voice, the server will generate a more empathetic and comforting response.

[0830] User Interaction

[0831] 1. User: Initiate conversation

[0832] Users can start interacting with the replicas through a dedicated application, and when a user initiates an interaction, the device sends text and voice messages to the server.

[0833] 2. Device: Send Message

[0834] The terminal transmits the input text message and voice data to the server.

[0835] 3. Server: Response Generation

[0836] The server's chatbot generates an appropriate response based on the received message and emotional state. For example, if the user says "Today was tough" in a tired voice, the server generates a response like "Today was tough. Let's take a break."

[0837] 4. Server: Speech synthesis

[0838] The server converts the generated text response into voice using a speech synthesis model, which outputs the text response as a voice response.

[0839] 5. Terminal: Receives response

[0840] The terminal displays and plays back the text and voice responses received from the server, allowing the user to confirm the responses and continue the dialogue.

[0841] Specific examples and prompts for the generative AI model

[0842] Specific examples

[0843] For example, if a specific individual has often talked about "movies" in the past, when the user asks "What movie did you see recently?", the chatbot on the server will generate a response such as "There was a new action movie recently that was good. You should definitely check it out" based on the user's preferences and emotional state. Also, by recognizing the user's emotional state, if the user is feeling down, for example, an encouraging response such as "Cheer up. Let's find a movie we can enjoy together" will be generated.

[0844] Prompt Sentence Examples

[0845] 1. Prompt to analyze user text data

[0846] "Analyze text data from diaries, text messages, and social media posts using morphological and dependency analysis. Extract language patterns, frequently occurring topics, and specific reaction tendencies for each user to create a user profile."

[0847] 2. Prompt to request voice data analysis

[0848] "Use a speech recognition engine to convert the speech data into text, then perform spectral analysis to extract the frequency components and intonation of the speech. Based on the results, model the speech characteristics."

[0849] 3. Prompt to generate a chatbot

[0850] "Based on the results of analyzing the text data, create a chatbot that learns the user's language patterns. This chatbot should be able to generate natural responses in everyday conversations and on specific topics."

[0851] 4. Prompt to request emotion engine analysis

[0852] "Analyze voice data for characteristics such as tone, speed, and volume, and text data for context and emotional trends in chosen words. Based on the results, recognize the user's emotions and update the user profile."

[0853] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0854] System program processing flow

[0855] Step 1: Data entry

[0856] Users use smartphones or personal computers to input text data in various formats, such as diary entries, text messages, emails, and social media posts, into dedicated applications.

[0857] Input: Text data (e.g., diary entries, messages, posts)

[0858] Output: Collected text data

[0859] The user uses the same application to record their own voice, and the recorded voice data includes the user's characteristic phrasing and tone of voice.

[0860] Input: Audio data (e.g., recording of everyday conversation)

[0861] Output: Collected audio data

[0862] Step 2: Send data

[0863] The device encrypts the collected text and voice data and sends it to a server over a secure connection.

[0864] Input: Collected text and audio data

[0865] Output: Sending encrypted data to the server

[0866] After transmission, the terminal notifies the user that the data has been successfully transmitted.

[0867] Input: Send result

[0868] Output: User notification

[0869] Step 3: Data storage

[0870] The server stores the received text and voice data in a database, where data is organized and managed for each user.

[0871] Input: Received text and audio data

[0872] Output: Data stored in a database

[0873] Step 4: Text data analysis

[0874] The server performs morphological analysis and breaks down the text data into words.

[0875] Input: Text data

[0876] Output: A list of decomposed words

[0877] The server then uses dependency analysis to analyze grammatical and semantic relationships.

[0878] Input: A list of decomposed words

[0879] Output: Structural data showing grammatical and semantic relationships

[0880] Step 5: Audio data analysis

[0881] The server converts the voice data into text data using a voice recognition engine.

[0882] Input: Audio data

[0883] Output: Character data

[0884] In addition, the server performs spectral analysis to extract the frequency components and intonation of the voice.

[0885] Input: Audio data

[0886] Output: Frequency content and intonation data

[0887] Step 6: Create a user profile

[0888] The server creates a user profile based on the results of the analysis of the text and voice data.

[0889] Input: Analysis results (text data and audio data)

[0890] Output: User profile

[0891] Step 7: Chatbot generation

[0892] The server generates a chatbot that learns the user's language patterns based on the user profile.

[0893] Input: User Profile

[0894] Output: Chatbot

[0895] Step 8: Building a speech synthesis model

[0896] Based on the results of the voice data analysis, the server constructs a voice synthesis model that reproduces the voice characteristics of a specific individual.

[0897] Input: Audio data analysis results

[0898] Output: Speech synthesis model

[0899] Step 9: Emotion Recognition and Profile Update

[0900] The server uses an emotion engine to recognize the user's emotions and reflect them in the profile.

[0901] Input: Voice and text data

[0902] Output: Updated user profile

[0903] Step 10: Start a conversation

[0904] The user initiates a conversation with the created replica through a dedicated application. When the conversation begins, the terminal sends the user's text and voice messages to the server.

[0905] Input: User's message (text and voice)

[0906] Output: Send to server

[0907] Step 11: Send the message

[0908] The terminal transmits the input text message and voice data to the server.

[0909] Input: User message

[0910] Output: Data sent to the server

[0911] Step 12: Response Generation

[0912] The chatbot on the server generates an appropriate response based on the received message and emotional state.

[0913] Input: User message and emotion data

[0914] Output: Response message

[0915] Step 13: Speech synthesis

[0916] The server uses a speech synthesis model to convert the generated text response into a voice response.

[0917] Input: Text response

[0918] Output: Voice response

[0919] Step 14: Receive a response

[0920] The terminal displays and plays back the text and voice responses received from the server, allowing the user to confirm the responses and continue the dialogue.

[0921] Input: Response from the server (text and audio)

[0922] Output: Displayed and played response

[0923] (Application example 2)

[0924] 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."

[0925] Conventional food delivery services only suggest generic menus without considering the user's individual preferences or emotional state, resulting in a lack of personalized user experience. Even systems that reflect the user's preferences and conversational style lack the ability to generate appropriate responses based on the user's emotions, resulting in low user satisfaction.

[0926] 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.

[0927] In this invention, the server includes means for collecting data on the hobbies, preferences, and conversation style of a specific individual, means for analyzing the collected data to create a profile, means for generating a replica of the specific individual based on the profile, means for interacting with a user using the generated replica, means for analyzing the emotional state of the user, means for adjusting responses to the user based on the emotional state, means for recommending foods based on the individual's eating preferences and past ordering history, and means for using the responses in the food recommendation, thereby enabling a personalized food delivery experience that reflects the user's individual preferences and emotions.

[0928] "The hobbies and preferences of a particular individual" refers to the things that a particular individual likes or is interested in.

[0929] "Conversational style" refers to the language patterns and speaking characteristics that a particular individual uses in everyday interactions.

[0930] "Data collection means" refers to devices or methods for collecting information such as text, audio, or images.

[0931] "Means for analyzing data to create profiles" means processes or software that analyze collected data and generate profiles that reflect the characteristics of specific individuals.

[0932] "Means for generating a replica of a specific individual" refers to equipment or methods for creating a virtual entity (replica) that imitates a specific individual based on the results of analysis.

[0933] "Means for interacting with the user" refers to the interface and functions that allow the replica to communicate with the user.

[0934] "Means for analyzing emotional state" refers to analytical engines and algorithms for reading emotions from the user's voice or text.

[0935] "Means for adjusting responses" refers to techniques or systems for appropriately changing the content of responses generated based on the user's emotional state.

[0936] "Dietary preferences" refer to the types of foods and beverages that a particular individual prefers.

[0937] "Past Order History" refers to a record of food and beverage orders a user has placed.

[0938] "Food recommendation tools" refers to algorithms or systems that suggest appropriate foods based on a user's preferences and ordering history.

[0939] The present invention provides a system for enhancing user experience based on the tastes, preferences, and conversational styles of specific individuals. In particular, an example of application to a food delivery service will be described in detail.

[0940] System Overview

[0941] The system uses the following major hardware and software:

[0942] User's device (smartphone, tablet, etc.)

[0943] Server (performs data collection and analysis, profile generation, replica generation, and response generation)

[0944] Sentiment analysis engine (Hugging Face Transformers)

[0945] Response generation engine (OpenAI's GPT-3)

[0946] Data collection

[0947] Users can use the device to input various types of text data, such as diary entries, text messages, emails, and social media posts, as well as record their own voice. This data is then sent to a server via a secure connection and stored in a database.

[0948] Profile Generation

[0949] The server analyzes the collected text and speech data. It uses morphological and dependency structure analysis to extract language patterns and frequently occurring topics from the text data, and creates a user profile based on these analysis results. It also converts speech data into text using a speech recognition engine and performs spectral analysis to model speech characteristics.

[0950] Replication and Interaction

[0951] The server then generates a chatbot that has learned the target individual's language patterns based on the profile. This chatbot is capable of generating natural responses based on everyday conversations and specific topics. It also trains a speech synthesis engine based on the results of voice data analysis, building a speech synthesis model that reproduces the specific individual's voice characteristics.

[0952] Sentiment Analysis and Response Adjustment

[0953] The server uses an emotion analysis engine to recognize emotions from the user's voice and text data. Based on the recognized emotional state, the profile is updated and the generated response is adjusted. For example, if the user says "Today was tough" in a tired voice, the server will generate a compassionate response such as "Today was tough. Let's take a break."

[0954] Food recommendation function

[0955] Furthermore, the system recommends foods based on an individual's dietary preferences and past ordering history. It combines the user's emotional state and profile information to make optimal food suggestions. For example, if a user says, "I'm feeling low today," the server will recommend foods along with a response like, "Maybe ice cream would be good. Something sweet will definitely cheer you up."

[0956] Specific examples

[0957] When a user types "What's your recommendation today?" into a food concierge app, the system analyzes the words for emotional nuances and suggests the best food options. An example prompt could be, "Generate a response as if the user were talking to a close friend: What's your recommendation today?"

[0958] This enables a personalized food delivery experience that reflects the user's individual preferences and emotions.

[0959] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0960] Step 1:

[0961] A user uses a device to input text data in various formats, such as diary entries, text messages, emails, and social media posts, and records audio data. This data is sent from the device to a server via a secure connection. The input data is text data and audio data, and is sent via secure communication. The output is a notification to the server that the data has been successfully received.

[0962] Step 2:

[0963] The server stores the received text and voice data in a database. The server organizes and manages this data, saving it in a different folder for each user. The input at this stage is the text and voice data sent by the user, and the output is storage in the database.

[0964] Step 3:

[0965] The server performs morphological and dependency structure analysis on the collected text data. Specifically, it uses a natural language processing engine to analyze the text data and extract language patterns, frequently occurring topics, and specific response trends. The input is the text data stored in the database, and the output is a user profile based on the analysis results.

[0966] Step 4:

[0967] The server uses a speech recognition engine to convert the speech data into text and performs spectral analysis, which extracts the frequency components and intonation of the speech and models the speech characteristics. The input is the speech data stored in the database, and the output is the speech characteristic model.

[0968] Step 5:

[0969] The server generates a chatbot that learns the target individual's language patterns based on the profile information. The generated chatbot has the ability to generate natural responses to everyday conversations and specific topics. The input is text data and the results of its analysis, and the output is the chatbot.

[0970] Step 6:

[0971] The server trains the speech synthesis engine based on the voice characteristic model and constructs a voice synthesis model that reproduces the voice characteristics of a specific individual. The input is the voice characteristic model and the output is the voice synthesis model.

[0972] Step 7:

[0973] The server uses an emotion analysis engine to recognize emotions from the user's text and voice. It obtains emotion labels (e.g., "positive" or "negative") and adds them to the profile. The input is the user's text and voice data, and the output is the emotion labels.

[0974] Step 8:

[0975] The server processes the generated response based on the recognized emotional state, modifying the tone and content of the response based on the emotional label. The input is the emotional label and the chatbot's base response, and the output is the optimized response.

[0976] Step 9:

[0977] The server recommends optimal foods based on the user's dietary preferences and past ordering history, suggesting foods that match the user's immediate emotional state and long-term preferences. The input is user profile information and emotional state, and the output is a list of recommended foods.

[0978] Step 10:

[0979] The user initiates a conversation with the chatbot from their device and receives food recommendations. They then check the response and list of recommended foods received from the server and reflect these in their next conversation or order. The input is the response from the server and the list of recommended foods, and the output is increased user satisfaction.

[0980] 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.

[0981] 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.

[0982] 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.

[0983] [Third embodiment]

[0984] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0985] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0986] 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).

[0987] 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.

[0988] 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.

[0989] 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).

[0990] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for 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.

[0991] 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.

