system

Abstract

We provide the system. [Solution] A means of acquiring user voice data, A means for converting the above audio data into text data, A means of obtaining the user's past interaction data, A means for generating feature data by analyzing the above text data and the user's past interaction data, A generative AI means that generates personalized responses based on the above feature data, A means for providing the user with a personalized response generated by the above-mentioned generative AI means, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance as a response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Conventional voice assistant systems and personalization engines are limited in providing responses based on user preferences and interests, and more detailed personalization is required. Also, there are privacy concerns regarding the storage and use of personal information. Therefore, there is a need for a system that effectively utilizes the content of user utterances and past interaction data, manages feature data in a form where personal information cannot be identified, and provides personalized responses.

Means for Solving the Problems

[0005] The present invention provides a means for acquiring user voice data and means for converting voice data into text data. It also includes means for acquiring the user's past interaction data and means for analyzing this data to generate feature data. Furthermore, it provides a system that includes a generative AI means for generating personalized responses based on the generated feature data and means for providing the personalized responses generated by the generative AI means to the user. This system can manage the feature data in a way that does not identify individuals and can provide responses that reflect the user's preferences and interests.

[0006] "Voice data" refers to data that represents the voice information spoken by the user to the device.

[0007] "Text data" refers to data in written form obtained by converting audio data.

[0008] "Interaction data" refers to information such as data and operation history that a user has performed with the system or other related servers.

[0009] "Feature data" refers to data that reflects user preferences and interests, obtained by analyzing user text data and interaction data.

[0010] "Generative AI" refers to artificial intelligence technology that can generate new data based on input data.

[0011] "Personalized responses" refer to responses that provide answers and information optimized for the user, based on the user's characteristic data.

[0012] "Means" refers to methods or devices used to achieve a specific function or goal. [Brief explanation of the drawing]

[0013] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

MODE FOR CARRYING OUT THE INVENTION

[0014] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0015] First, the terms used in the following description will be explained.

[0016] In the following embodiments, a numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0017] In the following embodiments, a numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0018] In the following embodiments, a numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, and the like.

[0019] In the following embodiments, a numbered communication I / F (Interface) is an interface including a communication processor and 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), or Bluetooth (registered trademark), and the like.

[0020] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0021] [First Embodiment]

[0022] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0023] As shown in Figure 1, the 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.

[0024] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0025] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0026] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0027] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0028] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0029] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0031] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0032] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0033] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0034] The system for carrying out the present invention has the function of acquiring user voice data, analyzing it, and providing a personalized response. The system mainly includes a terminal, a server, and a generative AI.

[0035] Acquisition and conversion of audio data

[0036] The user speaks a question or command into their smartphone. For example, they might say, "What's the weather like today?" This voice data is captured by the device and sent to a server in real time. The server passes the received voice data to a natural language processing (NLP) module, which converts the voice data into text data. The converted text data will be in the form of "What's the weather like today?"

[0037] Acquisition of customer data

[0038] The server uses the user's identification information to retrieve the user's past interaction data from relevant customer databases (e.g., databases of telecommunications companies or related services). This data includes the applications the user has used and their past search history. For example, it might be confirmed that the user frequently uses a particular weather app.

[0039] Data analysis and feature data generation

[0040] The server analyzes text data and the user's past interaction data to extract user preferences and interests. This generates characteristic data about the user. For example, it might be identified that the user tends to go out on sunny days. This characteristic data is managed in a way that does not identify individuals.

[0041] Response generation by generative AI

[0042] The server trains a generative AI model based on the generated feature data. The generative AI uses the user's feature data to generate personalized responses. For example, it creates responses that reflect the user's interests, such as, "It's sunny today. It might get a little cloudy in the afternoon, but you can also check with your favorite weather app."

[0043] Providing a response

[0044] The device receives personalized responses generated by generative AI models and provides them to the user in voice or text format. For example, if a user asks about a specific cafe, the system will recommend cafes that suit the user's preferences and provide details about their locations.

[0045] Specific example

[0046] Situation

[0047] The following outlines the process when a user uses their smartphone to ask for the location of a cafe.

[0048] 1. User: "Can you recommend a nearby cafe?"

[0049] 2. The device captures the audio and sends the data to the server.

[0050] 3. The server converts the voice data into text. The result of the conversion is, "Please tell me about a nearby cafe."

[0051] 4. The server uses the identification information to retrieve the user's past interaction data from the relevant customer database. For example, it might be confirmed that the user prefers quiet cafes.

[0052] 5. The server analyzes the data and generates feature data based on user preferences.

[0053] 6. The server uses a generative AI model to generate personalized responses that reflect the user's characteristics.

[0054] 7. The terminal provides the user with a generated response. For example, it might respond, "There's a quiet cafe nearby that you might like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[0055] As described above, this invention can provide users with optimized information by personalizing responses based on the user's utterances and past data.

[0056] The following describes the processing flow.

[0057] Step 1:

[0058] The user speaks to their smartphone. For example, they might say, "What's the weather like today?"

[0059] Step 2:

[0060] The device captures the user's speech using its microphone and sends that audio data to the server in real time.

[0061] Step 3:

[0062] The server receives the audio data and passes it to a natural language processing (NLP) module, which converts the audio data into text data. The result of this conversion is "What's the weather like today?"

[0063] Step 4:

[0064] The server uses the user's identification information to retrieve the user's past interaction data from databases of software providers and integrated services. This data includes the applications the user uses and their past search history.

[0065] Step 5:

[0066] The server analyzes text data and acquired interaction data using machine learning algorithms to generate feature data based on user preferences and interests. For example, it might identify that a user prefers to use a particular weather app or tends to go out on sunny days.

[0067] Step 6:

[0068] The server trains a generative AI model based on the analyzed feature data. The generative AI is then adapted to generate personalized responses using the user's feature data.

[0069] Step 7:

[0070] The device generates personalized responses through a generative AI model. These responses reflect the user's preferences and past behavior. For example, they might say, "It's sunny today. It might get a little cloudy this afternoon."

[0071] Step 8:

[0072] The device provides the user with a personalized response via voice or text. The user can then review the response and decide on their next course of action. In this way, the system provides the user with optimized information.

[0073] (Example 1)

[0074] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0075] Conventional voice assistant systems have struggled to generate personalized responses by fully utilizing user preferences and past interaction data. Furthermore, in order to provide information best suited to each individual user, there has been insufficient management of user-specific data and adequate measures to protect personal information. This invention aims to solve these problems and provide users with appropriate and personalized information quickly.

[0076] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0077] In this invention, the server includes means for acquiring user voice data, means for converting voice data into text data, means for acquiring the user's past interaction data, means for analyzing and generating feature data, means for training a generative AI model, means for generating personalized responses, and means for providing responses to the user. This makes it possible to provide personalized responses quickly and accurately based on the user's preferences and interests.

[0078] A "user" is someone who uses the system to input voice data and receive a personalized response.

[0079] "Voice data" refers to the audio recorded when a user speaks into their smartphone or other device.

[0080] "Text data" refers to data obtained by converting audio data into text format.

[0081] "Interaction data" refers to a record of a user's past interactions with the system.

[0082] "Feature data" refers to data that indicates characteristics such as user preferences and interests.

[0083] A "generative AI model" is an artificial intelligence model designed to generate personalized responses based on user characteristic data.

[0084] "Means" refers to the devices or methods used to achieve an objective.

[0085] A "server" is a central processing unit that performs audio data processing, text conversion, and acquisition and analysis of interaction data.

[0086] A "terminal" is a device that a user uses to input voice data and receive a personalized response.

[0087] A "personalized response" is a customized reply generated based on the user's characteristic data.

[0088] The system for carrying out the present invention acquires user voice data, analyzes it, and provides a personalized response. The system components include a terminal, a server, and a generative AI model.

[0089] overview

[0090] The user speaks a question or command into a device such as a smartphone. For example, they might say, "What's the weather like today?" This voice data is captured by the device and sent to the server in real time. The server passes the received voice data to a natural language processing (NLP) module, which converts the voice data into text data. The converted text data will be in the form of "What's the weather like today?".

[0091] Acquisition and conversion of audio data

[0092] When a user speaks a question or command to their smartphone, the device uses its built-in microphone to capture voice data. This voice data is then transmitted to a server in real time via an internet connection. Specifically, the software used is an audio capture module, and the hardware used is the smartphone's built-in microphone.

[0093] Server-based processing

[0094] The server passes the received audio data to a natural language processing module, which uses text conversion software such as Google® Speech-to-Text API or IBM Watson® to convert the audio data into text data. This converted text data is then used for subsequent analysis.

[0095] User data collection and analysis

[0096] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. This data includes the applications the user has used and their past search history. For example, it might confirm that the user frequently uses a particular weather app.

[0097] The server analyzes text data and the user's past interaction data to extract user preferences and interests. This generates feature data about the user. Analysis tools such as Python's Pandas and Scikit-learn are used for the analysis. For example, if it is found that a user tends to go out on sunny days, this information is saved as feature data.

[0098] Response generation using generative AI models

[0099] The server trains a generative AI model based on the generated feature data. Examples of generative AI models used include GPT-3® and BERT. The generative AI model uses the user's feature data to generate personalized responses. For example, a possible response might be, "It's sunny today. It might get a little cloudy in the afternoon, but you can check this with your usual weather app."

[0100] Providing a response

[0101] Finally, the server sends the generated personalized response to the device. The device then provides this response to the user in either voice or text format. In the voice format, a Text-to-Speech (TTS) engine is used, and the response is delivered via the smartphone's notification system, for example. For instance, the user's screen might display a message such as, "It's sunny today. It might get a little cloudy in the afternoon, but you can also check your favorite weather app."

[0102] Examples of specific prompt messages

[0103] The following are examples of prompt messages in specific situations.

[0104] User: "Can you recommend a nearby cafe?"

[0105] Response: "There's a quiet cafe nearby that you might like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[0106] In this way, the present invention can provide personalized responses based on the user's voice input and analysis of past data.

[0107] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0108] Step 1:

[0109] The user speaks questions or commands into a device such as a smartphone. For example, they might say, "What's the weather like today?" The input is the user's voice data, and the output is the voice signal captured by the device's microphone. The device converts this voice into digital data and sends it to the server using a communication module.

[0110] Step 2:

[0111] The server receives audio data sent from the terminal. The input is digitized audio data, and the output is audio data passed to the NLP module. The server passes this data to the Natural Language Processing (NLP) module, which converts the audio data into text data. Software used includes Google Speech-to-Text API and IBM Watson.