[0992] 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.

[0993] 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.

[0994] 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.

[0995] 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."

[0996] The present invention is a system that collects data on the hobbies, preferences, and conversational style of a specific individual, analyzes the data to create a profile, and generates a replica based on this profile. The generated replica can interact with the user and supports both voice and text. Specific embodiments are described below.

[0997] Data collection

[0998] 1. Terminal: Data entry

[0999] Users use devices such as smartphones and personal computers to input text data in various formats, such as diaries, text messages, emails, and posts on social media.

[1000] The user uses an application on the device to record their own voice, and this voice data contains the target individual's characteristic phrasing and tone of voice.

[1001] 2. Terminal: Data transmission

[1002] The device transmits the collected text and voice data to the server over a secure connection, after which the device notifies the user that the data was successfully transmitted.

[1003] Data analysis

[1004] 1. Server: Data storage

[1005] The server stores the received text data and voice data in a database, which is organized and managed for each user.

[1006] 2. Server: Text data analysis

[1007] The server uses morphological and dependency structure analysis to analyze the text data for language patterns, frequent topics, and specific response tendencies, and the analysis results are stored in a database as user profiles.

[1008] 3. Server: Voice data analysis

[1009] The server uses a voice recognition engine to convert the voice data into text, which is then used for analysis.

[1010] The target individual's vocal characteristics are modeled through spectral analysis, which extracts the frequency components and intonation of the voice.

[1011] Replication

[1012] 1. Server: Chatbot generation

[1013] Based on the results of analyzing the text data, the server generates a chatbot that mimics the conversational style and language patterns of the target individual, and is capable of generating natural responses in everyday conversations and on specific topics.

[1014] 2. Server: Building a speech synthesis model

[1015] The server trains the speech synthesis engine based on the analysis of the voice data and builds a speech synthesis model that reproduces the voice characteristics of a specific individual, allowing it to convert text responses into voice responses.

[1016] User Interaction

[1017] 1. User: Initiate conversation

[1018] Surviving family members and friends can then interact with the replica through a dedicated application, for example by typing, "What should I cook today?"

[1019] 2. Device: Send Message

[1020] The terminal sends this input message to the server.

[1021] 3. Server: Response Generation

[1022] The chatbot on the server generates an appropriate response based on the received message, for example, "How about making curry today?" in text and voice.

[1023] 4. Terminal: Receives response

[1024] The terminal displays and plays the text and voice responses received from the server, allowing the user to confirm and proceed with the next dialogue.

[1025] Specific examples

[1026] For example, if a particular individual frequently talks about movies they liked in the past, then based on that conversation log, when a user asks, "What movie have you seen recently?", the server's chatbot can generate a response such as, "You used to like action movies, so check out these new action movies!"

[1027] This system allows surviving family and friends to receive emotional support by interacting with a replica that faithfully reproduces the individual's hobbies, preferences, and conversation style.

[1028] The processing flow will be explained below.

[1029] Step 1:

[1030] Users use the device's applications to enter information about their diary, conversation logs, and hobbies and preferences, including text messages, emails, and social media posts.

[1031] Users can also use the device's recording function to record voice data, including everyday conversations and voice messages on specific topics.

[1032] Step 2:

[1033] The device sends the collected text and voice data to a server over a secure connection, and the data is encrypted before transmission.

[1034] After the transmission is complete, the terminal notifies the user that the data has been successfully transmitted.

[1035] Step 3:

[1036] The server stores the received text data and voice data in a database.

[1037] The database is divided by user, and each user's data is organized and managed.

[1038] Step 4:

[1039] The server analyzes the collected text data using morphological analysis and dependency structure analysis.

[1040] Specifically, it identifies the parts of speech of words in a sentence, analyzes the structure of the sentence, and extracts frequently occurring words, phrases, and frequently discussed topics, thereby identifying the user's language patterns.

[1041] Step 5:

[1042] The server converts the voice data into text data using a voice recognition engine.

[1043] The converted text data is also analyzed and stored in a database. The server also performs spectral analysis to extract the frequency components of the voice and the features of each phoneme.

[1044] Step 6:

[1045] The server then creates a user profile based on the analysis results, which includes the user's language patterns, interests, and vocal characteristics.

[1046] The profile is stored in a database for later processing.

[1047] Step 7:

[1048] The server generates a chatbot that learns the language patterns of the target user based on the user profile.

[1049] The chatbot is configured to respond to a user's specific phrases and topical trends.

[1050] Step 8:

[1051] Based on the results of the voice data analysis, the server trains the voice synthesis engine and builds a voice synthesis model that reproduces the user's voice characteristics.

[1052] This speech synthesis model is used to convert the generated text response into a speech response.

[1053] Step 9:

[1054] Users can initiate interactions with the created replicas through dedicated applications.

[1055] When the user initiates a conversation, the terminal sends the user's message to the server.

[1056] Step 10:

[1057] The chatbot on the server generates an appropriate response based on the received message.

[1058] For example, if a user asks, "What should I cook today?" the server generates a text response saying, "How about making curry?"

[1059] Step 11:

[1060] The server converts the generated text response into voice using a speech synthesis model.

[1061] This causes the text response to be output as a voice response.

[1062] Step 12:

[1063] The terminal displays and plays back the text and audio responses received from the server.

[1064] The user can then confirm the response and continue the dialogue.

[1065] Example 1

[1066] 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."

[1067] In modern society, after a specific individual passes away, there is a need to build an interactive dialogue system based on that individual's hobbies, preferences, and conversational style. However, there is still a lack of technology to efficiently collect available data, accurately analyze it, and conduct natural dialogue. The purpose of this invention is to solve this problem and provide a system that generates a replica based on the hobbies, preferences, and conversational style of a specific individual, enabling dialogue with the user.

[1068] 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.

[1069] In this invention, the server includes: means for a user to input text data and voice data using a terminal; means for transmitting the input text data and voice data to the server via a secure connection; means for the server to store the received text data and voice data in a database; means for the server to analyze the text data using morphological analysis and dependency structure analysis to extract language patterns and frequent topics; means for the server to convert the voice data into text using a voice recognition engine and model voice characteristics through spectral analysis; means for the server to generate a chatbot and a voice synthesis model that imitate the conversation style and language patterns of a specific individual based on the analysis results of the text data and voice data; and means for engaging in a dialogue with the user using the generated chatbot and voice synthesis model. This makes it possible to faithfully reproduce the hobbies, preferences, and conversation style of a specific individual and realize natural, interactive dialogue.

[1070] A "terminal" is an electronic device used by a user, such as a smartphone or a personal computer.

[1071] "Text data" refers to data of sentences or character strings that a user inputs using a terminal.

[1072] "Audio data" refers to data of audio recorded by a user using a terminal.

[1073] A "secure connection" is an encrypted communication method that protects data from being intercepted or tampered with by third parties during transmission.

[1074] A "server" is a computer system connected to the Internet that receives, analyzes, and stores text and audio data.

[1075] A "database" is a data structure for efficiently organizing and managing data within a server.

[1076] "Morphological analysis" is a natural language processing technique that divides text data into words and identifies their parts of speech in order to understand their meaning.

[1077] "Dependency structure analysis" is a natural language processing technique that analyzes the relationships between words in text data and reveals the structure of sentences.

[1078] A "speech recognition engine" is a software or hardware system for converting voice data into text data.

[1079] "Spectral analysis" is a technique for analyzing the frequency components and intonation of voice data.

[1080] "Voice characteristics" refer to the unique characteristics of individual voices, including, for example, tone, pitch, and intonation.

[1081] A "chatbot" is an automated response system that uses text data to interact with users.

[1082] A "voice synthesis model" is a model for converting text data into voice data based on analyzed voice characteristics.

[1083] A "profile" is information generated based on collected data that summarizes a specific individual's hobbies, preferences, and conversation style.

[1084] The present invention is a system that collects data on the hobbies, preferences, and conversational style of a specific individual, analyzes the data to create a profile, and generates a replica based on this profile. The system analyzes text data and voice data entered by a user using a terminal, and generates a chatbot and a voice synthesis model that mimics the conversational style and language patterns of the specific individual. The following describes how to specifically implement the present invention.

[1085] Data Collection and Transmission

[1086] Users use devices such as smartphones and PCs to input text data in various formats, such as diaries, text messages, emails, and social media posts. For example, a user might input "It was a nice day today" into a dedicated smartphone app. The user might also use an application on the device to record their own voice. This voice data contains the target individual's characteristic phrasing and tone of voice. For example, a user might record and save the voice saying "I slept well today."

[1087] The device sends the collected text data and voice data to the server via a secure connection (such as SSL / TLS). For example, the device sends text data such as "The weather was nice today" and voice data such as "I slept well today" to the server. After sending, the device notifies the user that the data has been sent successfully. For example, the device displays a message on the screen saying "The data has been sent successfully."

[1088] Data Storage and Analysis

[1089] The server stores the received text data and voice data in a database. This database is a data structure for organizing and managing data for each user. For example, the server stores the text data "The weather was nice today" and the voice file "I slept well today" in the "user123" folder.

[1090] The server then uses morphological analysis and dependency structure analysis to analyze the text data for language patterns, frequently occurring topics, and specific response trends. The software used includes the Python morphological analysis library "MeCab" and the dependency structure analysis engine "spaCy." For example, the server analyzes "The weather was nice today" and recognizes that "weather" is a frequently occurring topic and that it is a pleasant day. The analysis results are saved in a database as a user profile. For example, the server can add information to the profile such as "I have many good experiences with the weather."

[1091] The server also uses a speech recognition engine to convert the voice data into text. For this, it uses "Google Cloud Speech-to-Text." For example, it converts the voice data "I slept well today" into text. It then performs spectral analysis to extract the frequency components and intonation of the voice. For this, it uses "Praat." For example, the server analyzes and models the intonation and pitch of the voice "I slept well today."

[1092] Replication and User Interaction

[1093] The server generates a chatbot that mimics the target individual's conversation style and language patterns based on the results of analyzing the text data. The generative AI model used is "OpenAI GPT-4." For example, based on information that the server has "many positive experiences related to the weather," it generates a chatbot that prefers topics related to sunny days.

[1094] The server then trains a speech synthesis engine based on the results of the voice data analysis to build a speech synthesis model that can reproduce the voice characteristics of a specific individual. Engines used include Google Cloud Text-to-Speech and Amazon Polly. For example, the server builds a voice model that can reproduce the tone and tempo of the phrase "I slept well tonight."

[1095] Surviving family and friends can start a conversation with the created replica through a dedicated application. For example, a family member can type, "What should we cook today?" The device sends this input message to the server. The chatbot on the server generates an appropriate response based on the received message. For example, it might generate a response in text and voice, such as, "How about cooking curry today?" The device displays and plays back the text and voice response received from the server. The user can confirm this and carry out the next conversation.

[1096] For example, if a particular individual frequently talked about their favorite movies in the past, when a user asks "What movie did you see recently?" based on the conversation log, the chatbot on the server can generate a response such as "You used to like action movies, so check out the new action movies!" In this way, surviving family and friends can receive emotional support by interacting with a replica that faithfully reproduces the hobbies, preferences, and conversation style of the particular individual.

[1097] Example prompt sentence:

[1098] "What activities have you been interested in lately?"

[1099] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1100] Step 1:

[1101] A user inputs text data and voice data using a device. Using an application on a smartphone or computer, the user inputs text data in the form of a diary entry, text message, email, or social media post. The user also uses an application to record their own voice. For example, a user inputs text data such as "It was nice weather today" into a dedicated smartphone application and records voice such as "I slept well today." The input data is text data and voice data.

[1102] Step 2:

[1103] The device sends the collected text and voice data to the server through a secure connection. Specifically, the data is sent using encrypted communication such as SSL / TLS. When the transmission is complete, the device displays a notification to the user stating, "The data was sent successfully." The input data is the text and voice data entered by the user, and the output data is a notification indicating that the data has been sent.

[1104] Step 3:

[1105] The server stores the received text data and voice data in a database. Specifically, the server saves the text data "It was nice weather today" and the voice file "I slept well today" in the "user123" folder. The input data is the transmitted text data and voice data, and the output data is the data stored in the database.

[1106] Step 4:

[1107] The server analyzes the text data using morphological analysis and dependency structure analysis. Specifically, it uses the Python morphological analysis library "MeCab" and the dependency structure analysis engine "spaCy" to extract language patterns, frequently occurring topics, and specific reaction trends from the text data. For example, the server analyzes the text "It was nice weather today" and recognizes that "weather" is a frequently occurring topic. The input data is text data, and the output data is the analyzed language patterns and frequently occurring topics.

[1108] Step 5:

[1109] The server converts the voice data into text using a speech recognition engine. Specifically, it uses "Google Cloud Speech-to-Text" to convert the voice data "I slept well today" into text. It also performs spectral analysis to extract the frequency components and intonation of the voice. For example, using "Praat," the server analyzes and models the intonation and pitch of the voice "I slept well today." The input data is the voice data, and the output data is the analyzed voice characteristics and converted text data.

[1110] Step 6:

[1111] Based on the analysis results of the text and voice data, the server generates a chatbot and voice synthesis model that mimics the conversation style and language patterns of a specific individual. Specifically, it uses the generative AI model "OpenAI GPT-4" to generate a chatbot based on the text data, and builds a voice model using the voice synthesis engines "Google Cloud Text-to-Speech" and "Amazon Polly." The input data is the analysis results, and the output data is the chatbot and voice synthesis model.

[1112] Step 7:

[1113] The user initiates a conversation with the generated replica through a terminal. Specifically, the user inputs a message into a dedicated application, such as "What should I cook today?" The input message is sent to the server, and the chatbot on the server generates a response in text and voice, such as "How about making curry?" The input data is the message from the user, and the output data is the text and voice response from the chatbot.

[1114] Step 8:

[1115] The terminal displays and plays back the text and voice response received from the server. Specifically, it displays the chatbot's response text on the terminal's display and plays back the response aloud. For example, the terminal displays the text "How about making curry?" on the screen and plays back the voice. This allows the user to carry out the next dialogue. The input data is the text and voice response from the server, and the output data is the displayed text and played back voice.

[1116] An example of a prompt sentence: "What activities have you been interested in lately?"

[1117] (Application example 1)

[1118] 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."

[1119] In today's brick-and-mortar stores, it is difficult to provide personalized product recommendations and assistance to each individual consumer. Conventional services often provide general information and uniform customer service to all customers, failing to address individual preferences and needs. This can result in low user satisfaction and a loss of purchasing motivation. Furthermore, there is a lack of technology to effectively utilize profiles that utilize a specific individual's conversation style and preferences using a visual display device. Therefore, there is a need for the development of a system that can provide personalized product recommendations and assistance based on a specific individual's profile through a visual display device.

[1120] 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.

[1121] In this invention, the server includes means for collecting data on the hobbies, preferences, and conversation style of a specific individual, means for analyzing the collected data to create a profile, means for generating a replica of the specific individual based on the profile, means for interacting with the user using the generated replica, and means for providing customized product suggestions and assistance based on the analyzed profile using a visual display device attached to the user's communication device. This makes it possible to effectively provide personalized product suggestions and assistance based on the profile of a specific individual in a physical store.

[1122] "Hobbies" are preferences regarding the things and activities that a particular individual likes.

[1123] A "conversational style" is the linguistic expression or method of communication that a particular individual normally uses.

[1124] A "data collection instrument" is a device or method for collecting information about the tastes, preferences, and speaking styles of a particular individual.

[1125] A "profile" is a collection of information that describes the characteristics of a particular individual, created based on collected data.

[1126] A "replica" is an interactive virtual assistant generated based on a specific individual's profile.

[1127] The "interaction means" is a method or device for communicating with a user using the generated replica.

[1128] A "visual display device" is a communication device worn by a user that visually displays text and images.

[1129] "Product Recommendation" is a method or device aimed at recommending personalized products and services based on a user's profile.

[1130] "Assistance" refers to help or support provided based on a specific individual's preferences and conversational style.

[1131] This invention is a "personalized shopping assistant" system that provides personalized product recommendations and assistance in brick-and-mortar stores. It uses a visual display device attached to a user's communication device. Specifically, it is implemented using smart glasses. The system of this invention suggests products and services based on the user's tastes, preferences, and conversation style, providing an experience tailored to individual needs.

[1132] System Configuration and Operation

[1133] 1. Hardware and Software:

[1134] Hardware: Smart glasses (with voice recognition microphone and speaker), server, microphone

[1135] Software: GPT-2, speech recognition library (speech_recognition), speech synthesis engine (pyttsx3), Python

[1136] The smart glasses recognize the user's voice and send the data to the server, which analyzes the data and generates a response based on the user's profile, which is presented visually and audibly to the smart glasses.

[1137] 2. Data collection and analysis:

[1138] The user puts on the smart glasses and speaks their questions or requests. The speech recognition library, speech_recognition, converts the speech into text data.

[1139] The voice and text data sent to the server is used to generate appropriate responses based on the profile using GPT-2.

[1140] 3. Create a profile:

[1141] The server collects data about the user's interests and conversational style and creates a profile based on that data, including the types of products the user likes and conversational habits.

[1142] Through a speech recognition engine and speech synthesis engine, the user's language patterns and voice characteristics are analyzed and reflected in the profile.

[1143] 4. Generate and present the response:

[1144] When a user speaks to the smart glasses, the server creates a prompt based on the content and generates a response using the GPT-2 model.

[1145] The generated text response is converted to speech using the speech synthesis engine pyttsx3 and played through the smart glasses' speakers, and also displayed as text on the visual display.

[1146] Examples:

[1147] For example, if a user speaks to the smart glasses, "What kind of food would you like today?", the microphone in the smart glasses will capture the voice, and the speech recognition library, speech_recognition, will convert it into text. This text data will be sent to the server and processed as the following prompt sentence:

[1148] User: What would you like to eat today?

[1149] Assistant: The weather is nice today, why not enjoy a barbecue?

[1150] The responses are converted to speech and played back to the user, and are also displayed as text on a visual display, allowing for personalized product recommendations and assistance based on the user's preferences and conversational style.

[1151] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1152] Step 1:

[1153] User: Voice Input

[1154] Users wear smart glasses and input questions or requests by voice.

[1155] Input: User's voice data

[1156] Output: Audio data is captured

[1157] Specific operation: A microphone installed in the smart glasses collects the user's voice.

[1158] Step 2:

[1159] Device: Voice recognition

[1160] The device uses the speech recognition library speech_recognition to convert the voice data into text data.

[1161] Input: Captured audio data

[1162] Output: Text data

[1163] Specific operation: Audio data collected from the microphone is sent to the speech recognition library, and corresponding text data is generated.

[1164] Step 3:

[1165] Terminal: Data transmission

[1166] The terminal transmits the converted text data to the server over a secure connection.

[1167] Input: Text data

[1168] Output: Text data sent to the server

[1169] Specific operation: The converted text data is encrypted and sent securely to the server.

[1170] Step 4:

[1171] Server:Profile Creation

[1172] The server analyzes the received text data and creates a profile of the user's interests, preferences and conversation style.

[1173] Input: Text data

[1174] Output: User profile

[1175] Specific operation: Through morphological analysis and dependency structure analysis, language patterns and preferences are extracted from text data and a profile is generated.

[1176] Step 5:

[1177] Server: Response Generation

[1178] The server uses the generative AI model GPT-2 to create appropriate prompts and generate responses based on the user's profile.

[1179] Input: Profile and text data

[1180] Output: Text response

[1181] How it works: A prompt sentence is input into the GPT-2 model, which generates a text response based on the user's interests and preferences.

[1182] Example prompt sentence:

[1183] User: What would you like to eat today?

[1184] Assistant: The weather is nice today, why not enjoy a barbecue?

[1185] Step 6:

[1186] Server: Speech synthesis

[1187] The server converts the text response into speech using the speech synthesis engine pyttsx3.

[1188] Input: Text response

[1189] Output: Audio data

[1190] What it does: The generated text response is fed into a speech synthesis engine to generate a voice that sounds similar to a human voice.

[1191] Step 7:

[1192] Terminal: Response presentation

[1193] The terminal plays the generated audio data through the speaker of the smart glasses and displays the text data on a visual display.

[1194] Input: Audio and text data

[1195] Output: Presents the response to the user

[1196] Specific Actions: Provide a response to the user by playing audio data over a speaker and displaying text data on a visual display device.

[1197] 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.

[1198] The present invention is a system that collects data on a specific individual's hobbies, preferences, and conversation style, analyzes the data, and creates a profile. It then generates a replica based on the profile, and combines this with an emotion engine that recognizes the user's emotions. The generated replica can converse with the user and adjusts its responses according to the user's emotional state. Specific embodiments are described below.

[1199] Data collection

[1200] 1. Terminal: Data entry

[1201] Users use devices such as smartphones and personal computers to input text data in various formats, such as diaries, text messages, emails, and posts on social media.

[1202] The user uses an application on the device to record their own voice, and this voice data contains the target individual's characteristic phrasing and tone of voice.

[1203] 2. Terminal: Data transmission

[1204] The device transmits the collected text and voice data to the server over a secure connection, after which the device notifies the user that the data was successfully transmitted.

[1205] Data analysis

[1206] 1. Server: Data storage

[1207] The server stores the received text data and voice data in a database.

[1208] Within the database, data is organized and managed for each user.

[1209] 2. Server: Text data analysis

[1210] The server uses morphological and dependency structure analysis to analyze the text data for language patterns, frequent topics, and specific response tendencies, and the analysis results are stored in a database as user profiles.

[1211] 3. Server: Voice data analysis

[1212] The server uses a voice recognition engine to convert the voice data into text, and based on this, the voice data is also used for analysis.

[1213] The server performs spectral analysis to extract the frequency components and intonation of the voice, thereby modeling the voice characteristics.

[1214] Replication

[1215] 1. Server: Chatbot generation

[1216] Based on the results of analyzing the text data, the server generates a chatbot that has learned the target individual's language patterns and is capable of generating natural responses in everyday conversations and on specific topics.

[1217] 2. Server: Building a speech synthesis model

[1218] The server trains the speech synthesis engine based on the analysis of the voice data and builds a speech synthesis model that reproduces the voice characteristics of a specific individual, allowing it to convert text responses into voice responses.

[1219] Adding an Emotion Engine

[1220] 1. Server: User emotion recognition

[1221] The server uses an emotion engine to recognize emotions from the user's voice and text data, for example by analyzing voice characteristics such as tone, speed, and volume, and from text data by analyzing the context and emotional tendencies of selected words.

[1222] 2. Server: Emotion-based profile update

[1223] Based on the recognized emotions, the user profile is updated, which includes the user's emotional tendencies and current emotional state.

[1224] 3. Server: Emotion-based response generation

[1225] The generated response is tailored based on the emotional state recognized by the emotion engine. For example, if the user asks a question in a sad voice, the server will generate a more empathetic and comforting response.

[1226] User Interaction

[1227] 1. User: Initiate conversation

[1228] Family and friends can then interact with the replica through a dedicated application, which sends text and voice messages to the server.

[1229] 2. Device: Send Message

[1230] The terminal transmits the input text message and voice data to the server.

[1231] 3. Server: Response Generation

[1232] The server's chatbot generates an appropriate response based on the received message and emotional state. For example, if the user says "Today was tough" in a tired voice, the server generates a response like "Today was tough. Let's take a break."

[1233] 4. Server: Speech synthesis

[1234] The server converts the generated text response into voice using a speech synthesis model, which outputs the text response as a voice response.

[1235] 5. Terminal: Receives response

[1236] The terminal displays and plays back the text and voice responses received from the server, allowing the user to confirm the responses and continue the dialogue.

[1237] Specific examples

[1238] For example, if a specific individual has often talked about "movies" in the past, when the user asks "What movie did you see recently?", the chatbot on the server will generate a response such as "There was a new action movie recently that was good. You should definitely check it out" based on the user's preferences and emotional state. Also, by recognizing the user's emotional state, if the user is feeling down, for example, an encouraging response such as "Cheer up. Let's find a movie we can enjoy together" will be generated.

[1239] This system allows surviving family and friends to receive emotional support by interacting with a replica that faithfully reproduces the individual's hobbies, preferences, and conversation style. Furthermore, the emotional engine recognizes the user's emotional state and generates responses accordingly, enabling more empathetic and natural conversations.

[1240] The processing flow will be explained below.

[1241] Step 1:

[1242] Users can use the device's applications to enter information about their diary, conversation log, and hobbies and interests in text format, and also record voice messages using the device's recording function.

[1243] Step 2:

[1244] The device sends the collected text and voice data to the server over a secure connection, and the user is notified once the data has been sent.

[1245] Step 3:

[1246] The server stores the received text data in a database, and simultaneously stores the audio data in the database.

[1247] Step 4:

[1248] The server analyzes the received text data using morphological and dependency analysis to identify the user's language patterns and frequent topics and create a profile.

[1249] Step 5:

[1250] The server converts the voice data into text data using a speech recognition engine, and then performs spectral analysis to extract features for each frequency component and phoneme.

[1251] Step 6:

[1252] The server stores a user profile based on the analysis results in a database, which includes language patterns, interests, and voice characteristics.

[1253] Step 7:

[1254] The server generates a chatbot based on the profile, which is trained to replicate the language patterns of the specific individual and to engage in natural conversations.