[0112] Step 3:

[0113] The server parses the converted text data. The input is text data ("What's the weather like today?"), and the output is a query based on what the user asked. This query is stored for use in the next processing step.

[0114] Step 4:

[0115] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. The input is the user's identification information (user ID and registered email address), and the output is the user's past interaction data, which includes the applications the user used and their search history.

[0116] Step 5:

[0117] The server analyzes text data and the user's past interaction data. The input consists of text data and interaction data, while the output is feature data indicating the user's preferences and interests. This feature data includes information such as the user's tendency to go out on sunny days. Python libraries such as Pandas and Scikit-learn are used for the analysis.

[0118] Step 6:

[0119] The server trains a generative AI model based on the generated feature data. The input is the feature data, and the output is the trained generative AI model. Examples of generative AI models used include GPT-3 and BERT.

[0120] Step 7:

[0121] The server uses a generative AI model to generate personalized responses for the user. The input is a trained generative AI model and user feature data, and the output is a personalized response. For example, a response like, "It's sunny today. It might get a little cloudy in the afternoon, but you can also check with your favorite weather app," might be generated.

[0122] Step 8:

[0123] The server sends the generated personalized response to the terminal. The input is the personalized response, and the output is the response data sent to the terminal.

[0124] Step 9:

[0125] The device provides the user with the received response in either voice or text format. The input is the response data sent from the server, and the output is the personalized response presented to the user. In voice format, a Text-to-Speech (TTS) engine is used, and the response is delivered via the smartphone's notification system, for example. For instance, the user's screen might display a message such as, "It's sunny today. It might get a little cloudy in the afternoon, but you can also check with your favorite weather app."

[0126] In this way, the system analyzes the user's voice data and provides a response optimized for the user.

[0127] (Application Example 1)

[0128] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0129] Modern food delivery services are expected to rely heavily on voice commands for ordering and recommendations. However, current systems struggle to provide personalized recommendations based on users' past preferences and interests, and voice recognition for ordering also has accuracy issues. As a result, the user experience is not fully optimized.

[0130] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0131] In this invention, the server includes means for acquiring voice data, means for converting voice data into text data, means for acquiring the user's past interaction data, means for analyzing the text data and interaction data to generate feature data, means for generating personalized responses using generative AI, means for providing the generated responses, means for recommending restaurants and dishes, and means for placing food delivery orders using voice recognition. This enables the provision of restaurant and dish recommendations based on the user's preferences and interests, and further enables highly accurate food delivery orders using voice recognition.

[0132] "Audio data" refers to the voice information spoken by the user, and the captured acoustic signal.

[0133] "Text data" refers to information in string format generated by analyzing audio data.

[0134] "Interaction data" refers to information such as the actions, orders, and search history that a user has performed in the past.

[0135] "Feature data" refers to information about user preferences and interests, generated based on text data and interaction data.

[0136] "Generative AI" refers to artificial intelligence technology that receives feature data as input and generates personalized responses.

[0137] "Response" refers to the answers and recommendation information provided to the user by a generative AI.

[0138] A "server" refers to a computer system that is responsible for processing tasks such as analyzing audio data, acquiring interaction data, and executing generative AI.

[0139] "Restaurant and food recommendations" refers to a function that suggests appropriate restaurants and dishes based on the user's preferences and interests.

[0140] "Speech recognition" refers to the technology that analyzes speech data and converts it into text data.

[0141] "Food delivery" refers to a service that delivers food and beverages to the user's location.

[0142] This invention is a system that acquires user voice data, analyzes it, and provides personalized responses. This system mainly consists of a smartphone, a server, and a generative AI.

[0143] Acquisition of audio data

[0144] When a user speaks a question or command into their smartphone, voice data is captured. For example, they might say, "What restaurant would you recommend for dinner tonight?" This voice data is captured by the device and sent to the server in real time.

[0145] Audio data conversion

[0146] The server passes the received audio data to a natural language processing (NLP) module, which converts the audio data into text data. The converted text data will be in the form of "What restaurant would you recommend for dinner tonight?"

[0147] Acquisition of interaction data

[0148] The server uses the user's identification information to retrieve the user's past interaction data from relevant databases. This data includes the applications the user has used and their past order history. For example, it might confirm that the user has a preference for ordering Japanese food in the past.

[0149] Data analysis and feature data generation

[0150] The server analyzes text data and the user's past interaction data to extract user preferences and interests. This generates characteristic data about the user. For example, it might be identified that the user tends to prefer Japanese food. This characteristic data is managed in a way that does not identify individuals.

[0151] Response generation by generative AI

[0152] The server trains a generative AI model based on the generated feature data. The generative AI uses the user's feature data to generate personalized responses. For example, it might create a response that reflects the user's interests, such as, "There's a Japanese restaurant called XYZ nearby. It has good reviews, and is especially known for its delicious sashimi."

[0153] Providing a response

[0154] The device receives personalized responses generated by generative AI models and provides them to the user in voice or text format. For example, if a user asks, "What restaurants would you recommend for dinner tonight?", the system will recommend a Japanese restaurant based on the user's preferences and provide details about its location.

[0155] Specific example

[0156] The following outlines the process when a user uses their smartphone to request restaurant recommendations.

[0157] 1. User: "What restaurants would you recommend for dinner tonight?"

[0158] 2. The device captures the audio and sends the data to the server.

[0159] 3. The server converts the audio data into text. The result of the conversion will be, "What restaurant would you recommend for dinner tonight?"

[0160] 4. The server uses the identification information to retrieve the user's past interaction data from the relevant database. For example, it might confirm that the user prefers Japanese food.

[0161] 5. The server analyzes the data and generates feature data based on user preferences.

[0162] 6. The server uses a generative AI model to generate personalized responses that reflect the user's characteristics. For example, it might respond, "There's a Japanese restaurant called XYZ nearby. It has good reviews, especially for its sashimi."

[0163] 7. The terminal provides the generated response to the user.

[0164] Example of a prompt:

[0165] The user likes Japanese food and is looking for highly-rated restaurants. Text: Tell me about nearby Japanese restaurants. Please recommend restaurants that reflect these criteria.

[0166] In this way, the system can analyze the user's voice and recommend the most suitable restaurants and dishes based on the user's preferences. This improves the user's food delivery experience.

[0167] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0168] Step 1:

[0169] The user verbally asks a question into their smartphone. For example, they might ask, "What restaurants would you recommend for dinner tonight?" This audio data is captured by the device. The input data is an audio signal, and the output is audio data.

[0170] Step 2:

[0171] The device sends the captured audio data to the server. Since the transmitted audio data cannot be analyzed directly, speech recognition is required.

[0172] Step 3:

[0173] The server uses a speech recognition module to convert speech data into text data. This process is performed using natural language processing (NLP) techniques. The input data is speech data, and the output is text data such as "What restaurant would you recommend for dinner tonight?"

[0174] Step 4:

[0175] The server uses the user's identification information to retrieve the user's past interaction data from the relevant database. This data includes past order history and search history. The input data is the user's identification information, and the output is interaction data.

[0176] Step 5:

[0177] The server analyzes text data and the user's past interaction data to generate feature data. This process involves data analysis to extract the user's preferences and interests. The input data consists of text data and interaction data, and the output is feature data that indicates the user's preferences and interests.

[0178] Step 6:

[0179] The server trains a generative AI model using the generated feature data. The generative AI model is used to generate personalized responses based on the feature data. The input data is the feature data, and the output is a personalized response based on the generated prompt sentences.

[0180] Step 7:

[0181] The server sends the generated personalized response to the terminal. This response is then ready to be provided to the user. The input data is the generated response, and the output is the response data directed to the terminal.

[0182] Step 8:

[0183] The terminal provides the user with personalized responses received from the server. These responses are delivered in either voice or text format. Input data is the response data, and output is the information provided to the user. For example, the user might be notified with something like, "There's a Japanese restaurant called XYZ nearby. It has high ratings in online reviews, and is especially known for its delicious sashimi."

[0184] Example of a prompt:

[0185] The user likes Japanese food and is looking for highly-rated restaurants. Text: Tell me about nearby Japanese restaurants. Please recommend restaurants that reflect these criteria.

[0186] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0187] The system for carrying out the present invention has the function of acquiring user voice data, analyzing it, and providing a personalized response. The system mainly includes a terminal, a server, a generative AI, and an emotion engine.

[0188] Acquisition and conversion of audio data

[0189] The user speaks a question or command into their smartphone. For example, they might say, "What's the weather like today?" This voice data is captured by the device and sent to a server in real time. The server passes the received voice data to a natural language processing (NLP) module, which converts the voice data into text data. The converted text data will be in the form of "What's the weather like today?"

[0190] Analysis of emotional data

[0191] After receiving text data, the server passes it to the emotion engine. The emotion engine analyzes the user's emotions from the voice and text data and generates emotion data such as joy, surprise, anger, and sadness. For example, the statement "What's the weather like today?" might be analyzed as containing the user's anxiety or worry.

[0192] Acquisition of customer data

[0193] The server uses the user's identification information to retrieve the user's past interaction data from relevant customer databases (e.g., databases of telecommunications companies or related services). This data includes the applications the user has used and their past search history. For example, it might be confirmed that the user frequently uses a particular weather app.

[0194] Data analysis and feature data generation

[0195] The server uses machine learning algorithms to analyze text data, sentiment data, and the user's past interaction data to generate characteristic data based on the user's preferences and interests. For example, it might identify that the user tends to go out on sunny days or that their current speech contains anxiety. This characteristic data is managed in a way that does not identify individuals.

[0196] Response generation by generative AI

[0197] The server trains a generative AI model based on the generated feature and sentiment data. The generative AI uses the user's feature and sentiment data to generate personalized responses. For example, it might create a response that also takes the user's emotions into account, such as, "It's sunny today. It might get a little cloudy in the afternoon, but don't worry. You can also check with your favorite weather app."

[0198] Providing a response

[0199] The device receives personalized responses generated by generative AI models and provides them to the user in voice or text format. For example, if a user asks about cafe locations, the system recommends cafes that suit the user's preferences and current mood, and provides details about the locations. If the user is tired, it might recommend a quiet cafe.

[0200] Specific example

[0201] Situation

[0202] The following outlines the process when a user uses their smartphone to ask for the location of a cafe.

[0203] 1. User: "Can you recommend a nearby cafe?"

[0204] 2. The device captures the audio and sends the data to the server.

[0205] 3. The server converts the voice data into text. The result of the conversion is, "Please tell me about a nearby cafe."

[0206] 4. The server uses the identification information to retrieve the user's past interaction data from the relevant customer database. For example, it might be confirmed that the user prefers quiet cafes.

[0207] 5. The server uses an emotion engine to generate emotion data based on the user's utterances. For example, it might analyze that the user is seeking relaxation.

[0208] 6. The server analyzes the data to generate feature data and sentiment data based on user preferences.

[0209] 7. The server uses generative AI models to generate personalized responses that reflect the user's characteristics and emotions.

[0210] 8. The terminal provides the user with a generated response. For example, it might respond, "There is a quiet, relaxing cafe nearby that you might like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[0211] The present invention, through the process described above, provides information optimized for the user by personalizing responses based on the user's speech content, past data, and even emotional data.

[0212] The following describes the processing flow.

[0213] Step 1:

[0214] The user speaks to their smartphone. For example, they might say, "What's the weather like today?"

[0215] Step 2:

[0216] The device captures the user's speech using its microphone and sends that audio data to the server in real time.

[0217] Step 3:

[0218] The server receives the audio data and passes it to a natural language processing (NLP) module, which converts the audio data into text data. The result of this conversion is "What's the weather like today?"

[0219] Step 4:

[0220] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. This data includes the applications the user has used and their past search history. For example, it might be confirmed that the user frequently uses a particular weather app.

[0221] Step 5:

[0222] The server passes text data to the emotion engine, which analyzes the user's emotions. The emotion engine identifies whether the user is feeling emotions such as joy, surprise, anger, or anxiety based on the tone and content of the voice. For example, the statement "What's the weather like today?" might be analyzed as containing anxiety.

[0223] Step 6:

[0224] The server analyzes sentiment data, text data, and acquired interaction data using machine learning algorithms to generate feature data. This feature data includes things like whether a user tends to go out on sunny days, whether they prefer a particular weather app, and their current emotional state.

[0225] Step 7:

[0226] The server trains a generative AI model based on the generated feature and sentiment data. The generative AI model is configured to generate personalized responses that reflect the user's feature and sentiment data.

[0227] Step 8:

[0228] The device uses a generative AI model to generate personalized responses. These responses take into account the user's preferences and current mood. For example, it might say something like, "It's sunny today. It might get a little cloudy in the afternoon, but don't worry. You can also check your favorite weather app."

[0229] Step 9:

[0230] The device provides the user with a generated, personalized response via voice or text. The user can then review the response and decide on their next action. For example, if the user asks about cafe locations, the system recommends cafes that suit the user's preferences and current mood, and provides details about those locations.

[0231] (Example 2)

[0232] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0233] Conventional speech recognition systems only convert user speech into text and are unable to provide responses that take into account the user's emotions or past interactions. As a result, they tend to provide generic answers without offering truly personalized and useful information to the user. Furthermore, there is a lack of means to generate and securely manage feature data based on individual user preferences and emotions. This invention aims to solve these problems.

[0234] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0235] In this invention, the server includes means for acquiring user voice data, means for converting voice data into text data, means for analyzing user emotion data, means for acquiring user past interaction data, means for generating feature data by analyzing text data, emotion data and the user's past interaction data, generative AI means for generating personalized responses based on the feature data and emotion data, and means for providing the generated personalized responses to the user. This makes it possible to provide more appropriate and personalized responses that take into account the user's emotions and past interactions.

[0236] "User" refers to an individual or group that uses this system.

[0237] "Voice data" refers to data that records voice information spoken by a user in digital format.

[0238] "Text data" refers to string information obtained by converting audio data using natural language processing technology.

[0239] "Emotional data" refers to data that indicates the emotional state of a user, analyzed and extracted from their voice and text.

[0240] "Past interaction data" refers to information such as the user's past actions, usage history, and search history.

[0241] "Feature data" refers to characteristic information obtained by analyzing a user's preferences, interests, emotions, etc.

[0242] "Generative AI methods" refer to artificial intelligence models that generate optimal responses for users based on feature data and sentiment data.

[0243] "Encryption" refers to the process of transforming information to protect it from unauthorized access.

[0244] "Analysis" refers to the process of thoroughly examining data and extracting its meaning.

[0245] A "speech recognition algorithm" refers to an algorithm that converts speech data into text data.

[0246] A "natural language processing (NLP) module" refers to a software module that implements technology to analyze audio data and convert it into corresponding text data.

[0247] A "personalized response" refers to a response created with the individual user's characteristics and emotions in mind.

[0248] The system for carrying out the present invention has the function of acquiring user voice data, analyzing it, and providing a personalized response. This system mainly includes a user, a terminal, a server, a generative AI model, and an emotion engine.

[0249] First, the user speaks a question or command into a device such as a smartphone. For example, voice input such as "What's the weather like today?" or "Can you tell me about a nearby cafe?" is performed. This voice data is captured by the device and then sent to the server in real time. The hardware used by the device includes a microphone and a communication module, and the transmitted data is encrypted to ensure reliability.

[0250] The server passes the received audio data to a natural language processing (NLP) module, which converts the audio data into text data. This conversion process uses a speech recognition algorithm, so that, for example, an audio statement like "Please tell me about a nearby cafe" is converted into similar text data.

[0251] Next, the server sends the converted text data to the emotion engine for analysis. The emotion engine analyzes the user's emotions from the audio and text data and generates emotion data such as joy, surprise, anger, and sadness. For example, it extracts emotions such as tension and relaxation from the user's tone and way of speaking.

[0252] User interaction data is also a crucial element. Based on user identification information, the server retrieves data such as past operation history, usage history, and search history from relevant databases. This allows for a detailed understanding of what the user likes and what their tendencies are.

[0253] The server combines and analyzes the obtained text data, sentiment data, and past interaction data to generate feature data. This feature data includes user preferences and interests, and machine learning algorithms are used for analysis.

[0254] Based on the generated feature and sentiment data, the server uses a generative AI model to create personalized responses. This allows for the creation of appropriate responses tailored to the user's specific situation and emotions. For example, it could respond with something like, "There's a quiet, relaxing cafe nearby that you'll like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[0255] Ultimately, the generated personalized response is delivered to the user via the device. The response is communicated to the user in voice or text format, and relevant information is specifically presented.

[0256] Specific example

[0257] For example, consider the following scenario where a user asks their smartphone, "Please tell me about nearby cafes."

[0258] 1. The user asks, "Can you tell me about a nearby cafe?"

[0259] 2. The device captures the audio and sends the data to the server.

[0260] 3. The server converts the voice data into text. The result of the conversion is "Please tell me about a nearby cafe."

[0261] 4. The server uses an emotion engine to generate emotion data based on the user's utterances. For example, it might analyze that the user is seeking relaxation.

[0262] 5. The server retrieves the user's past interaction data from the relevant database. For example, it might confirm that the user prefers quiet cafes.

[0263] 6. The server analyzes text data, sentiment data, and past interaction data to generate feature data.

[0264] 7. The server uses a generative AI model to generate personalized responses that reflect the user's characteristics and emotions.

[0265] 8. The terminal provides the user with a generated response. It responds, "There is a quiet, relaxing cafe nearby that you might like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[0266] In this way, the system can personalize its responses based on the user's utterances, past data, and even emotional data, providing information optimized for the user.

[0267] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0268] Step 1: User voice input

[0269] The user speaks into their smartphone. For example, they might say, "Please tell me about a nearby cafe." The device uses its microphone to capture the audio.

[0270] Input: User's voice data

[0271] Output: Captured audio data

[0272] Step 2: Sending the audio data

[0273] The device transmits the captured audio data to the server in real time. During this process, the data is encrypted and securely transferred over the internet.

[0274] Input: Captured audio data

[0275] Output: Audio data sent to the server

[0276] Step 3: Convert speech to text

[0277] The server passes the received audio data to a natural language processing (NLP) module, which converts the audio data into text data. A speech recognition algorithm is used for this process. For example, the audio "Please tell me about a nearby cafe" is converted into text.

[0278] Input: Sent audio data

[0279] Output: Text data "Please tell me about nearby cafes."

[0280] Step 4: Analysis of emotional data

[0281] The server passes the converted text data to the emotion engine. The emotion engine analyzes the user's emotions from the audio and text data and generates emotion data. For example, it might analyze that the user is seeking relaxation.

[0282] Input: Transformed text data

[0283] Output: User's emotional data (e.g., relaxed)

[0284] Step 5: Acquisition of customer data

[0285] Based on the user's identification information, the server retrieves customer data such as past operation history and usage history from the relevant database. This includes the content frequently searched by the user and the history of applications used.

[0286] Input: User's identification information

[0287] Output: User's past interaction data

[0288] Step 6: Analysis of data and generation of feature data

[0289] The server performs analysis based on the obtained text data, emotional data, and past interaction data. Using machine learning algorithms, it generates feature data representing the user's preferences and interests. For example, a feature that the user prefers a quiet café is extracted.

[0290] Input: Text data, emotional data, past interaction data

[0291] Output: User's feature data

[0292] Step 7: Generation of responses by generative AI

[0293] The server operates the generative AI model based on the generated feature data and emotional data. The generative AI model generates personalized responses according to the user's characteristics. For example, a response such as "There is a quiet and relaxing café that you like near here. The most highly rated one is 'Café XYZ', and it is known for its delicious coffee" is generated.

[0294] Input: Feature data, sentiment data

[0295] Output: Personalized response

[0296] Step 8: Provide a response

[0297] The device receives personalized responses sent from the server and provides them to the user in voice or text format. Users can receive these responses through their smartphone screen or voice assistant.

[0298] Input: Personalized response

[0299] Output: Provides a response to the user.

[0300] (Application Example 2)

[0301] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0302] In today's shopping environment, personalized services that cater to the diverse needs and emotions of consumers are in demand. However, conventional systems struggle to effectively utilize user voice and emotional data, combining them with past interaction history to provide optimal responses. This makes it difficult to deliver a truly satisfying shopping experience. Furthermore, the lack of technology to generate personalized responses in real time is also a challenge.

[0303] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0304] In this invention, the server includes means for acquiring the user's voice data, means for converting the voice data into text data, and means for acquiring the user's past interaction data. As a result, in a shopping environment, it becomes possible to analyze questions and commands based on the user's voice in real time, and further combine past interaction data and sentiment data to provide a personalized response to the user. Also, by using generative AI means, it is possible to realize the function of an optimal shopping assistant that reflects the user's current sentiment and past purchase history.