[1255] Step 8:

[1256] The server uses the speech analysis results to train a speech synthesis engine and build a speech synthesis model that replicates the voice characteristics of a particular individual, which is then used to convert the text response into a speech response.

[1257] Step 9:

[1258] The user initiates a dialogue with the created replica through a dedicated application. At the beginning of the dialogue, the terminal sends the user's input to the server.

[1259] Step 10:

[1260] The server uses an emotion engine to recognize emotions from user input data, analyzing voice tone, speed, volume, etc. from voice data, and emotional trends from text data.

[1261] Step 11:

[1262] The server's chatbot generates more empathetic and appropriate responses based on the emotions recognized by the emotion engine. For example, if the user speaks in a sad voice, the server will generate a comforting response.

[1263] Step 12:

[1264] The server converts the generated text response into voice using a speech synthesis model, which outputs the text response as a voice response that reflects the appropriate emotion.

[1265] Step 13:

[1266] The device displays and plays the text and voice responses received from the server to the user, who can then confirm the responses and continue the dialogue.

[1267] Example 2

[1268] 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."

[1269] In modern society, emotional support gained through memories and conversations with specific individuals is highly valued. However, when a specific individual has passed away or is far away, means of sharing conversations and memories with that person are limited. Conventional technologies simply play back text or audio data, and do not realize natural conversations that reflect the user's emotions, hobbies, and preferences. Therefore, the present invention calls for a new system that faithfully reproduces the hobbies, preferences, and conversation style of a specific individual, enabling conversations that correspond to the user's emotional state.

[1270] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes: means for collecting data on the hobbies, preferences, and conversation style of a specific individual; means for encrypting the collected text data and voice data and transmitting it through a secure connection; means for analyzing the collected data to create a user profile; means for generating a chatbot and a voice synthesis model for the specific individual based on the user profile; means for recognizing the user's emotions using an emotion engine and updating the profile; and means for interacting with the user using the generated replica and adjusting responses based on the user's emotional state. This makes it possible to interactively interact with a replica that faithfully reproduces the characteristics of a specific individual and obtain responses that correspond to the user's emotional state.

[1271] 1. "Specified Individual" refers to a specific person about whom data is collected and analyzed by the system.

[1272] 2. "Preferences" refers to information about a particular individual's preferred activities, interests, and preferences.

[1273] 3. "Conversational style" refers to a particular individual's speaking style, phrasing, language patterns, and communication characteristics.

[1274] 4. "Data Collection Instrument" means the part of the system that captures text and voice data about a specific individual.

[1275] 5. "Data transmission means" refers to the function for encrypting collected data and transmitting it securely to the server.

[1276] 6. "Data analysis means" refers to the part of the system that analyzes collected data to extract and understand hobbies, preferences, and conversation styles.

[1277] 7. "User profile" refers to information that reproduces the characteristics of a specific individual, generated based on analyzed hobbies, preferences, and conversation style.

[1278] 8. "Chatbot generation means" refers to a function for generating a chatbot that reproduces the language patterns of a specific individual based on a user profile.

[1279] 9. "Voice synthesis model generation means" refers to a function for generating a voice synthesis model that reproduces the voice characteristics of a specific individual based on the results of voice data analysis.

[1280] 10. "Emotion engine" refers to software or algorithms that recognize and analyze user emotions.

[1281] 11. "Profile Updater" refers to functionality for updating a user profile based on the user's emotional state recognized by the Emotion Engine.

[1282] 12. "Response adjustment means" refers to a function that enables the generated replica to generate a response according to the emotional state during interaction with the user.

[1283] 13. "Replica" refers to a simulation that faithfully reproduces the tastes, preferences, and conversation style of a specific individual.

[1284] The present invention is a system that collects data on a specific individual's hobbies, preferences, and conversation style, analyzes the data, and creates a profile. A replica is generated based on this profile, and by combining it with an emotion engine that recognizes the user's emotions, a system is provided that can interact with the user. Specific embodiments are described below.

[1285] Data collection

[1286] 1. Terminal: Data entry

[1287] Users use devices such as smartphones and PCs to input various forms of text data, such as diaries, text messages, emails, and posts on social media. Users also use applications on their devices to record their own voice. This voice data includes the target individual's characteristic phrasing and tone of voice.

[1288] 2. Terminal: Data transmission

[1289] The device encrypts the collected text and voice data and sends it to the server over a secure connection, after which the device notifies the user that the data has been successfully sent.

[1290] Data analysis

[1291] 1. Server: Data storage

[1292] The server stores the received text and voice data in a database, where data is organized and managed for each user.

[1293] 2. Server: Text data analysis

[1294] The server uses morphological and dependency structure analysis to analyze the text data for language patterns, frequent topics, and specific response tendencies, and the analysis results are stored in a database as user profiles.

[1295] 3. Server: Voice data analysis

[1296] The server converts the voice data into text using a speech recognition engine. Based on this, the voice data is also analyzed. The server also performs spectral analysis to extract the frequency components and intonation of the voice and model the voice characteristics.

[1297] Replication

[1298] 1. Server: Chatbot generation

[1299] Based on the results of analyzing the text data, the server generates a chatbot that has learned the target individual's language patterns and is capable of generating natural responses in everyday conversations and on specific topics.

[1300] 2. Server: Building a speech synthesis model

[1301] The server trains the speech synthesis engine based on the analysis of the voice data and builds a speech synthesis model that reproduces the voice characteristics of a specific individual, allowing it to convert text responses into voice responses.

[1302] Adding an Emotion Engine

[1303] 1. Server: User emotion recognition

[1304] The server uses an emotion engine to recognize emotions from the user's voice and text data, for example by analyzing voice characteristics such as tone, speed, and volume, and from text data by analyzing the context and emotional tendencies of selected words.

[1305] 2. Server: Emotion-based profile update

[1306] Based on the recognized emotions, the user profile is updated, which includes the user's emotional tendencies and current emotional state.

[1307] 3. Server: Emotion-based response generation

[1308] The generated response is tailored based on the perceived emotional state: for example, if the user asks a question in a sad voice, the server will generate a more empathetic and comforting response.

[1309] User Interaction

[1310] 1. User: Initiate conversation

[1311] Users can start interacting with the replicas through a dedicated application, and when a user initiates an interaction, the device sends text and voice messages to the server.

[1312] 2. Device: Send Message

[1313] The terminal transmits the input text message and voice data to the server.

[1314] 3. Server: Response Generation

[1315] The server's chatbot generates an appropriate response based on the received message and emotional state. For example, if the user says "Today was tough" in a tired voice, the server generates a response like "Today was tough. Let's take a break."

[1316] 4. Server: Speech synthesis

[1317] The server converts the generated text response into voice using a speech synthesis model, which outputs the text response as a voice response.

[1318] 5. Terminal: Receives response

[1319] The terminal displays and plays back the text and voice responses received from the server, allowing the user to confirm the responses and continue the dialogue.

[1320] Specific examples and prompts for the generative AI model

[1321] Specific examples

[1322] For example, if a specific individual has often talked about "movies" in the past, when the user asks "What movie did you see recently?", the chatbot on the server will generate a response such as "There was a new action movie recently that was good. You should definitely check it out" based on the user's preferences and emotional state. Also, by recognizing the user's emotional state, if the user is feeling down, for example, an encouraging response such as "Cheer up. Let's find a movie we can enjoy together" will be generated.

[1323] Prompt Sentence Examples

[1324] 1. Prompt to analyze user text data

[1325] "Analyze text data from diaries, text messages, and social media posts using morphological and dependency analysis. Extract language patterns, frequently occurring topics, and specific reaction tendencies for each user to create a user profile."

[1326] 2. Prompt to request voice data analysis

[1327] "Use a speech recognition engine to convert the speech data into text, then perform spectral analysis to extract the frequency components and intonation of the speech. Based on the results, model the speech characteristics."

[1328] 3. Prompt to generate a chatbot

[1329] "Based on the results of analyzing the text data, create a chatbot that learns the user's language patterns. This chatbot should be able to generate natural responses in everyday conversations and on specific topics."

[1330] 4. Prompt to request emotion engine analysis

[1331] "Analyze voice data for characteristics such as tone, speed, and volume, and text data for context and emotional trends in chosen words. Based on the results, recognize the user's emotions and update the user profile."

[1332] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1333] System program processing flow

[1334] Step 1: Data entry

[1335] Users use smartphones or personal computers to input text data in various formats, such as diary entries, text messages, emails, and social media posts, into dedicated applications.

[1336] Input: Text data (e.g., diary entries, messages, posts)

[1337] Output: Collected text data

[1338] The user uses the same application to record their own voice, and the recorded voice data includes the user's characteristic phrasing and tone of voice.

[1339] Input: Audio data (e.g., recording of everyday conversation)

[1340] Output: Collected audio data

[1341] Step 2: Send data

[1342] The device encrypts the collected text and voice data and sends it to a server over a secure connection.

[1343] Input: Collected text and audio data

[1344] Output: Sending encrypted data to the server

[1345] After transmission, the terminal notifies the user that the data has been successfully transmitted.

[1346] Input: Send result

[1347] Output: User notification

[1348] Step 3: Data storage

[1349] The server stores the received text and voice data in a database, where data is organized and managed for each user.

[1350] Input: Received text and audio data

[1351] Output: Data stored in a database

[1352] Step 4: Text data analysis

[1353] The server performs morphological analysis and breaks down the text data into words.

[1354] Input: Text data

[1355] Output: A list of decomposed words

[1356] The server then uses dependency analysis to analyze grammatical and semantic relationships.

[1357] Input: A list of decomposed words

[1358] Output: Structural data showing grammatical and semantic relationships

[1359] Step 5: Audio data analysis

[1360] The server converts the voice data into text data using a voice recognition engine.

[1361] Input: Audio data

[1362] Output: Character data

[1363] In addition, the server performs spectral analysis to extract the frequency components and intonation of the voice.

[1364] Input: Audio data

[1365] Output: Frequency content and intonation data

[1366] Step 6: Create a user profile

[1367] The server creates a user profile based on the results of the analysis of the text and voice data.

[1368] Input: Analysis results (text data and audio data)

[1369] Output: User profile

[1370] Step 7: Chatbot generation

[1371] The server generates a chatbot that learns the user's language patterns based on the user profile.

[1372] Input: User Profile

[1373] Output: Chatbot

[1374] Step 8: Building a speech synthesis model

[1375] Based on the results of the voice data analysis, the server constructs a voice synthesis model that reproduces the voice characteristics of a specific individual.

[1376] Input: Audio data analysis results

[1377] Output: Speech synthesis model

[1378] Step 9: Emotion Recognition and Profile Update

[1379] The server uses an emotion engine to recognize the user's emotions and reflect them in the profile.

[1380] Input: Voice and text data

[1381] Output: Updated user profile

[1382] Step 10: Start a conversation

[1383] The user initiates a conversation with the created replica through a dedicated application. When the conversation begins, the terminal sends the user's text and voice messages to the server.

[1384] Input: User's message (text and voice)

[1385] Output: Send to server

[1386] Step 11: Send the message

[1387] The terminal transmits the input text message and voice data to the server.

[1388] Input: User message

[1389] Output: Data sent to the server

[1390] Step 12: Response Generation

[1391] The chatbot on the server generates an appropriate response based on the received message and emotional state.

[1392] Input: User message and emotion data

[1393] Output: Response message

[1394] Step 13: Speech synthesis

[1395] The server uses a speech synthesis model to convert the generated text response into a voice response.

[1396] Input: Text response

[1397] Output: Voice response

[1398] Step 14: Receive a response

[1399] The terminal displays and plays back the text and voice responses received from the server, allowing the user to confirm the responses and continue the dialogue.

[1400] Input: Response from the server (text and audio)

[1401] Output: Displayed and played response

[1402] (Application example 2)

[1403] 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."

[1404] Conventional food delivery services only suggest generic menus without considering the user's individual preferences or emotional state, resulting in a lack of personalized user experience. Even systems that reflect the user's preferences and conversational style lack the ability to generate appropriate responses based on the user's emotions, resulting in low user satisfaction.

[1405] 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.

[1406] In this invention, the server includes means for collecting data on the hobbies, preferences, and conversation style of a specific individual, means for analyzing the collected data to create a profile, means for generating a replica of the specific individual based on the profile, means for interacting with a user using the generated replica, means for analyzing the emotional state of the user, means for adjusting responses to the user based on the emotional state, means for recommending foods based on the individual's eating preferences and past ordering history, and means for using the responses in the food recommendation, thereby enabling a personalized food delivery experience that reflects the user's individual preferences and emotions.

[1407] "The hobbies and preferences of a particular individual" refers to the things that a particular individual likes or is interested in.