[0305] The "user's voice data" is a voice signal including words and commands uttered by the user.

[0306] The "text data" is data obtained by converting voice data into characters and sentences.

[0307] The "user's past interaction data" is data related to operations and activities previously performed by the user, and includes, for example, past purchase history and logs of used applications.

[0308] The "feature data" is data indicating the user's preferences and sentiment, extracted by analyzing the text data and past interaction data.

[0309] The "sentiment data" is data indicating the sentiment state analyzed from the user's utterance content and voice tone, and includes joy, surprise, anger, sadness, anxiety, etc.

[0310] The "generative AI means" is means for generating a personalized response based on the user's feature data and sentiment data using machine learning algorithms and generative models.

[0311] The "personalized response" indicates an optimized individual answer or recommendation considering the user's individual preferences and sentiment.

[0312] "Means of delivery" refers to the means of communicating the generated personalized response to the user in the form of voice or text.

[0313] The system for carrying out the present invention has the function of acquiring user voice data, analyzing it, and providing a personalized response. The system mainly includes a terminal, a server, a generative AI, and an emotion engine. The configuration and processing of each are described in detail below.

[0314] Acquisition and conversion of audio data

[0315] The user speaks a question or command into their smartphone or smart glasses. For example, they might say, "Which coffee maker do you recommend nearby?" This voice data is captured by the device and sent to a server in real time. The server passes the received voice data to a natural language processing (NLP) module, which converts the voice data into text data. The converted text data will be in the form of "Which coffee maker do you recommend nearby?"

[0316] Analysis of emotional data

[0317] After the server receives the text data, it passes it to the emotion engine. The emotion engine analyzes the user's emotions from the voice and text data and generates emotion data such as joy, relief, surprise, and excitement. For example, if the user says, "What's a good coffee maker nearby?", the emotion engine might analyze that the user is excited.

[0318] Acquisition of customer data

[0319] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. This data includes items the user has previously purchased and services they have used. For example, a record of the user having previously purchased a coffee maker may be found.

[0320] Data analysis and feature data generation

[0321] The server uses machine learning algorithms to analyze text data, sentiment data, and the user's past interaction data to generate characteristic data based on the user's preferences and interests. For example, it might identify that the user has a history of liking espresso or that their current statement contains excitement. This characteristic data is managed in a way that does not identify individuals.

[0322] Response generation by generative AI

[0323] The server generates personalized responses using a generative AI model based on the generated feature and sentiment data. The generative AI uses the user's feature and sentiment data to create an optimal, individually tailored response. For example, it might generate a response that takes the user's emotions into account, such as, "You've always liked espresso, so we recommend the latest espresso maker A. It's on sale right now, so please take this opportunity to try it."

[0324] Providing a response

[0325] The device receives personalized responses generated by a generative AI model and provides them to the user in voice or text format. For example, it might respond, "Nearby, we have coffee maker A, which is perfect for espresso, and it's currently on sale."

[0326] Explanation of specific examples

[0327] For example, if a user asks, "Which coffee maker do you recommend near me?", the system converts the voice into text, performs sentiment analysis, and then, referencing past purchase history and preferences, generates a response such as, "Since you've always liked espresso, we recommend the latest espresso maker A. It's currently on sale, so please take this opportunity to try it."

[0328] Example of a prompt

[0329] User comment: "What coffee maker would you recommend near me?"

[0330] Emotion: {"neg": 0.0, "neu": 0.5, "pos": 0.5, "compound": 0.5}

[0331] Customer information: {"name": "User", "preferences": "Espresso", "history": ["Coffee beans", "Coffee machine"]}

[0332] response:"

[0333] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0334] Step 1:

[0335] Acquisition of user voice data

[0336] The user speaks questions or commands into their smartphone or smart glasses. For example, they might say, "Which coffee maker do you recommend nearby?" This voice data is captured by the device and sent to the server in real time.

[0337] Input: Audio data

[0338] Output: Audio data

[0339] Step 2:

[0340] Converting audio data to text

[0341] The server passes the received audio data to a natural language processing (NLP) module, which converts the audio data into text data. The converted text data will be in the form of "What is the best coffee maker nearby?"

[0342] Input: Audio data

[0343] Output: Text data

[0344] Step 3:

[0345] Analysis of emotional data

[0346] After the server receives the text data, it passes it to the emotion engine. The emotion engine analyzes the user's emotions from the voice and text data and generates emotion data such as joy, relief, surprise, and excitement. For example, if the user says, "What's a good coffee maker nearby?", the emotion engine might analyze that the user is excited.

[0347] Input: Audio data, text data

[0348] Output: Sentiment data

[0349] Step 4:

[0350] Acquisition of customer data

[0351] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. This data includes items the user has previously purchased and services they have used. For example, a record of the user having previously purchased a coffee maker may be found.

[0352] Input: User identification information

[0353] Output: Past interaction data

[0354] Step 5:

[0355] Data analysis and feature data generation

[0356] The server uses machine learning algorithms to analyze text data, sentiment data, and past user interaction data to generate characteristic data based on the user's preferences and interests. For example, it might reveal that the user tends to like espresso and identify if their current statement contains excitement. This characteristic data is managed in a way that does not identify individuals.

[0357] Input: Text data, sentiment data, historical interaction data

[0358] Output: Feature data

[0359] Step 6:

[0360] Response generation by generative AI

[0361] The server uses a generative AI model to generate personalized responses based on the generated feature data and sentiment data. The generative AI utilizes the user's feature data and sentiment data to create appropriate responses tailored to each individual, such as, "You've always liked espresso, so we recommend the latest espresso maker A. It's on sale right now, so please take this opportunity to try it."

[0362] Input: Feature data, sentiment data

[0363] Output: Personalized response

[0364] Step 7:

[0365] Providing a response

[0366] The device receives personalized responses generated by a generative AI model and provides them to the user in voice or text format. For example, it might respond, "Nearby, we have coffee maker A, which is perfect for espresso, and it's currently on sale."

[0367] Input: Personalized response

[0368] Output: Voice or text response

[0369] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0370] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0371] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0372] [Second Embodiment]

[0373] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0374] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0375] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0376] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0377] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0378] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0379] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0380] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0381] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0382] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0383] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0384] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0385] The system for carrying out the present invention has the function of acquiring user voice data, analyzing it, and providing a personalized response. The system mainly includes a terminal, a server, and a generative AI.

[0386] Acquisition and conversion of audio data

[0387] The user speaks a question or command into their smartphone. For example, they might say, "What's the weather like today?" This voice data is captured by the device and sent to a server in real time. The server passes the received voice data to a natural language processing (NLP) module, which converts the voice data into text data. The converted text data will be in the form of "What's the weather like today?"

[0388] Acquisition of customer data

[0389] The server uses the user's identification information to retrieve the user's past interaction data from relevant customer databases (e.g., databases of telecommunications companies or related services). This data includes the applications the user has used and their past search history. For example, it might be confirmed that the user frequently uses a particular weather app.

[0390] Data analysis and feature data generation

[0391] The server analyzes text data and the user's past interaction data to extract user preferences and interests. This generates characteristic data about the user. For example, it might be identified that the user tends to go out on sunny days. This characteristic data is managed in a way that does not identify individuals.

[0392] Response generation by generative AI

[0393] The server trains a generative AI model based on the generated feature data. The generative AI uses the user's feature data to generate personalized responses. For example, it creates responses that reflect the user's interests, such as, "It's sunny today. It might get a little cloudy in the afternoon, but you can also check with your favorite weather app."

[0394] Providing a response

[0395] The device receives personalized responses generated by generative AI models and provides them to the user in voice or text format. For example, if a user asks about a specific cafe, the system will recommend cafes that suit the user's preferences and provide details about their locations.

[0396] Specific example

[0397] Situation

[0398] The following outlines the process when a user uses their smartphone to ask for the location of a cafe.

[0399] 1. User: "Can you recommend a nearby cafe?"

[0400] 2. The device captures the audio and sends the data to the server.

[0401] 3. The server converts the voice data into text. The result of the conversion is, "Please tell me about a nearby cafe."

[0402] 4. The server uses the identification information to retrieve the user's past interaction data from the relevant customer database. For example, it might be confirmed that the user prefers quiet cafes.

[0403] 5. The server analyzes the data and generates feature data based on user preferences.

[0404] 6. The server uses a generative AI model to generate personalized responses that reflect the user's characteristics.

[0405] 7. The terminal provides the user with a generated response. For example, it might respond, "There's a quiet cafe nearby that you might like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[0406] As described above, this invention can provide users with optimized information by personalizing responses based on the user's utterances and past data.

[0407] The following describes the processing flow.

[0408] Step 1:

[0409] The user speaks to their smartphone. For example, they might say, "What's the weather like today?"

[0410] Step 2:

[0411] The device captures the user's speech using its microphone and sends that audio data to the server in real time.

[0412] Step 3:

[0413] The server receives the audio data and passes it to a natural language processing (NLP) module, which converts the audio data into text data. The result of this conversion is "What's the weather like today?"

[0414] Step 4:

[0415] The server uses the user's identification information to retrieve the user's past interaction data from databases of software providers and integrated services. This data includes the applications the user uses and their past search history.

[0416] Step 5:

[0417] The server analyzes text data and acquired interaction data using machine learning algorithms to generate feature data based on user preferences and interests. For example, it might identify that a user prefers to use a particular weather app or tends to go out on sunny days.

[0418] Step 6:

[0419] The server trains a generative AI model based on the analyzed feature data. The generative AI is then adapted to generate personalized responses using the user's feature data.

[0420] Step 7:

[0421] The device generates personalized responses through a generative AI model. These responses reflect the user's preferences and past behavior. For example, they might say, "It's sunny today. It might get a little cloudy this afternoon."

[0422] Step 8:

[0423] The device provides the user with a personalized response via voice or text. The user can then review the response and decide on their next course of action. In this way, the system provides the user with optimized information.

[0424] (Example 1)

[0425] Next, we will describe Example 1. 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."

[0426] Conventional voice assistant systems have struggled to generate personalized responses by fully utilizing user preferences and past interaction data. Furthermore, in order to provide information best suited to each individual user, there has been insufficient management of user-specific data and adequate measures to protect personal information. This invention aims to solve these problems and provide users with appropriate and personalized information quickly.

[0427] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0428] In this invention, the server includes means for acquiring user voice data, means for converting voice data into text data, means for acquiring the user's past interaction data, means for analyzing and generating feature data, means for training a generative AI model, means for generating personalized responses, and means for providing responses to the user. This makes it possible to provide personalized responses quickly and accurately based on the user's preferences and interests.

[0429] A "user" is someone who uses the system to input voice data and receive a personalized response.

[0430] "Voice data" refers to the audio recorded when a user speaks into their smartphone or other device.

[0431] "Text data" refers to data obtained by converting audio data into text format.

[0432] "Interaction data" refers to a record of a user's past interactions with the system.

[0433] "Feature data" refers to data that indicates characteristics such as user preferences and interests.

[0434] A "generative AI model" is an artificial intelligence model designed to generate personalized responses based on user characteristic data.

[0435] "Means" refers to the devices or methods used to achieve an objective.

[0436] A "server" is a central processing unit that performs audio data processing, text conversion, and acquisition and analysis of interaction data.

[0437] A "terminal" is a device that a user uses to input voice data and receive a personalized response.

[0438] A "personalized response" is a customized reply generated based on the user's characteristic data.

[0439] The system for carrying out the present invention acquires user voice data, analyzes it, and provides a personalized response. The system components include a terminal, a server, and a generative AI model.

[0440] overview

[0441] The user speaks a question or command into a device such as a smartphone. For example, they might say, "What's the weather like today?" This voice data is captured by the device and sent to the server in real time. The server passes the received voice data to a natural language processing (NLP) module, which converts the voice data into text data. The converted text data will be in the form of "What's the weather like today?".

[0442] Acquisition and conversion of audio data

[0443] When a user speaks a question or command to their smartphone, the device uses its built-in microphone to capture voice data. This voice data is then transmitted to a server in real time via an internet connection. Specifically, the software used is an audio capture module, and the hardware used is the smartphone's built-in microphone.

[0444] Server-based processing

[0445] The server passes the received audio data to a natural language processing module, which uses text conversion software such as the Google Speech-to-Text API or IBM Watson to convert the audio data into text. This converted text data is then used for subsequent analysis.

[0446] User data collection and analysis

[0447] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. This data includes the applications the user has used and their past search history. For example, it might confirm that the user frequently uses a particular weather app.

[0448] The server analyzes text data and the user's past interaction data to extract user preferences and interests. This generates feature data about the user. Analysis tools such as Python's Pandas and Scikit-learn are used for the analysis. For example, if it is found that a user tends to go out on sunny days, this information is saved as feature data.

[0449] Response generation using generative AI models

[0450] The server trains a generative AI model based on the generated feature data. Examples of generative AI models used include GPT-3 and BERT. The generative AI model uses the user's feature data to generate personalized responses. For example, a possible response might be, "It's sunny today. It might get a little cloudy this afternoon, but you can check this with your usual weather app."

[0451] Providing a response

[0452] Finally, the server sends the generated personalized response to the device. The device then provides this response to the user in either voice or text format. In the voice format, a Text-to-Speech (TTS) engine is used, and the response is delivered via the smartphone's notification system, for example. For instance, the user's screen might display a message such as, "It's sunny today. It might get a little cloudy in the afternoon, but you can also check your favorite weather app."

[0453] Examples of specific prompt messages

[0454] The following are examples of prompt messages in specific situations.

[0455] User: "Can you recommend a nearby cafe?"

[0456] Response: "There's a quiet cafe nearby that you might like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[0457] In this way, the present invention can provide personalized responses based on the user's voice input and analysis of past data.

[0458] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0459] Step 1:

[0460] The user speaks questions or commands into a device such as a smartphone. For example, they might say, "What's the weather like today?" The input is the user's voice data, and the output is the voice signal captured by the device's microphone. The device converts this voice into digital data and sends it to the server using a communication module.

[0461] Step 2:

[0462] The server receives audio data sent from the terminal. The input is digitized audio data, and the output is audio data passed to the NLP module. The server passes this data to the Natural Language Processing (NLP) module, which converts the audio data into text data. Software used includes Google Speech-to-Text API and IBM Watson.

[0463] Step 3:

[0464] The server parses the converted text data. The input is text data ("What's the weather like today?"), and the output is a query based on what the user asked. This query is stored for use in the next processing step.

[0465] Step 4:

[0466] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. The input is the user's identification information (user ID and registered email address), and the output is the user's past interaction data, which includes the applications the user used and their search history.

[0467] Step 5:

[0468] The server analyzes text data and the user's past interaction data. The input consists of text data and interaction data, while the output is feature data indicating the user's preferences and interests. This feature data includes information such as the user's tendency to go out on sunny days. Python libraries such as Pandas and Scikit-learn are used for the analysis.

[0469] Step 6:

[0470] The server trains a generative AI model based on the generated feature data. The input is the feature data, and the output is the trained generative AI model. Examples of generative AI models used include GPT-3 and BERT.

[0471] Step 7:

[0472] The server uses a generative AI model to generate personalized responses for the user. The input is a trained generative AI model and user feature data, and the output is a personalized response. For example, a response like, "It's sunny today. It might get a little cloudy in the afternoon, but you can also check with your favorite weather app," might be generated.

[0473] Step 8:

[0474] The server sends the generated personalized response to the terminal. The input is the personalized response, and the output is the response data sent to the terminal.

[0475] Step 9:

[0476] The device provides the user with the received response in either voice or text format. The input is the response data sent from the server, and the output is the personalized response presented to the user. In voice format, a Text-to-Speech (TTS) engine is used, and the response is delivered via the smartphone's notification system, for example. For instance, the user's screen might display a message such as, "It's sunny today. It might get a little cloudy in the afternoon, but you can also check with your favorite weather app."

[0477] In this way, the system analyzes the user's voice data and provides a response optimized for the user.

[0478] (Application Example 1)

[0479] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0480] Modern food delivery services are expected to rely heavily on voice commands for ordering and recommendations. However, current systems struggle to provide personalized recommendations based on users' past preferences and interests, and voice recognition for ordering also has accuracy issues. As a result, the user experience is not fully optimized.

[0481] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0482] In this invention, the server includes means for acquiring voice data, means for converting voice data into text data, means for acquiring the user's past interaction data, means for analyzing the text data and interaction data to generate feature data, means for generating personalized responses using generative AI, means for providing the generated responses, means for recommending restaurants and dishes, and means for placing food delivery orders using voice recognition. This enables the provision of restaurant and dish recommendations based on the user's preferences and interests, and further enables highly accurate food delivery orders using voice recognition.

[0483] "Audio data" refers to the voice information spoken by the user, and the captured acoustic signal.

[0484] "Text data" refers to information in string format generated by analyzing audio data.

[0485] "Interaction data" refers to information such as the actions, orders, and search history that a user has performed in the past.

[0486] "Feature data" refers to information about user preferences and interests, generated based on text data and interaction data.

[0487] "Generative AI" refers to artificial intelligence technology that receives feature data as input and generates personalized responses.

[0488] "Response" refers to the answers and recommendation information provided to the user by a generative AI.

[0489] A "server" refers to a computer system that is responsible for processing tasks such as analyzing audio data, acquiring interaction data, and executing generative AI.

[0490] "Restaurant and food recommendations" refers to a function that suggests appropriate restaurants and dishes based on the user's preferences and interests.

[0491] "Speech recognition" refers to the technology that analyzes speech data and converts it into text data.

[0492] "Food delivery" refers to a service that delivers food and beverages to the user's location.

[0493] This invention is a system that acquires user voice data, analyzes it, and provides personalized responses. This system mainly consists of a smartphone, a server, and a generative AI.

[0494] Acquisition of audio data

[0495] When a user speaks a question or command into their smartphone, voice data is captured. For example, they might say, "What restaurant would you recommend for dinner tonight?" This voice data is captured by the device and sent to the server in real time.

[0496] Audio data conversion

[0497] The server passes the received audio data to a natural language processing (NLP) module, which converts the audio data into text data. The converted text data will be in the form of "What restaurant would you recommend for dinner tonight?"

[0498] Acquisition of interaction data

[0499] The server uses the user's identification information to retrieve the user's past interaction data from relevant databases. This data includes the applications the user has used and their past order history. For example, it might confirm that the user has a preference for ordering Japanese food in the past.

[0500] Data analysis and feature data generation

[0501] The server analyzes text data and the user's past interaction data to extract user preferences and interests. This generates characteristic data about the user. For example, it might be identified that the user tends to prefer Japanese food. This characteristic data is managed in a way that does not identify individuals.

[0502] Response generation by generative AI

[0503] The server trains a generative AI model based on the generated feature data. The generative AI uses the user's feature data to generate personalized responses. For example, it might create a response that reflects the user's interests, such as, "There's a Japanese restaurant called XYZ nearby. It has good reviews, and is especially known for its delicious sashimi."

[0504] Providing a response

[0505] The device receives personalized responses generated by generative AI models and provides them to the user in voice or text format. For example, if a user asks, "What restaurants would you recommend for dinner tonight?", the system will recommend a Japanese restaurant based on the user's preferences and provide details about its location.

[0506] Specific example

[0507] The following outlines the process when a user uses their smartphone to request restaurant recommendations.

[0508] 1. User: "What restaurants would you recommend for dinner tonight?"

[0509] 2. The device captures the audio and sends the data to the server.

[0510] 3. The server converts the audio data into text. The result of the conversion will be, "What restaurant would you recommend for dinner tonight?"

[0511] 4. The server uses the identification information to retrieve the user's past interaction data from the relevant database. For example, it might confirm that the user prefers Japanese food.

[0512] 5. The server analyzes the data and generates feature data based on user preferences.

[0513] 6. The server uses a generative AI model to generate personalized responses that reflect the user's characteristics. For example, it might respond, "There's a Japanese restaurant called XYZ nearby. It has good reviews, especially for its sashimi."

[0514] 7. The terminal provides the generated response to the user.

[0515] Example of a prompt:

[0516] The user likes Japanese food and is looking for highly-rated restaurants. Text: Tell me about nearby Japanese restaurants. Please recommend restaurants that reflect these criteria.

[0517] In this way, the system can analyze the user's voice and recommend the most suitable restaurants and dishes based on the user's preferences. This improves the user's food delivery experience.

[0518] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0519] Step 1:

[0520] The user verbally asks a question into their smartphone. For example, they might ask, "What restaurants would you recommend for dinner tonight?" This audio data is captured by the device. The input data is an audio signal, and the output is audio data.

[0521] Step 2:

[0522] The device sends the captured audio data to the server. Since the transmitted audio data cannot be analyzed directly, speech recognition is required.