[1408] "Conversational style" refers to the language patterns and speaking characteristics that a particular individual uses in everyday interactions.

[1409] "Data collection means" refers to devices or methods for collecting information such as text, audio, or images.

[1410] "Means for analyzing data to create profiles" means processes or software that analyze collected data and generate profiles that reflect the characteristics of specific individuals.

[1411] "Means for generating a replica of a specific individual" refers to equipment or methods for creating a virtual entity (replica) that imitates a specific individual based on the results of analysis.

[1412] "Means for interacting with the user" refers to the interface and functions that allow the replica to communicate with the user.

[1413] "Means for analyzing emotional state" refers to analytical engines and algorithms for reading emotions from the user's voice or text.

[1414] "Means for adjusting responses" refers to techniques or systems for appropriately changing the content of responses generated based on the user's emotional state.

[1415] "Dietary preferences" refer to the types of foods and beverages that a particular individual prefers.

[1416] "Past Order History" refers to a record of food and beverage orders a user has placed.

[1417] "Food recommendation tools" refers to algorithms or systems that suggest appropriate foods based on a user's preferences and ordering history.

[1418] The present invention provides a system for enhancing user experience based on the tastes, preferences, and conversational styles of specific individuals. In particular, an example of application to a food delivery service will be described in detail.

[1419] System Overview

[1420] The system uses the following major hardware and software:

[1421] User's device (smartphone, tablet, etc.)

[1422] Server (performs data collection and analysis, profile generation, replica generation, and response generation)

[1423] Sentiment analysis engine (Hugging Face Transformers)

[1424] Response generation engine (OpenAI's GPT-3)

[1425] Data collection

[1426] Users can use the device to input various types of text data, such as diary entries, text messages, emails, and social media posts, as well as record their own voice. This data is then sent to a server via a secure connection and stored in a database.

[1427] Profile Generation

[1428] The server analyzes the collected text and speech data. It uses morphological and dependency structure analysis to extract language patterns and frequently occurring topics from the text data, and creates a user profile based on these analysis results. It also converts speech data into text using a speech recognition engine and performs spectral analysis to model speech characteristics.

[1429] Replication and Interaction

[1430] The server then generates a chatbot that has learned the target individual's language patterns based on the profile. This chatbot is capable of generating natural responses based on everyday conversations and specific topics. It also trains a speech synthesis engine based on the results of voice data analysis, building a speech synthesis model that reproduces the specific individual's voice characteristics.

[1431] Sentiment Analysis and Response Adjustment

[1432] The server uses an emotion analysis engine to recognize emotions from the user's voice and text data. Based on the recognized emotional state, the profile is updated and the generated response is adjusted. For example, if the user says "Today was tough" in a tired voice, the server will generate a compassionate response such as "Today was tough. Let's take a break."

[1433] Food recommendation function

[1434] Furthermore, the system recommends foods based on an individual's dietary preferences and past ordering history. It combines the user's emotional state and profile information to make optimal food suggestions. For example, if a user says, "I'm feeling low today," the server will recommend foods along with a response like, "Maybe ice cream would be good. Something sweet will definitely cheer you up."

[1435] Specific examples

[1436] When a user types "What's your recommendation today?" into a food concierge app, the system analyzes the words for emotional nuances and suggests the best food options. An example prompt could be, "Generate a response as if the user were talking to a close friend: What's your recommendation today?"

[1437] This enables a personalized food delivery experience that reflects the user's individual preferences and emotions.

[1438] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1439] Step 1:

[1440] A user uses a device to input text data in various formats, such as diary entries, text messages, emails, and social media posts, and records audio data. This data is sent from the device to a server via a secure connection. The input data is text data and audio data, and is sent via secure communication. The output is a notification to the server that the data has been successfully received.

[1441] Step 2:

[1442] The server stores the received text and voice data in a database. The server organizes and manages this data, saving it in a different folder for each user. The input at this stage is the text and voice data sent by the user, and the output is storage in the database.

[1443] Step 3:

[1444] The server performs morphological and dependency structure analysis on the collected text data. Specifically, it uses a natural language processing engine to analyze the text data and extract language patterns, frequently occurring topics, and specific response trends. The input is the text data stored in the database, and the output is a user profile based on the analysis results.

[1445] Step 4:

[1446] The server uses a speech recognition engine to convert the speech data into text and performs spectral analysis, which extracts the frequency components and intonation of the speech and models the speech characteristics. The input is the speech data stored in the database, and the output is the speech characteristic model.

[1447] Step 5:

[1448] The server generates a chatbot that learns the target individual's language patterns based on the profile information. The generated chatbot has the ability to generate natural responses to everyday conversations and specific topics. The input is text data and the results of its analysis, and the output is the chatbot.

[1449] Step 6:

[1450] The server trains the speech synthesis engine based on the voice characteristic model and constructs a voice synthesis model that reproduces the voice characteristics of a specific individual. The input is the voice characteristic model and the output is the voice synthesis model.

[1451] Step 7:

[1452] The server uses an emotion analysis engine to recognize emotions from the user's text and voice. It obtains emotion labels (e.g., "positive" or "negative") and adds them to the profile. The input is the user's text and voice data, and the output is the emotion labels.

[1453] Step 8:

[1454] The server processes the generated response based on the recognized emotional state, modifying the tone and content of the response based on the emotional label. The input is the emotional label and the chatbot's base response, and the output is the optimized response.

[1455] Step 9:

[1456] The server recommends optimal foods based on the user's dietary preferences and past ordering history, suggesting foods that match the user's immediate emotional state and long-term preferences. The input is user profile information and emotional state, and the output is a list of recommended foods.

[1457] Step 10:

[1458] The user initiates a conversation with the chatbot from their device and receives food recommendations. They then check the response and list of recommended foods received from the server and reflect these in their next conversation or order. The input is the response from the server and the list of recommended foods, and the output is increased user satisfaction.

[1459] 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.

[1460] 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.

[1461] 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.

[1462] [Fourth embodiment]

[1463] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1464] 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.

[1465] 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).

[1466] 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.

[1467] 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.

[1468] 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).

[1469] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for 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.

[1470] 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.

[1471] 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.

[1472] 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.

[1473] 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.

[1474] 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.

[1475] 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."

[1476] The present invention is a system that collects data on the hobbies, preferences, and conversational style of a specific individual, analyzes the data to create a profile, and generates a replica based on this profile. The generated replica can interact with the user and supports both voice and text. Specific embodiments are described below.

[1477] Data collection

[1478] 1. Terminal: Data entry

[1479] Users use devices such as smartphones and personal computers to input text data in various formats, such as diaries, text messages, emails, and posts on social media.

[1480] The user uses an application on the device to record their own voice, and this voice data contains the target individual's characteristic phrasing and tone of voice.

[1481] 2. Terminal: Data transmission

[1482] The device transmits the collected text and voice data to the server over a secure connection, after which the device notifies the user that the data was successfully transmitted.

[1483] Data analysis

[1484] 1. Server: Data storage

[1485] The server stores the received text data and voice data in a database, which is organized and managed for each user.

[1486] 2. Server: Text data analysis

[1487] The server uses morphological and dependency structure analysis to analyze the text data for language patterns, frequent topics, and specific response tendencies, and the analysis results are stored in a database as user profiles.

[1488] 3. Server: Voice data analysis

[1489] The server uses a voice recognition engine to convert the voice data into text, which is then used for analysis.

[1490] The target individual's vocal characteristics are modeled through spectral analysis, which extracts the frequency components and intonation of the voice.

[1491] Replication

[1492] 1. Server: Chatbot generation

[1493] Based on the results of analyzing the text data, the server generates a chatbot that mimics the conversational style and language patterns of the target individual, and is capable of generating natural responses in everyday conversations and on specific topics.

[1494] 2. Server: Building a speech synthesis model

[1495] The server trains the speech synthesis engine based on the analysis of the voice data and builds a speech synthesis model that reproduces the voice characteristics of a specific individual, allowing it to convert text responses into voice responses.

[1496] User Interaction

[1497] 1. User: Initiate conversation

[1498] Surviving family members and friends can then interact with the replica through a dedicated application, for example by typing, "What should I cook today?"

[1499] 2. Device: Send Message

[1500] The terminal sends this input message to the server.

[1501] 3. Server: Response Generation

[1502] The chatbot on the server generates an appropriate response based on the received message, for example, "How about making curry today?" in text and voice.

[1503] 4. Terminal: Receives response

[1504] The terminal displays and plays the text and voice responses received from the server, allowing the user to confirm and proceed with the next dialogue.

[1505] Specific examples

[1506] For example, if a particular individual frequently talks about movies they liked in the past, then based on that conversation log, when a user asks, "What movie have you seen recently?", the server's chatbot can generate a response such as, "You used to like action movies, so check out these new action movies!"

[1507] This system allows surviving family and friends to receive emotional support by interacting with a replica that faithfully reproduces the individual's hobbies, preferences, and conversation style.

[1508] The processing flow will be explained below.

[1509] Step 1:

[1510] Users use the device's applications to enter information about their diary, conversation logs, and hobbies and preferences, including text messages, emails, and social media posts.

[1511] Users can also use the device's recording function to record voice data, including everyday conversations and voice messages on specific topics.

[1512] Step 2:

[1513] The device sends the collected text and voice data to a server over a secure connection, and the data is encrypted before transmission.

[1514] After the transmission is complete, the terminal notifies the user that the data has been successfully transmitted.

[1515] Step 3:

[1516] The server stores the received text data and voice data in a database.

[1517] The database is divided by user, and each user's data is organized and managed.

[1518] Step 4:

[1519] The server analyzes the collected text data using morphological analysis and dependency structure analysis.

[1520] Specifically, it identifies the parts of speech of words in a sentence, analyzes the structure of the sentence, and extracts frequently occurring words, phrases, and frequently discussed topics, thereby identifying the user's language patterns.

[1521] Step 5:

[1522] The server converts the voice data into text data using a voice recognition engine.

[1523] The converted text data is also analyzed and stored in a database. The server also performs spectral analysis to extract the frequency components of the voice and the features of each phoneme.

[1524] Step 6:

[1525] The server then creates a user profile based on the analysis results, which includes the user's language patterns, interests, and vocal characteristics.

[1526] The profile is stored in a database for later processing.

[1527] Step 7:

[1528] The server generates a chatbot that learns the language patterns of the target user based on the user profile.

[1529] The chatbot is configured to respond to a user's specific phrases and topical trends.

[1530] Step 8:

[1531] Based on the results of the voice data analysis, the server trains the voice synthesis engine and builds a voice synthesis model that reproduces the user's voice characteristics.

[1532] This speech synthesis model is used to convert the generated text response into a speech response.

[1533] Step 9:

[1534] Users can initiate interactions with the created replicas through dedicated applications.

[1535] When the user initiates a conversation, the terminal sends the user's message to the server.

[1536] Step 10:

[1537] The chatbot on the server generates an appropriate response based on the received message.

[1538] For example, if a user asks, "What should I cook today?" the server generates a text response saying, "How about making curry?"

[1539] Step 11:

[1540] The server converts the generated text response into voice using a speech synthesis model.

[1541] This causes the text response to be output as a voice response.

[1542] Step 12:

[1543] The terminal displays and plays back the text and audio responses received from the server.

[1544] The user can then confirm the response and continue the dialogue.

[1545] Example 1

[1546] 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."

[1547] In modern society, after a specific individual passes away, there is a need to build an interactive dialogue system based on that individual's hobbies, preferences, and conversational style. However, there is still a lack of technology to efficiently collect available data, accurately analyze it, and conduct natural dialogue. The purpose of this invention is to solve this problem and provide a system that generates a replica based on the hobbies, preferences, and conversational style of a specific individual, enabling dialogue with the user.

[1548] 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.

[1549] In this invention, the server includes: means for a user to input text data and voice data using a terminal; means for transmitting the input text data and voice data to the server via a secure connection; means for the server to store the received text data and voice data in a database; means for the server to analyze the text data using morphological analysis and dependency structure analysis to extract language patterns and frequent topics; means for the server to convert the voice data into text using a voice recognition engine and model voice characteristics through spectral analysis; means for the server to generate a chatbot and a voice synthesis model that imitate the conversation style and language patterns of a specific individual based on the analysis results of the text data and voice data; and means for engaging in a dialogue with the user using the generated chatbot and voice synthesis model. This makes it possible to faithfully reproduce the hobbies, preferences, and conversation style of a specific individual and realize natural, interactive dialogue.

[1550] A "terminal" is an electronic device used by a user, such as a smartphone or a personal computer.

[1551] "Text data" refers to data of sentences or character strings that a user inputs using a terminal.

[1552] "Audio data" refers to data of audio recorded by a user using a terminal.

[1553] A "secure connection" is an encrypted communication method that protects data from being intercepted or tampered with by third parties during transmission.

[1554] A "server" is a computer system connected to the Internet that receives, analyzes, and stores text and audio data.

[1555] A "database" is a data structure for efficiently organizing and managing data within a server.

[1556] "Morphological analysis" is a natural language processing technique that divides text data into words and identifies their parts of speech in order to understand their meaning.

[1557] "Dependency structure analysis" is a natural language processing technique that analyzes the relationships between words in text data and reveals the structure of sentences.

[1558] A "speech recognition engine" is a software or hardware system for converting voice data into text data.

[1559] "Spectral analysis" is a technique for analyzing the frequency components and intonation of voice data.

[1560] "Voice characteristics" refer to the unique characteristics of individual voices, including, for example, tone, pitch, and intonation.

[1561] A "chatbot" is an automated response system that uses text data to interact with users.

[1562] A "voice synthesis model" is a model for converting text data into voice data based on analyzed voice characteristics.

[1563] A "profile" is information generated based on collected data that summarizes a specific individual's hobbies, preferences, and conversation style.

[1564] The present invention is a system that collects data on the hobbies, preferences, and conversational style of a specific individual, analyzes the data to create a profile, and generates a replica based on this profile. The system analyzes text data and voice data entered by a user using a terminal, and generates a chatbot and a voice synthesis model that mimics the conversational style and language patterns of the specific individual. The following describes how to specifically implement the present invention.

[1565] Data Collection and Transmission

[1566] Users use devices such as smartphones and PCs to input text data in various formats, such as diaries, text messages, emails, and social media posts. For example, a user might input "It was a nice day today" into a dedicated smartphone app. The user might also use an application on the device to record their own voice. This voice data contains the target individual's characteristic phrasing and tone of voice. For example, a user might record and save the voice saying "I slept well today."

[1567] The device sends the collected text data and voice data to the server via a secure connection (such as SSL / TLS). For example, the device sends text data such as "The weather was nice today" and voice data such as "I slept well today" to the server. After sending, the device notifies the user that the data has been sent successfully. For example, the device displays a message on the screen saying "The data has been sent successfully."

[1568] Data Storage and Analysis

[1569] The server stores the received text data and voice data in a database. This database is a data structure for organizing and managing data for each user. For example, the server stores the text data "The weather was nice today" and the voice file "I slept well today" in the "user123" folder.

[1570] The server then uses morphological analysis and dependency structure analysis to analyze the text data for language patterns, frequently occurring topics, and specific response trends. The software used includes the Python morphological analysis library "MeCab" and the dependency structure analysis engine "spaCy." For example, the server analyzes "The weather was nice today" and recognizes that "weather" is a frequently occurring topic and that it is a pleasant day. The analysis results are saved in a database as a user profile. For example, the server can add information to the profile such as "I have many good experiences with the weather."

[1571] The server also uses a speech recognition engine to convert the voice data into text. For this, it uses "Google Cloud Speech-to-Text." For example, it converts the voice data "I slept well today" into text. It then performs spectral analysis to extract the frequency components and intonation of the voice. For this, it uses "Praat." For example, the server analyzes and models the intonation and pitch of the voice "I slept well today."

[1572] Replication and User Interaction

[1573] The server generates a chatbot that mimics the target individual's conversation style and language patterns based on the results of analyzing the text data. The generative AI model used is "OpenAI GPT-4." For example, based on information that the server has "many positive experiences related to the weather," it generates a chatbot that prefers topics related to sunny days.

[1574] The server then trains a speech synthesis engine based on the results of the voice data analysis to build a speech synthesis model that can reproduce the voice characteristics of a specific individual. Engines used include Google Cloud Text-to-Speech and Amazon Polly. For example, the server builds a voice model that can reproduce the tone and tempo of the phrase "I slept well tonight."

[1575] Surviving family and friends can start a conversation with the created replica through a dedicated application. For example, a family member can type, "What should we cook today?" The device sends this input message to the server. The chatbot on the server generates an appropriate response based on the received message. For example, it might generate a response in text and voice, such as, "How about cooking curry today?" The device displays and plays back the text and voice response received from the server. The user can confirm this and carry out the next conversation.

[1576] For example, if a particular individual frequently talked about their favorite movies in the past, when a user asks "What movie did you see recently?" based on the conversation log, the chatbot on the server can generate a response such as "You used to like action movies, so check out the new action movies!" In this way, surviving family and friends can receive emotional support by interacting with a replica that faithfully reproduces the hobbies, preferences, and conversation style of the particular individual.

[1577] Example prompt sentence:

[1578] "What activities have you been interested in lately?"

[1579] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1580] Step 1:

[1581] A user inputs text data and voice data using a device. Using an application on a smartphone or computer, the user inputs text data in the form of a diary entry, text message, email, or social media post. The user also uses an application to record their own voice. For example, a user inputs text data such as "It was nice weather today" into a dedicated smartphone application and records voice such as "I slept well today." The input data is text data and voice data.

[1582] Step 2:

[1583] The device sends the collected text and voice data to the server through a secure connection. Specifically, the data is sent using encrypted communication such as SSL / TLS. When the transmission is complete, the device displays a notification to the user stating, "The data was sent successfully." The input data is the text and voice data entered by the user, and the output data is a notification indicating that the data has been sent.

[1584] Step 3:

[1585] The server stores the received text data and voice data in a database. Specifically, the server saves the text data "It was nice weather today" and the voice file "I slept well today" in the "user123" folder. The input data is the transmitted text data and voice data, and the output data is the data stored in the database.

[1586] Step 4:

[1587] The server analyzes the text data using morphological analysis and dependency structure analysis. Specifically, it uses the Python morphological analysis library "MeCab" and the dependency structure analysis engine "spaCy" to extract language patterns, frequently occurring topics, and specific reaction trends from the text data. For example, the server analyzes the text "It was nice weather today" and recognizes that "weather" is a frequently occurring topic. The input data is text data, and the output data is the analyzed language patterns and frequently occurring topics.

[1588] Step 5:

[1589] The server converts the voice data into text using a speech recognition engine. Specifically, it uses "Google Cloud Speech-to-Text" to convert the voice data "I slept well today" into text. It also performs spectral analysis to extract the frequency components and intonation of the voice. For example, using "Praat," the server analyzes and models the intonation and pitch of the voice "I slept well today." The input data is the voice data, and the output data is the analyzed voice characteristics and converted text data.

[1590] Step 6:

[1591] Based on the analysis results of the text and voice data, the server generates a chatbot and voice synthesis model that mimics the conversation style and language patterns of a specific individual. Specifically, it uses the generative AI model "OpenAI GPT-4" to generate a chatbot based on the text data, and builds a voice model using the voice synthesis engines "Google Cloud Text-to-Speech" and "Amazon Polly." The input data is the analysis results, and the output data is the chatbot and voice synthesis model.

[1592] Step 7:

[1593] The user initiates a conversation with the generated replica through a terminal. Specifically, the user inputs a message into a dedicated application, such as "What should I cook today?" The input message is sent to the server, and the chatbot on the server generates a response in text and voice, such as "How about making curry?" The input data is the message from the user, and the output data is the text and voice response from the chatbot.

[1594] Step 8:

[1595] The terminal displays and plays back the text and voice response received from the server. Specifically, it displays the chatbot's response text on the terminal's display and plays back the response aloud. For example, the terminal displays the text "How about making curry?" on the screen and plays back the voice. This allows the user to carry out the next dialogue. The input data is the text and voice response from the server, and the output data is the displayed text and played back voice.

[1596] An example of a prompt sentence: "What activities have you been interested in lately?"

[1597] (Application example 1)

[1598] 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."

[1599] In today's brick-and-mortar stores, it is difficult to provide personalized product recommendations and assistance to each individual consumer. Conventional services often provide general information and uniform customer service to all customers, failing to address individual preferences and needs. This can result in low user satisfaction and a loss of purchasing motivation. Furthermore, there is a lack of technology to effectively utilize profiles that utilize a specific individual's conversation style and preferences using a visual display device. Therefore, there is a need for the development of a system that can provide personalized product recommendations and assistance based on a specific individual's profile through a visual display device.

[1600] 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.

[1601] In this invention, the server includes means for collecting data on the hobbies, preferences, and conversation style of a specific individual, means for analyzing the collected data to create a profile, means for generating a replica of the specific individual based on the profile, means for interacting with the user using the generated replica, and means for providing customized product suggestions and assistance based on the analyzed profile using a visual display device attached to the user's communication device. This makes it possible to effectively provide personalized product suggestions and assistance based on the profile of a specific individual in a physical store.

[1602] "Hobbies" are preferences regarding the things and activities that a particular individual likes.

[1603] A "conversational style" is the linguistic expression or method of communication that a particular individual normally uses.

[1604] A "data collection instrument" is a device or method for collecting information about the tastes, preferences, and speaking styles of a particular individual.

[1605] A "profile" is a collection of information that describes the characteristics of a particular individual, created based on collected data.

[1606] A "replica" is an interactive virtual assistant generated based on a specific individual's profile.

[1607] The "interaction means" is a method or device for communicating with a user using the generated replica.

[1608] A "visual display device" is a communication device worn by a user that visually displays text and images.

[1609] "Product Recommendation" is a method or device aimed at recommending personalized products and services based on a user's profile.

[1610] "Assistance" refers to help or support provided based on a specific individual's preferences and conversational style.

[1611] This invention is a "personalized shopping assistant" system that provides personalized product recommendations and assistance in brick-and-mortar stores. It uses a visual display device attached to a user's communication device. Specifically, it is implemented using smart glasses. The system of this invention suggests products and services based on the user's tastes, preferences, and conversation style, providing an experience tailored to individual needs.

[1612] System Configuration and Operation

[1613] 1. Hardware and Software:

[1614] Hardware: Smart glasses (with voice recognition microphone and speaker), server, microphone

[1615] Software: GPT-2, speech recognition library (speech_recognition), speech synthesis engine (pyttsx3), Python

[1616] The smart glasses recognize the user's voice and send the data to the server, which analyzes the data and generates a response based on the user's profile, which is presented visually and audibly to the smart glasses.

[1617] 2. Data collection and analysis:

[1618] The user puts on the smart glasses and speaks their questions or requests. The speech recognition library, speech_recognition, converts the speech into text data.

[1619] The voice and text data sent to the server is used to generate appropriate responses based on the profile using GPT-2.

[1620] 3. Create a profile:

[1621] The server collects data about the user's interests and conversational style and creates a profile based on that data, including the types of products the user likes and conversational habits.

[1622] Through a speech recognition engine and speech synthesis engine, the user's language patterns and voice characteristics are analyzed and reflected in the profile.

[1623] 4. Generate and present the response:

[1624] When a user speaks to the smart glasses, the server creates a prompt based on the content and generates a response using the GPT-2 model.

[1625] The generated text response is converted to speech using the speech synthesis engine pyttsx3 and played through the smart glasses' speakers, and also displayed as text on the visual display.

[1626] Examples:

[1627] For example, if a user speaks to the smart glasses, "What kind of food would you like today?", the microphone in the smart glasses will capture the voice, and the speech recognition library, speech_recognition, will convert it into text. This text data will be sent to the server and processed as the following prompt sentence:

[1628] User: What would you like to eat today?

[1629] Assistant: The weather is nice today, why not enjoy a barbecue?

[1630] The responses are converted to speech and played back to the user, and are also displayed as text on a visual display, allowing for personalized product recommendations and assistance based on the user's preferences and conversational style.

[1631] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1632] Step 1:

[1633] User: Voice Input

[1634] Users wear smart glasses and input questions or requests by voice.

[1635] Input: User's voice data

[1636] Output: Audio data is captured

[1637] Specific operation: A microphone installed in the smart glasses collects the user's voice.

[1638] Step 2:

[1639] Device: Voice recognition

[1640] The device uses the speech recognition library speech_recognition to convert the voice data into text data.

[1641] Input: Captured audio data

[1642] Output: Text data

[1643] Specific operation: Audio data collected from the microphone is sent to the speech recognition library, and corresponding text data is generated.

[1644] Step 3:

[1645] Terminal: Data transmission

[1646] The terminal transmits the converted text data to the server over a secure connection.

[1647] Input: Text data

[1648] Output: Text data sent to the server

[1649] Specific operation: The converted text data is encrypted and sent securely to the server.

[1650] Step 4:

[1651] Server:Profile Creation

[1652] The server analyzes the received text data and creates a profile of the user's interests, preferences and conversation style.

[1653] Input: Text data

[1654] Output: User profile

[1655] Specific operation: Through morphological analysis and dependency structure analysis, language patterns and preferences are extracted from text data and a profile is generated.