[0523] Step 3:

[0524] The server uses a speech recognition module to convert speech data into text data. This process is performed using natural language processing (NLP) techniques. The input data is speech data, and the output is text data such as "What restaurant would you recommend for dinner tonight?"

[0525] Step 4:

[0526] The server uses the user's identification information to retrieve the user's past interaction data from the relevant database. This data includes past order history and search history. The input data is the user's identification information, and the output is interaction data.

[0527] Step 5:

[0528] The server analyzes text data and the user's past interaction data to generate feature data. This process involves data analysis to extract the user's preferences and interests. The input data consists of text data and interaction data, and the output is feature data that indicates the user's preferences and interests.

[0529] Step 6:

[0530] The server trains a generative AI model using the generated feature data. The generative AI model is used to generate personalized responses based on the feature data. The input data is the feature data, and the output is a personalized response based on the generated prompt sentences.

[0531] Step 7:

[0532] The server sends the generated personalized response to the terminal. This response is then ready to be provided to the user. The input data is the generated response, and the output is the response data directed to the terminal.

[0533] Step 8:

[0534] The terminal provides the user with personalized responses received from the server. These responses are delivered in either voice or text format. Input data is the response data, and output is the information provided to the user. For example, the user might be notified with something like, "There's a Japanese restaurant called XYZ nearby. It has high ratings in online reviews, and is especially known for its delicious sashimi."

[0535] Example of a prompt:

[0536] The user likes Japanese food and is looking for highly-rated restaurants. Text: Tell me about nearby Japanese restaurants. Please recommend restaurants that reflect these criteria.

[0537] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0538] The system for carrying out the present invention has the function of acquiring user voice data, analyzing it, and providing a personalized response. The system mainly includes a terminal, a server, a generative AI, and an emotion engine.

[0539] Acquisition and conversion of audio data

[0540] The user speaks a question or command into their smartphone. For example, they might say, "What's the weather like today?" This voice data is captured by the device and sent to a server in real time. The server passes the received voice data to a natural language processing (NLP) module, which converts the voice data into text data. The converted text data will be in the form of "What's the weather like today?"

[0541] Analysis of emotional data

[0542] After receiving text data, the server passes it to the emotion engine. The emotion engine analyzes the user's emotions from the voice and text data and generates emotion data such as joy, surprise, anger, and sadness. For example, the statement "What's the weather like today?" might be analyzed as containing the user's anxiety or worry.

[0543] Acquisition of customer data

[0544] The server uses the user's identification information to retrieve the user's past interaction data from relevant customer databases (e.g., databases of telecommunications companies or related services). This data includes the applications the user has used and their past search history. For example, it might be confirmed that the user frequently uses a particular weather app.

[0545] Data analysis and feature data generation

[0546] The server uses machine learning algorithms to analyze text data, sentiment data, and the user's past interaction data to generate characteristic data based on the user's preferences and interests. For example, it might identify that the user tends to go out on sunny days or that their current speech contains anxiety. This characteristic data is managed in a way that does not identify individuals.

[0547] Response generation by generative AI

[0548] The server trains a generative AI model based on the generated feature and sentiment data. The generative AI uses the user's feature and sentiment data to generate personalized responses. For example, it creates responses that also take the user's emotions into account, such as, "It's sunny today. It might get a little cloudy in the afternoon, but don't worry. You can also check with your favorite weather app."

[0549] Providing a response

[0550] The device receives personalized responses generated by generative AI models and provides them to the user in voice or text format. For example, if a user asks about cafe locations, the system recommends cafes that suit the user's preferences and current mood, and provides details about the locations. If the user is tired, it might recommend a quiet cafe.

[0551] Specific example

[0552] Situation

[0553] The following outlines the process when a user uses their smartphone to ask for the location of a cafe.

[0554] 1. User: "Can you recommend a nearby cafe?"

[0555] 2. The device captures the audio and sends the data to the server.

[0556] 3. The server converts the voice data into text. The result of the conversion is, "Please tell me about a nearby cafe."

[0557] 4. The server uses the identification information to retrieve the user's past interaction data from the relevant customer database. For example, it might be confirmed that the user prefers quiet cafes.

[0558] 5. The server uses an emotion engine to generate emotion data based on the user's utterances. For example, it might analyze that the user is seeking relaxation.

[0559] 6. The server analyzes the data to generate feature data and sentiment data based on user preferences.

[0560] 7. The server uses generative AI models to generate personalized responses that reflect the user's characteristics and emotions.

[0561] 8. The terminal provides the user with a generated response. For example, it might respond, "There is a quiet, relaxing cafe nearby that you might like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[0562] The present invention, through the process described above, provides information optimized for the user by personalizing responses based on the user's speech content, past data, and even emotional data.

[0563] The following describes the processing flow.

[0564] Step 1:

[0565] The user speaks to their smartphone. For example, they might say, "What's the weather like today?"

[0566] Step 2:

[0567] The device captures the user's speech using its microphone and sends that audio data to the server in real time.

[0568] Step 3:

[0569] The server receives the audio data and passes it to a natural language processing (NLP) module, which converts the audio data into text data. The result of this conversion is "What's the weather like today?"

[0570] Step 4:

[0571] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. This data includes the applications the user has used and their past search history. For example, it might be confirmed that the user frequently uses a particular weather app.

[0572] Step 5:

[0573] The server passes text data to the emotion engine, which analyzes the user's emotions. The emotion engine identifies whether the user is feeling emotions such as joy, surprise, anger, or anxiety based on the tone and content of the voice. For example, the statement "What's the weather like today?" might be analyzed as containing anxiety.

[0574] Step 6:

[0575] The server analyzes sentiment data, text data, and acquired interaction data using machine learning algorithms to generate feature data. This feature data includes things like whether a user tends to go out on sunny days, whether they prefer a particular weather app, and their current emotional state.

[0576] Step 7:

[0577] The server trains a generative AI model based on the generated feature and sentiment data. The generative AI model is configured to generate personalized responses that reflect the user's feature and sentiment data.

[0578] Step 8:

[0579] The device uses a generative AI model to generate personalized responses. These responses take into account the user's preferences and current mood. For example, it might say something like, "It's sunny today. It might get a little cloudy in the afternoon, but don't worry. You can also check your favorite weather app."

[0580] Step 9:

[0581] The device provides the user with a generated, personalized response via voice or text. The user can then review the response and decide on their next action. For example, if the user asks about cafe locations, the system recommends cafes that suit the user's preferences and current mood, and provides details about those locations.

[0582] (Example 2)

[0583] Next, we will describe Example 2. 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".

[0584] Conventional speech recognition systems only convert user speech into text and are unable to provide responses that take into account the user's emotions or past interactions. As a result, they tend to provide generic answers without offering truly personalized and useful information to the user. Furthermore, there is a lack of means to generate and securely manage feature data based on individual user preferences and emotions. This invention aims to solve these problems.

[0585] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0586] In this invention, the server includes means for acquiring user voice data, means for converting voice data into text data, means for analyzing user emotion data, means for acquiring user past interaction data, means for generating feature data by analyzing text data, emotion data and the user's past interaction data, generative AI means for generating personalized responses based on the feature data and emotion data, and means for providing the generated personalized responses to the user. This makes it possible to provide more appropriate and personalized responses that take into account the user's emotions and past interactions.

[0587] "User" refers to an individual or group that uses this system.

[0588] "Voice data" refers to data that records voice information spoken by a user in digital format.

[0589] "Text data" refers to string information obtained by converting audio data using natural language processing technology.

[0590] "Emotional data" refers to data that indicates the emotional state of a user, analyzed and extracted from their voice and text.

[0591] "Past interaction data" refers to information such as the user's past actions, usage history, and search history.

[0592] "Feature data" refers to characteristic information obtained by analyzing a user's preferences, interests, emotions, etc.

[0593] "Generative AI methods" refer to artificial intelligence models that generate optimal responses for users based on feature data and sentiment data.

[0594] "Encryption" refers to the process of transforming information to protect it from unauthorized access.

[0595] "Analysis" refers to the process of thoroughly examining data and extracting its meaning.

[0596] A "speech recognition algorithm" refers to an algorithm that converts speech data into text data.

[0597] A "natural language processing (NLP) module" refers to a software module that implements technology to analyze audio data and convert it into corresponding text data.

[0598] A "personalized response" refers to a response created with the individual user's characteristics and emotions in mind.

[0599] The system for carrying out the present invention has the function of acquiring user voice data, analyzing it, and providing a personalized response. This system mainly includes a user, a terminal, a server, a generative AI model, and an emotion engine.

[0600] First, the user speaks a question or command into a device such as a smartphone. For example, voice input such as "What's the weather like today?" or "Can you tell me about a nearby cafe?" is performed. This voice data is captured by the device and then sent to the server in real time. The hardware used by the device includes a microphone and a communication module, and the transmitted data is encrypted to ensure reliability.

[0601] The server passes the received audio data to a natural language processing (NLP) module, which converts the audio data into text data. This conversion process uses a speech recognition algorithm, so that, for example, an audio statement like "Please tell me about a nearby cafe" is converted into similar text data.

[0602] Next, the server sends the converted text data to the emotion engine for analysis. The emotion engine analyzes the user's emotions from the audio and text data and generates emotion data such as joy, surprise, anger, and sadness. For example, it extracts emotions such as tension and relaxation from the user's tone and way of speaking.

[0603] User interaction data is also a crucial element. Based on user identification information, the server retrieves data such as past operation history, usage history, and search history from relevant databases. This allows for a detailed understanding of what the user likes and what their tendencies are.

[0604] The server combines and analyzes the obtained text data, sentiment data, and past interaction data to generate feature data. This feature data includes user preferences and interests, and machine learning algorithms are used for analysis.

[0605] Based on the generated feature and sentiment data, the server uses a generative AI model to create personalized responses. This allows for the creation of appropriate responses tailored to the user's specific situation and emotions. For example, it could respond with something like, "There's a quiet, relaxing cafe nearby that you'll like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[0606] Ultimately, the generated personalized response is delivered to the user via the device. The response is communicated to the user in voice or text format, and relevant information is specifically presented.

[0607] Specific example

[0608] For example, consider the following scenario where a user asks their smartphone, "Please tell me about nearby cafes."

[0609] 1. The user asks, "Can you tell me about a nearby cafe?"

[0610] 2. The device captures the audio and sends the data to the server.

[0611] 3. The server converts the voice data into text. The result of the conversion is "Please tell me about a nearby cafe."

[0612] 4. The server uses an emotion engine to generate emotion data based on the user's utterances. For example, it might analyze that the user is seeking relaxation.