[1656] Step 5:

[1657] Server: Response Generation

[1658] The server uses the generative AI model GPT-2 to create appropriate prompts and generate responses based on the user's profile.

[1659] Input: Profile and text data

[1660] Output: Text response

[1661] How it works: A prompt sentence is input into the GPT-2 model, which generates a text response based on the user's interests and preferences.

[1662] Example prompt sentence:

[1663] User: What would you like to eat today?

[1664] Assistant: The weather is nice today, why not enjoy a barbecue?

[1665] Step 6:

[1666] Server: Speech synthesis

[1667] The server converts the text response into speech using the speech synthesis engine pyttsx3.

[1668] Input: Text response

[1669] Output: Audio data

[1670] What it does: The generated text response is fed into a speech synthesis engine to generate a voice that sounds similar to a human voice.

[1671] Step 7:

[1672] Terminal: Response presentation

[1673] The terminal plays the generated audio data through the speaker of the smart glasses and displays the text data on a visual display.

[1674] Input: Audio and text data

[1675] Output: Presents the response to the user

[1676] Specific Actions: Provide a response to the user by playing audio data over a speaker and displaying text data on a visual display device.

[1677] 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.

[1678] The present invention is a system that collects data on a specific individual's hobbies, preferences, and conversation style, analyzes the data, and creates a profile. It then generates a replica based on the profile, and combines this with an emotion engine that recognizes the user's emotions. The generated replica can converse with the user and adjusts its responses according to the user's emotional state. Specific embodiments are described below.

[1679] Data collection

[1680] 1. Terminal: Data entry

[1681] Users use devices such as smartphones and personal computers to input text data in various formats, such as diaries, text messages, emails, and posts on social media.

[1682] The user uses an application on the device to record their own voice, and this voice data contains the target individual's characteristic phrasing and tone of voice.

[1683] 2. Terminal: Data transmission

[1684] The device transmits the collected text and voice data to the server over a secure connection, after which the device notifies the user that the data was successfully transmitted.

[1685] Data analysis

[1686] 1. Server: Data storage

[1687] The server stores the received text data and voice data in a database.

[1688] Within the database, data is organized and managed for each user.

[1689] 2. Server: Text data analysis

[1690] The server uses morphological and dependency structure analysis to analyze the text data for language patterns, frequent topics, and specific response tendencies, and the analysis results are stored in a database as user profiles.

[1691] 3. Server: Voice data analysis

[1692] The server uses a voice recognition engine to convert the voice data into text, and based on this, the voice data is also used for analysis.

[1693] The server performs spectral analysis to extract the frequency components and intonation of the voice, thereby modeling the voice characteristics.

[1694] Replication

[1695] 1. Server: Chatbot generation

[1696] Based on the results of analyzing the text data, the server generates a chatbot that has learned the target individual's language patterns and is capable of generating natural responses in everyday conversations and on specific topics.

[1697] 2. Server: Building a speech synthesis model

[1698] The server trains the speech synthesis engine based on the analysis of the voice data and builds a speech synthesis model that reproduces the voice characteristics of a specific individual, allowing it to convert text responses into voice responses.

[1699] Adding an Emotion Engine

[1700] 1. Server: User emotion recognition

[1701] The server uses an emotion engine to recognize emotions from the user's voice and text data, for example by analyzing voice characteristics such as tone, speed, and volume, and from text data by analyzing the context and emotional tendencies of selected words.

[1702] 2. Server: Emotion-based profile update

[1703] Based on the recognized emotions, the user profile is updated, which includes the user's emotional tendencies and current emotional state.

[1704] 3. Server: Emotion-based response generation

[1705] The generated response is tailored based on the emotional state recognized by the emotion engine. For example, if the user asks a question in a sad voice, the server will generate a more empathetic and comforting response.

[1706] User Interaction

[1707] 1. User: Initiate conversation

[1708] Family and friends can then interact with the replica through a dedicated application, which sends text and voice messages to the server.

[1709] 2. Device: Send Message

[1710] The terminal transmits the input text message and voice data to the server.

[1711] 3. Server: Response Generation

[1712] The server's chatbot generates an appropriate response based on the received message and emotional state. For example, if the user says "Today was tough" in a tired voice, the server generates a response like "Today was tough. Let's take a break."

[1713] 4. Server: Speech synthesis

[1714] The server converts the generated text response into voice using a speech synthesis model, which outputs the text response as a voice response.

[1715] 5. Terminal: Receives response

[1716] The terminal displays and plays back the text and voice responses received from the server, allowing the user to confirm the responses and continue the dialogue.

[1717] Specific examples

[1718] For example, if a specific individual has often talked about "movies" in the past, when the user asks "What movie did you see recently?", the chatbot on the server will generate a response such as "There was a new action movie recently that was good. You should definitely check it out" based on the user's preferences and emotional state. Also, by recognizing the user's emotional state, if the user is feeling down, for example, an encouraging response such as "Cheer up. Let's find a movie we can enjoy together" will be generated.

[1719] This system allows surviving family and friends to receive emotional support by interacting with a replica that faithfully reproduces the individual's hobbies, preferences, and conversation style. Furthermore, the emotional engine recognizes the user's emotional state and generates responses accordingly, enabling more empathetic and natural conversations.

[1720] The processing flow will be explained below.

[1721] Step 1:

[1722] Users can use the device's applications to enter information about their diary, conversation log, and hobbies and interests in text format, and also record voice messages using the device's recording function.

[1723] Step 2:

[1724] The device sends the collected text and voice data to the server over a secure connection, and the user is notified once the data has been sent.

[1725] Step 3:

[1726] The server stores the received text data in a database, and simultaneously stores the audio data in the database.

[1727] Step 4:

[1728] The server analyzes the received text data using morphological and dependency analysis to identify the user's language patterns and frequent topics and create a profile.

[1729] Step 5:

[1730] The server converts the voice data into text data using a speech recognition engine, and then performs spectral analysis to extract features for each frequency component and phoneme.

[1731] Step 6:

[1732] The server stores a user profile based on the analysis results in a database, which includes language patterns, interests, and voice characteristics.

[1733] Step 7:

[1734] The server generates a chatbot based on the profile, which is trained to replicate the language patterns of the specific individual and to engage in natural conversations.

[1735] Step 8:

[1736] The server uses the speech analysis results to train a speech synthesis engine and build a speech synthesis model that replicates the voice characteristics of a particular individual, which is then used to convert the text response into a speech response.

[1737] Step 9:

[1738] The user initiates a dialogue with the created replica through a dedicated application. At the beginning of the dialogue, the terminal sends the user's input to the server.

[1739] Step 10:

[1740] The server uses an emotion engine to recognize emotions from user input data, analyzing voice tone, speed, volume, etc. from voice data, and emotional trends from text data.

[1741] Step 11:

[1742] The server's chatbot generates more empathetic and appropriate responses based on the emotions recognized by the emotion engine. For example, if the user speaks in a sad voice, the server will generate a comforting response.

[1743] Step 12:

[1744] The server converts the generated text response into voice using a speech synthesis model, which outputs the text response as a voice response that reflects the appropriate emotion.

[1745] Step 13:

[1746] The device displays and plays the text and voice responses received from the server to the user, who can then confirm the responses and continue the dialogue.

[1747] Example 2

[1748] 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."

[1749] In modern society, emotional support gained through memories and conversations with specific individuals is highly valued. However, when a specific individual has passed away or is far away, means of sharing conversations and memories with that person are limited. Conventional technologies simply play back text or audio data, and do not realize natural conversations that reflect the user's emotions, hobbies, and preferences. Therefore, the present invention calls for a new system that faithfully reproduces the hobbies, preferences, and conversation style of a specific individual, enabling conversations that correspond to the user's emotional state.

[1750] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes: means for collecting data on the hobbies, preferences, and conversation style of a specific individual; means for encrypting the collected text data and voice data and transmitting it through a secure connection; means for analyzing the collected data to create a user profile; means for generating a chatbot and a voice synthesis model for the specific individual based on the user profile; means for recognizing the user's emotions using an emotion engine and updating the profile; and means for interacting with the user using the generated replica and adjusting responses based on the user's emotional state. This makes it possible to interactively interact with a replica that faithfully reproduces the characteristics of a specific individual and obtain responses that correspond to the user's emotional state.

[1751] 1. "Specified Individual" refers to a specific person about whom data is collected and analyzed by the system.

[1752] 2. "Preferences" refers to information about a particular individual's preferred activities, interests, and preferences.

[1753] 3. "Conversational style" refers to a particular individual's speaking style, phrasing, language patterns, and communication characteristics.

[1754] 4. "Data Collection Instrument" means the part of the system that captures text and voice data about a specific individual.

[1755] 5. "Data transmission means" refers to the function for encrypting collected data and transmitting it securely to the server.

[1756] 6. "Data analysis means" refers to the part of the system that analyzes collected data to extract and understand hobbies, preferences, and conversation styles.

[1757] 7. "User profile" refers to information that reproduces the characteristics of a specific individual, generated based on analyzed hobbies, preferences, and conversation style.

[1758] 8. "Chatbot generation means" refers to a function for generating a chatbot that reproduces the language patterns of a specific individual based on a user profile.

[1759] 9. "Voice synthesis model generation means" refers to a function for generating a voice synthesis model that reproduces the voice characteristics of a specific individual based on the results of voice data analysis.

[1760] 10. "Emotion engine" refers to software or algorithms that recognize and analyze user emotions.

[1761] 11. "Profile Updater" refers to functionality for updating a user profile based on the user's emotional state recognized by the Emotion Engine.

[1762] 12. "Response adjustment means" refers to a function that enables the generated replica to generate a response according to the emotional state during interaction with the user.

[1763] 13. "Replica" refers to a simulation that faithfully reproduces the tastes, preferences, and conversation style of a specific individual.

[1764] The present invention is a system that collects data on a specific individual's hobbies, preferences, and conversation style, analyzes the data, and creates a profile. A replica is generated based on this profile, and by combining it with an emotion engine that recognizes the user's emotions, a system is provided that can interact with the user. Specific embodiments are described below.

[1765] Data collection

[1766] 1. Terminal: Data entry

[1767] Users use devices such as smartphones and PCs to input various forms of text data, such as diaries, text messages, emails, and posts on social media. Users also use applications on their devices to record their own voice. This voice data includes the target individual's characteristic phrasing and tone of voice.

[1768] 2. Terminal: Data transmission

[1769] The device encrypts the collected text and voice data and sends it to the server over a secure connection, after which the device notifies the user that the data has been successfully sent.

[1770] Data analysis

[1771] 1. Server: Data storage

[1772] The server stores the received text and voice data in a database, where data is organized and managed for each user.

[1773] 2. Server: Text data analysis

[1774] The server uses morphological and dependency structure analysis to analyze the text data for language patterns, frequent topics, and specific response tendencies, and the analysis results are stored in a database as user profiles.

[1775] 3. Server: Voice data analysis

[1776] The server converts the voice data into text using a speech recognition engine. Based on this, the voice data is also analyzed. The server also performs spectral analysis to extract the frequency components and intonation of the voice and model the voice characteristics.

[1777] Replication

[1778] 1. Server: Chatbot generation

[1779] Based on the results of analyzing the text data, the server generates a chatbot that has learned the target individual's language patterns and is capable of generating natural responses in everyday conversations and on specific topics.

[1780] 2. Server: Building a speech synthesis model

[1781] The server trains the speech synthesis engine based on the analysis of the voice data and builds a speech synthesis model that reproduces the voice characteristics of a specific individual, allowing it to convert text responses into voice responses.

[1782] Adding an Emotion Engine

[1783] 1. Server: User emotion recognition

[1784] The server uses an emotion engine to recognize emotions from the user's voice and text data, for example by analyzing voice characteristics such as tone, speed, and volume, and from text data by analyzing the context and emotional tendencies of selected words.

[1785] 2. Server: Emotion-based profile update

[1786] Based on the recognized emotions, the user profile is updated, which includes the user's emotional tendencies and current emotional state.

[1787] 3. Server: Emotion-based response generation

[1788] The generated response is tailored based on the perceived emotional state: for example, if the user asks a question in a sad voice, the server will generate a more empathetic and comforting response.

[1789] User Interaction

[1790] 1. User: Initiate conversation

[1791] Users can start interacting with the replicas through a dedicated application, and when a user initiates an interaction, the device sends text and voice messages to the server.

[1792] 2. Device: Send Message

[1793] The terminal transmits the input text message and voice data to the server.

[1794] 3. Server: Response Generation

[1795] The server's chatbot generates an appropriate response based on the received message and emotional state. For example, if the user says "Today was tough" in a tired voice, the server generates a response like "Today was tough. Let's take a break."

[1796] 4. Server: Speech synthesis

[1797] The server converts the generated text response into voice using a speech synthesis model, which outputs the text response as a voice response.

[1798] 5. Terminal: Receives response

[1799] The terminal displays and plays back the text and voice responses received from the server, allowing the user to confirm the responses and continue the dialogue.

[1800] Specific examples and prompts for the generative AI model

[1801] Specific examples

[1802] For example, if a specific individual has often talked about "movies" in the past, when the user asks "What movie did you see recently?", the chatbot on the server will generate a response such as "There was a new action movie recently that was good. You should definitely check it out" based on the user's preferences and emotional state. Also, by recognizing the user's emotional state, if the user is feeling down, for example, an encouraging response such as "Cheer up. Let's find a movie we can enjoy together" will be generated.

[1803] Prompt Sentence Examples

[1804] 1. Prompt to analyze user text data

[1805] "Analyze text data from diaries, text messages, and social media posts using morphological and dependency analysis. Extract language patterns, frequently occurring topics, and specific reaction tendencies for each user to create a user profile."

[1806] 2. Prompt to request voice data analysis

[1807] "Use a speech recognition engine to convert the speech data into text, then perform spectral analysis to extract the frequency components and intonation of the speech. Based on the results, model the speech characteristics."

[1808] 3. Prompt to generate a chatbot

[1809] "Based on the results of analyzing the text data, create a chatbot that learns the user's language patterns. This chatbot should be able to generate natural responses in everyday conversations and on specific topics."

[1810] 4. Prompt to request emotion engine analysis

[1811] "Analyze voice data for characteristics such as tone, speed, and volume, and text data for context and emotional trends in chosen words. Based on the results, recognize the user's emotions and update the user profile."

[1812] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1813] System program processing flow

[1814] Step 1: Data entry

[1815] Users use smartphones or personal computers to input text data in various formats, such as diary entries, text messages, emails, and social media posts, into dedicated applications.

[1816] Input: Text data (e.g., diary entries, messages, posts)

[1817] Output: Collected text data

[1818] The user uses the same application to record their own voice, and the recorded voice data includes the user's characteristic phrasing and tone of voice.

[1819] Input: Audio data (e.g., recording of everyday conversation)

[1820] Output: Collected audio data

[1821] Step 2: Send data

[1822] The device encrypts the collected text and voice data and sends it to a server over a secure connection.

[1823] Input: Collected text and audio data

[1824] Output: Sending encrypted data to the server

[1825] After transmission, the terminal notifies the user that the data has been successfully transmitted.

[1826] Input: Send result

[1827] Output: User notification

[1828] Step 3: Data storage

[1829] The server stores the received text and voice data in a database, where data is organized and managed for each user.

[1830] Input: Received text and audio data

[1831] Output: Data stored in a database

[1832] Step 4: Text data analysis

[1833] The server performs morphological analysis and breaks down the text data into words.

[1834] Input: Text data

[1835] Output: A list of decomposed words

[1836] The server then uses dependency analysis to analyze grammatical and semantic relationships.

[1837] Input: A list of decomposed words

[1838] Output: Structural data showing grammatical and semantic relationships

[1839] Step 5: Audio data analysis

[1840] The server converts the voice data into text data using a voice recognition engine.

[1841] Input: Audio data

[1842] Output: Character data

[1843] In addition, the server performs spectral analysis to extract the frequency components and intonation of the voice.

[1844] Input: Audio data

[1845] Output: Frequency content and intonation data

[1846] Step 6: Create a user profile

[1847] The server creates a user profile based on the results of the analysis of the text and voice data.

[1848] Input: Analysis results (text data and audio data)

[1849] Output: User profile

[1850] Step 7: Chatbot generation

[1851] The server generates a chatbot that learns the user's language patterns based on the user profile.

[1852] Input: User Profile

[1853] Output: Chatbot

[1854] Step 8: Building a speech synthesis model

[1855] Based on the results of the voice data analysis, the server constructs a voice synthesis model that reproduces the voice characteristics of a specific individual.

[1856] Input: Audio data analysis results

[1857] Output: Speech synthesis model

[1858] Step 9: Emotion Recognition and Profile Update

[1859] The server uses an emotion engine to recognize the user's emotions and reflect them in the profile.

[1860] Input: Voice and text data

[1861] Output: Updated user profile

[1862] Step 10: Start a conversation

[1863] The user initiates a conversation with the created replica through a dedicated application. When the conversation begins, the terminal sends the user's text and voice messages to the server.

[1864] Input: User's message (text and voice)

[1865] Output: Send to server

[1866] Step 11: Send the message

[1867] The terminal transmits the input text message and voice data to the server.

[1868] Input: User message

[1869] Output: Data sent to the server

[1870] Step 12: Response Generation

[1871] The chatbot on the server generates an appropriate response based on the received message and emotional state.

[1872] Input: User message and emotion data

[1873] Output: Response message

[1874] Step 13: Speech synthesis

[1875] The server uses a speech synthesis model to convert the generated text response into a voice response.

[1876] Input: Text response

[1877] Output: Voice response

[1878] Step 14: Receive a response

[1879] The terminal displays and plays back the text and voice responses received from the server, allowing the user to confirm the responses and continue the dialogue.

[1880] Input: Response from the server (text and audio)

[1881] Output: Displayed and played response

[1882] (Application example 2)

[1883] 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."

[1884] Conventional food delivery services only suggest generic menus without considering the user's individual preferences or emotional state, resulting in a lack of personalized user experience. Even systems that reflect the user's preferences and conversational style lack the ability to generate appropriate responses based on the user's emotions, resulting in low user satisfaction.

[1885] 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.

[1886] In this invention, the server includes means for collecting data on the hobbies, preferences, and conversation style of a specific individual, means for analyzing the collected data to create a profile, means for generating a replica of the specific individual based on the profile, means for interacting with a user using the generated replica, means for analyzing the emotional state of the user, means for adjusting responses to the user based on the emotional state, means for recommending foods based on the individual's eating preferences and past ordering history, and means for using the responses in the food recommendation, thereby enabling a personalized food delivery experience that reflects the user's individual preferences and emotions.

[1887] "The hobbies and preferences of a particular individual" refers to the things that a particular individual likes or is interested in.

[1888] "Conversational style" refers to the language patterns and speaking characteristics that a particular individual uses in everyday interactions.

[1889] "Data collection means" refers to devices or methods for collecting information such as text, audio, or images.

[1890] "Means for analyzing data to create profiles" means processes or software that analyze collected data and generate profiles that reflect the characteristics of specific individuals.

[1891] "Means for generating a replica of a specific individual" refers to equipment or methods for creating a virtual entity (replica) that imitates a specific individual based on the results of analysis.

[1892] "Means for interacting with the user" refers to the interface and functions that allow the replica to communicate with the user.

[1893] "Means for analyzing emotional state" refers to analytical engines and algorithms for reading emotions from the user's voice or text.

[1894] "Means for adjusting responses" refers to techniques or systems for appropriately changing the content of responses generated based on the user's emotional state.

[1895] "Dietary preferences" refer to the types of foods and beverages that a particular individual prefers.

[1896] "Past Order History" refers to a record of food and beverage orders a user has placed.

[1897] "Food recommendation tools" refers to algorithms or systems that suggest appropriate foods based on a user's preferences and ordering history.

[1898] The present invention provides a system for enhancing user experience based on the tastes, preferences, and conversational styles of specific individuals. In particular, an example of application to a food delivery service will be described in detail.

[1899] System Overview

[1900] The system uses the following major hardware and software:

[1901] User's device (smartphone, tablet, etc.)

[1902] Server (performs data collection and analysis, profile generation, replica generation, and response generation)

[1903] Sentiment analysis engine (Hugging Face Transformers)

[1904] Response generation engine (OpenAI's GPT-3)

[1905] Data collection

[1906] Users can use the device to input various types of text data, such as diary entries, text messages, emails, and social media posts, as well as record their own voice. This data is then sent to a server via a secure connection and stored in a database.

[1907] Profile Generation

[1908] The server analyzes the collected text and speech data. It uses morphological and dependency structure analysis to extract language patterns and frequently occurring topics from the text data, and creates a user profile based on these analysis results. It also converts speech data into text using a speech recognition engine and performs spectral analysis to model speech characteristics.

[1909] Replication and Interaction

[1910] The server then generates a chatbot that has learned the target individual's language patterns based on the profile. This chatbot is capable of generating natural responses based on everyday conversations and specific topics. It also trains a speech synthesis engine based on the results of voice data analysis, building a speech synthesis model that reproduces the specific individual's voice characteristics.

[1911] Sentiment Analysis and Response Adjustment

[1912] The server uses an emotion analysis engine to recognize emotions from the user's voice and text data. Based on the recognized emotional state, the profile is updated and the generated response is adjusted. For example, if the user says "Today was tough" in a tired voice, the server will generate a compassionate response such as "Today was tough. Let's take a break."

[1913] Food recommendation function

[1914] Furthermore, the system recommends foods based on an individual's dietary preferences and past ordering history. It combines the user's emotional state and profile information to make optimal food suggestions. For example, if a user says, "I'm feeling low today," the server will recommend foods along with a response like, "Maybe ice cream would be good. Something sweet will definitely cheer you up."

[1915] Specific examples

[1916] When a user types "What's your recommendation today?" into a food concierge app, the system analyzes the words for emotional nuances and suggests the best food options. An example prompt could be, "Generate a response as if the user were talking to a close friend: What's your recommendation today?"

[1917] This enables a personalized food delivery experience that reflects the user's individual preferences and emotions.

[1918] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1919] Step 1:

[1920] A user uses a device to input text data in various formats, such as diary entries, text messages, emails, and social media posts, and records audio data. This data is sent from the device to a server via a secure connection. The input data is text data and audio data, and is sent via secure communication. The output is a notification to the server that the data has been successfully received.

[1921] Step 2:

[1922] The server stores the received text and voice data in a database. The server organizes and manages this data, saving it in a different folder for each user. The input at this stage is the text and voice data sent by the user, and the output is storage in the database.

[1923] Step 3:

[1924] The server performs morphological and dependency structure analysis on the collected text data. Specifically, it uses a natural language processing engine to analyze the text data and extract language patterns, frequently occurring topics, and specific response trends. The input is the text data stored in the database, and the output is a user profile based on the analysis results.

[1925] Step 4:

[1926] The server uses a speech recognition engine to convert the speech data into text and performs spectral analysis, which extracts the frequency components and intonation of the speech and models the speech characteristics. The input is the speech data stored in the database, and the output is the speech characteristic model.

[1927] Step 5:

[1928] The server generates a chatbot that learns the target individual's language patterns based on the profile information. The generated chatbot has the ability to generate natural responses to everyday conversations and specific topics. The input is text data and the results of its analysis, and the output is the chatbot.

[1929] Step 6:

[1930] The server trains the speech synthesis engine based on the voice characteristic model and constructs a voice synthesis model that reproduces the voice characteristics of a specific individual. The input is the voice characteristic model and the output is the voice synthesis model.

[1931] Step 7:

[1932] The server uses an emotion analysis engine to recognize emotions from the user's text and voice. It obtains emotion labels (e.g., "positive" or "negative") and adds them to the profile. The input is the user's text and voice data, and the output is the emotion labels.

[1933] Step 8:

[1934] The server processes the generated response based on the recognized emotional state, modifying the tone and content of the response based on the emotional label. The input is the emotional label and the chatbot's base response, and the output is the optimized response.

[1935] Step 9:

[1936] The server recommends optimal foods based on the user's dietary preferences and past ordering history, suggesting foods that match the user's immediate emotional state and long-term preferences. The input is user profile information and emotional state, and the output is a list of recommended foods.

[1937] Step 10:

[1938] The user initiates a conversation with the chatbot from their device and receives food recommendations. They then check the response and list of recommended foods received from the server and reflect these in their next conversation or order. The input is the response from the server and the list of recommended foods, and the output is increased user satisfaction.

[1939] 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.

[1940] 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.

[1941] 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.

[1942] 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.

[1943] 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.

[1944] 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.

[1945] 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).

[1946] 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.

[1947] 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."

[1948] 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.

[1949] 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).

[1950] 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.

[1951] 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.

[1952] 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 in response to a request from the data processing device 12.

[1953] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1954] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1955] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1956] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1957] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1958] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1959] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, ...

Claims

1. means for collecting data regarding the tastes, preferences, and conversation styles of particular individuals; a means for analyzing the collected data to create a profile; means for generating a replica of a particular individual based on the profile; a means for interacting with a user using the generated replica; A system including:

2. means for analyzing the collected voice data to extract voice characteristics; means for generating an audio response based on the extracted audio characteristics; The system of claim 1 .

3. means for analyzing the collected text data to extract language patterns; means for generating a text response based on the extracted language patterns; means for interacting with a user using generated voice and text responses; The system of claim 1 .

Images