[0613] 5. The server retrieves the user's past interaction data from the relevant database. For example, it might confirm that the user prefers quiet cafes.

[0614] 6. The server analyzes text data, sentiment data, and past interaction data to generate feature data.

[0615] 7. The server uses a generative AI model to generate personalized responses that reflect the user's characteristics and emotions.

[0616] 8. The terminal provides the user with a generated response. It responds, "There is a quiet, relaxing cafe nearby that you might like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[0617] In this way, the system can personalize its responses based on the user's utterances, past data, and even emotional data, providing information optimized for the user.

[0618] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0619] Step 1: User voice input

[0620] The user speaks into their smartphone. For example, they might say, "Please tell me about a nearby cafe." The device uses its microphone to capture the audio.

[0621] Input: User's voice data

[0622] Output: Captured audio data

[0623] Step 2: Sending the audio data

[0624] The device transmits the captured audio data to the server in real time. During this process, the data is encrypted and securely transferred over the internet.

[0625] Input: Captured audio data

[0626] Output: Audio data sent to the server

[0627] Step 3: Convert speech to text

[0628] The server passes the received audio data to a natural language processing (NLP) module, which converts the audio data into text data. A speech recognition algorithm is used for this process. For example, the audio "Please tell me about a nearby cafe" is converted into text.

[0629] Input: Sent audio data

[0630] Output: Text data "Please tell me about nearby cafes."

[0631] Step 4: Analysis of emotional data

[0632] The server passes the converted text data to the emotion engine. The emotion engine analyzes the user's emotions from the audio and text data and generates emotion data. For example, it might analyze that the user is seeking relaxation.

[0633] Input: Converted text data

[0634] Output: User emotion data (e.g., relaxed)

[0635] Step 5: Obtain customer data

[0636] Based on the user's identification information, the server retrieves customer data, such as past operation history and usage history, from related databases. This includes the user's frequently searched topics and application usage history.

[0637] Input: User identification information

[0638] Output: User's past interaction data

[0639] Step 6: Data analysis and feature data generation

[0640] The server performs analysis based on the obtained text data, sentiment data, and past interaction data. Using machine learning algorithms, it generates feature data representing the user's preferences and interests. For example, the feature that the user prefers quiet cafes might be extracted.

[0641] Input: Text data, sentiment data, past interaction data

[0642] Output: User characteristic data

[0643] Step 7: Generating responses using generative AI

[0644] The server operates a generative AI model based on the generated feature and sentiment data. The generative AI model generates personalized responses tailored to the user's characteristics. For example, it might generate a response like, "There's a quiet, relaxing cafe nearby that you'll like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[0645] Input: Feature data, sentiment data

[0646] Output: Personalized response

[0647] Step 8: Provide a response

[0648] The device receives personalized responses sent from the server and provides them to the user in voice or text format. Users can receive these responses through their smartphone screen or voice assistant.

[0649] Input: Personalized response

[0650] Output: Provides a response to the user.

[0651] (Application Example 2)

[0652] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0653] In today's shopping environment, personalized services that cater to the diverse needs and emotions of consumers are in demand. However, conventional systems struggle to effectively utilize user voice and emotional data, combining them with past interaction history to provide optimal responses. This makes it difficult to deliver a truly satisfying shopping experience. Furthermore, the lack of technology to generate personalized responses in real time is also a challenge.

[0654] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0655] In this invention, the server includes means for acquiring user voice data, means for converting voice data into text data, and means for acquiring the user's past interaction data. This makes it possible to analyze user voice-based questions and commands in a shopping environment in real time, and to provide personalized responses to the user by combining past interaction data and emotion data. Furthermore, by using generative AI means, it is possible to realize the functionality of an optimal shopping assistant that reflects the user's current emotions and past purchase history.

[0656] "User voice data" refers to audio signals that include words and commands spoken by the user.

[0657] "Text data" refers to data obtained by converting audio data into text or written words.

[0658] "User's past interaction data" refers to data related to actions and activities the user has performed in the past, including, for example, past purchase history and logs of applications used.

[0659] "Feature data" refers to data that indicates user preferences and emotions, extracted by analyzing text data and past interaction data.

[0660] "Emotional data" refers to data that indicates the emotional state of a user, analyzed from their speech content and tone of voice, and includes emotions such as joy, surprise, anger, sadness, and anxiety.

[0661] "Generative AI methods" are means of generating personalized responses based on user characteristic data and sentiment data, using machine learning algorithms and generative models.

[0662] A "personalized response" refers to an optimized, individualized answer or recommendation that takes into account each user's preferences and emotions.

[0663] "Means of delivery" refers to the means of communicating the generated personalized response to the user in the form of voice or text.

[0664] The system for carrying out the present invention has the function of acquiring user voice data, analyzing it, and providing a personalized response. The system mainly includes a terminal, a server, a generative AI, and an emotion engine. The configuration and processing of each are described in detail below.

[0665] Acquisition and conversion of audio data

[0666] The user speaks a question or command into their smartphone or smart glasses. For example, they might say, "Which coffee maker do you recommend nearby?" This voice data is captured by the device and sent to a server in real time. The server passes the received voice data to a natural language processing (NLP) module, which converts the voice data into text data. The converted text data will be in the form of "Which coffee maker do you recommend nearby?"

[0667] Analysis of emotional data

[0668] After the server receives the text data, it passes it to the emotion engine. The emotion engine analyzes the user's emotions from the voice and text data and generates emotion data such as joy, relief, surprise, and excitement. For example, if the user says, "What's a good coffee maker nearby?", the emotion engine might analyze that the user is excited.

[0669] Acquisition of customer data

[0670] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. This data includes items the user has previously purchased and services they have used. For example, a record of the user having previously purchased a coffee maker may be found.

[0671] Data analysis and feature data generation

[0672] The server uses machine learning algorithms to analyze text data, sentiment data, and the user's past interaction data to generate characteristic data based on the user's preferences and interests. For example, it might identify that the user has a history of liking espresso or that their current statement contains excitement. This characteristic data is managed in a way that does not identify individuals.

[0673] Response generation by generative AI

[0674] The server generates personalized responses using a generative AI model based on the generated feature and sentiment data. The generative AI uses the user's feature and sentiment data to create an optimal, individually tailored response. For example, it might generate a response that takes the user's emotions into account, such as, "You've always liked espresso, so we recommend the latest espresso maker A. It's on sale right now, so please take this opportunity to try it."

[0675] Providing a response

[0676] The device receives personalized responses generated by a generative AI model and provides them to the user in voice or text format. For example, it might respond, "Nearby, we have coffee maker A, which is perfect for espresso, and it's currently on sale."

[0677] Explanation of specific examples

[0678] For example, if a user asks, "Which coffee maker do you recommend near me?", the system converts the voice into text, performs sentiment analysis, and then, referencing past purchase history and preferences, generates a response such as, "Since you've always liked espresso, we recommend the latest espresso maker A. It's currently on sale, so please take this opportunity to try it."

[0679] Example of a prompt

[0680] User comment: "What coffee maker would you recommend near me?"

[0681] Emotion: {"neg": 0.0, "neu": 0.5, "pos": 0.5, "compound": 0.5}

[0682] Customer information: {"name": "User", "preferences": "Espresso", "history": ["Coffee beans", "Coffee machine"]}

[0683] response:"

[0684] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0685] Step 1:

[0686] Acquisition of user voice data

[0687] The user speaks questions or commands into their smartphone or smart glasses. For example, they might say, "Which coffee maker do you recommend nearby?" This voice data is captured by the device and sent to the server in real time.

[0688] Input: Audio data

[0689] Output: Audio data

[0690] Step 2:

[0691] Converting audio data to text

[0692] The server passes the received audio data to a natural language processing (NLP) module, which converts the audio data into text data. The converted text data will be in the form of "What is the best coffee maker nearby?"

[0693] Input: Audio data

[0694] Output: Text data

[0695] Step 3:

[0696] Analysis of emotional data

[0697] After the server receives the text data, it passes it to the emotion engine. The emotion engine analyzes the user's emotions from the voice and text data and generates emotion data such as joy, relief, surprise, and excitement. For example, if the user says, "What's a good coffee maker nearby?", the emotion engine might analyze that the user is excited.

[0698] Input: Audio data, text data

[0699] Output: Sentiment data

[0700] Step 4:

[0701] Acquisition of customer data

[0702] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. This data includes items the user has previously purchased and services they have used. For example, a record of the user having previously purchased a coffee maker may be found.

[0703] Input: User identification information

[0704] Output: Past interaction data

[0705] Step 5:

[0706] Data analysis and feature data generation

[0707] The server uses machine learning algorithms to analyze text data, sentiment data, and past user interaction data to generate characteristic data based on the user's preferences and interests. For example, it might reveal that the user tends to like espresso and identify if their current statement contains excitement. This characteristic data is managed in a way that does not identify individuals.

[0708] Input: Text data, sentiment data, historical interaction data

[0709] Output: Feature data

[0710] Step 6:

[0711] Response generation by generative AI

[0712] The server uses a generative AI model to generate personalized responses based on the generated feature data and sentiment data. The generative AI utilizes the user's feature data and sentiment data to create appropriate responses tailored to each individual, such as, "You've always liked espresso, so we recommend the latest espresso maker A. It's on sale right now, so please take this opportunity to try it."

[0713] Input: Feature data, sentiment data

[0714] Output: Personalized response

[0715] Step 7:

[0716] Providing a response

[0717] The device receives personalized responses generated by a generative AI model and provides them to the user in voice or text format. For example, it might respond, "Nearby, we have coffee maker A, which is perfect for espresso, and it's currently on sale."

[0718] Input: Personalized response

[0719] Output: Voice or text response

[0720] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0721] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0722] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0723] [Third Embodiment]

[0724] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0725] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0726] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0727] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0728] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0729] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0730] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0731] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0732] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0733] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0734] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0735] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0736] The system for carrying out the present invention has the function of acquiring user voice data, analyzing it, and providing a personalized response. The system mainly includes a terminal, a server, and a generative AI.

[0737] Acquisition and conversion of audio data

[0738] The user speaks a question or command into their smartphone. For example, they might say, "What's the weather like today?" This voice data is captured by the device and sent to a server in real time. The server passes the received voice data to a natural language processing (NLP) module, which converts the voice data into text data. The converted text data will be in the form of "What's the weather like today?"

[0739] Acquisition of customer data

[0740] The server uses the user's identification information to retrieve the user's past interaction data from relevant customer databases (e.g., databases of telecommunications companies or related services). This data includes the applications the user has used and their past search history. For example, it might be confirmed that the user frequently uses a particular weather app.

[0741] Data analysis and feature data generation

[0742] The server analyzes text data and the user's past interaction data to extract user preferences and interests. This generates characteristic data about the user. For example, it might be identified that the user tends to go out on sunny days. This characteristic data is managed in a way that does not identify individuals.

[0743] Response generation by generative AI

[0744] The server trains a generative AI model based on the generated feature data. The generative AI uses the user's feature data to generate personalized responses. For example, it creates responses that reflect the user's interests, such as, "It's sunny today. It might get a little cloudy in the afternoon, but you can also check with your favorite weather app."

[0745] Providing a response

[0746] The device receives personalized responses generated by generative AI models and provides them to the user in voice or text format. For example, if a user asks about a specific cafe, the system will recommend cafes that suit the user's preferences and provide details about their locations.

[0747] Specific example

[0748] Situation

[0749] The following outlines the process when a user uses their smartphone to ask for the location of a cafe.

[0750] 1. User: "Can you recommend a nearby cafe?"

[0751] 2. The device captures the audio and sends the data to the server.

[0752] 3. The server converts the voice data into text. The result of the conversion is, "Please tell me about a nearby cafe."

[0753] 4. The server uses the identification information to retrieve the user's past interaction data from the relevant customer database. For example, it might be confirmed that the user prefers quiet cafes.

[0754] 5. The server analyzes the data and generates feature data based on user preferences.

[0755] 6. The server uses a generative AI model to generate personalized responses that reflect the user's characteristics.

[0756] 7. The terminal provides the user with a generated response. For example, it might respond, "There's a quiet cafe nearby that you might like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[0757] As described above, this invention can provide users with optimized information by personalizing responses based on the user's utterances and past data.

[0758] The following describes the processing flow.

[0759] Step 1:

[0760] The user speaks to their smartphone. For example, they might say, "What's the weather like today?"

[0761] Step 2:

[0762] The device captures the user's speech using its microphone and sends that audio data to the server in real time.

[0763] Step 3:

[0764] The server receives the audio data and passes it to a natural language processing (NLP) module, which converts the audio data into text data. The result of this conversion is "What's the weather like today?"

[0765] Step 4:

[0766] The server uses the user's identification information to retrieve the user's past interaction data from databases of software providers and integrated services. This data includes the applications the user uses and their past search history.

[0767] Step 5:

[0768] The server analyzes text data and acquired interaction data using machine learning algorithms to generate feature data based on user preferences and interests. For example, it might identify that a user prefers to use a particular weather app or tends to go out on sunny days.

[0769] Step 6:

[0770] The server trains a generative AI model based on the analyzed feature data. The generative AI is then adapted to generate personalized responses using the user's feature data.

[0771] Step 7:

[0772] The device generates personalized responses through a generative AI model. These responses reflect the user's preferences and past behavior. For example, they might say, "It's sunny today. It might get a little cloudy this afternoon."

[0773] Step 8:

[0774] The device provides the user with a personalized response via voice or text. The user can then review the response and decide on their next course of action. In this way, the system provides the user with optimized information.

[0775] (Example 1)

[0776] Next, we will describe Example 1. 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."

[0777] Conventional voice assistant systems have struggled to generate personalized responses by fully utilizing user preferences and past interaction data. Furthermore, in order to provide information best suited to each individual user, there has been insufficient management of user-specific data and adequate measures to protect personal information. This invention aims to solve these problems and provide users with appropriate and personalized information quickly.

[0778] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0779] In this invention, the server includes means for acquiring user voice data, means for converting voice data into text data, means for acquiring the user's past interaction data, means for analyzing and generating feature data, means for training a generative AI model, means for generating personalized responses, and means for providing responses to the user. This makes it possible to provide personalized responses quickly and accurately based on the user's preferences and interests.

[0780] A "user" is someone who uses the system to input voice data and receive a personalized response.

[0781] "Voice data" refers to the audio recorded when a user speaks into their smartphone or other device.

[0782] "Text data" refers to data obtained by converting audio data into text format.

[0783] "Interaction data" refers to a record of a user's past interactions with the system.

[0784] "Feature data" refers to data that indicates characteristics such as user preferences and interests.

[0785] A "generative AI model" is an artificial intelligence model designed to generate personalized responses based on user characteristic data.

[0786] "Means" refers to the devices or methods used to achieve an objective.

[0787] A "server" is a central processing unit that performs audio data processing, text conversion, and acquisition and analysis of interaction data.

[0788] A "terminal" is a device that a user uses to input voice data and receive a personalized response.

[0789] A "personalized response" is a customized reply generated based on the user's characteristic data.

[0790] The system for carrying out the present invention acquires user voice data, analyzes it, and provides a personalized response. The system components include a terminal, a server, and a generative AI model.

[0791] overview

[0792] The user speaks a question or command into a device such as a smartphone. For example, they might say, "What's the weather like today?" This voice data is captured by the device and sent to the server in real time. The server passes the received voice data to a natural language processing (NLP) module, which converts the voice data into text data. The converted text data will be in the form of "What's the weather like today?".

[0793] Acquisition and conversion of audio data

[0794] When a user speaks a question or command to their smartphone, the device uses its built-in microphone to capture voice data. This voice data is then transmitted to a server in real time via an internet connection. Specifically, the software used is an audio capture module, and the hardware used is the smartphone's built-in microphone.

[0795] Server-based processing

[0796] The server passes the received audio data to a natural language processing module, which uses text conversion software such as the Google Speech-to-Text API or IBM Watson to convert the audio data into text. This converted text data is then used for subsequent analysis.

[0797] User data collection and analysis

[0798] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. This data includes the applications the user has used and their past search history. For example, it might confirm that the user frequently uses a particular weather app.

[0799] The server analyzes text data and the user's past interaction data to extract user preferences and interests. This generates feature data about the user. Analysis tools such as Python's Pandas and Scikit-learn are used for the analysis. For example, if it is found that a user tends to go out on sunny days, this information is saved as feature data.

[0800] Response generation using generative AI models

[0801] The server trains a generative AI model based on the generated feature data. Examples of generative AI models used include GPT-3 and BERT. The generative AI model uses the user's feature data to generate personalized responses. For example, a possible response might be, "It's sunny today. It might get a little cloudy this afternoon, but you can check this with your usual weather app."

[0802] Providing a response

[0803] Finally, the server sends the generated personalized response to the device. The device then provides this response to the user in either voice or text format. In the voice format, a Text-to-Speech (TTS) engine is used, and the response is delivered via the smartphone's notification system, for example. For instance, the user's screen might display a message such as, "It's sunny today. It might get a little cloudy in the afternoon, but you can also check your favorite weather app."

[0804] Examples of specific prompt messages

[0805] The following are examples of prompt messages in specific situations.

[0806] User: "Can you recommend a nearby cafe?"

[0807] Response: "There's a quiet cafe nearby that you might like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[0808] In this way, the present invention can provide personalized responses based on the user's voice input and analysis of past data.

[0809] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0810] Step 1:

[0811] The user speaks questions or commands into a device such as a smartphone. For example, they might say, "What's the weather like today?" The input is the user's voice data, and the output is the voice signal captured by the device's microphone. The device converts this voice into digital data and sends it to the server using a communication module.

[0812] Step 2:

[0813] The server receives audio data sent from the terminal. The input is digitized audio data, and the output is audio data passed to the NLP module. The server passes this data to the Natural Language Processing (NLP) module, which converts the audio data into text data. Software used includes Google Speech-to-Text API and IBM Watson.

[0814] Step 3:

[0815] The server parses the converted text data. The input is text data ("What's the weather like today?"), and the output is a query based on what the user asked. This query is stored for use in the next processing step.

[0816] Step 4:

[0817] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. The input is the user's identification information (user ID and registered email address), and the output is the user's past interaction data, which includes the applications the user used and their search history.

[0818] Step 5:

[0819] The server analyzes text data and the user's past interaction data. The input consists of text data and interaction data, while the output is feature data indicating the user's preferences and interests. This feature data includes information such as the user's tendency to go out on sunny days. Python libraries such as Pandas and Scikit-learn are used for the analysis.

[0820] Step 6:

[0821] The server trains a generative AI model based on the generated feature data. The input is the feature data, and the output is the trained generative AI model. Examples of generative AI models used include GPT-3 and BERT.

[0822] Step 7:

[0823] The server uses a generative AI model to generate personalized responses for the user. The input is a trained generative AI model and user feature data, and the output is a personalized response. For example, a response like, "It's sunny today. It might get a little cloudy in the afternoon, but you can also check with your favorite weather app," might be generated.

[0824] Step 8:

[0825] The server sends the generated personalized response to the terminal. The input is the personalized response, and the output is the response data sent to the terminal.

[0826] Step 9:

[0827] The device provides the user with the received response in either voice or text format. The input is the response data sent from the server, and the output is the personalized response presented to the user. In voice format, a Text-to-Speech (TTS) engine is used, and the response is delivered via the smartphone's notification system, for example. For instance, the user's screen might display a message such as, "It's sunny today. It might get a little cloudy in the afternoon, but you can also check with your favorite weather app."

[0828] In this way, the system analyzes the user's voice data and provides a response optimized for the user.

[0829] (Application Example 1)

[0830] Next, we will explain Application Example 1. In the following explanation, 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."

[0831] Modern food delivery services are expected to rely heavily on voice commands for ordering and recommendations. However, current systems struggle to provide personalized recommendations based on users' past preferences and interests, and voice recognition for ordering also has accuracy issues. As a result, the user experience is not fully optimized.

[0832] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0833] In this invention, the server includes means for acquiring voice data, means for converting voice data into text data, means for acquiring the user's past interaction data, means for analyzing the text data and interaction data to generate feature data, means for generating personalized responses using generative AI, means for providing the generated responses, means for recommending restaurants and dishes, and means for placing food delivery orders using voice recognition. This enables the provision of restaurant and dish recommendations based on the user's preferences and interests, and further enables highly accurate food delivery orders using voice recognition.

[0834] "Audio data" refers to the voice information spoken by the user, and the captured acoustic signal.

[0835] "Text data" refers to information in string format generated by analyzing audio data.

[0836] "Interaction data" refers to information such as the actions, orders, and search history that a user has performed in the past.

[0837] "Feature data" refers to information about user preferences and interests, generated based on text data and interaction data.

[0838] "Generative AI" refers to artificial intelligence technology that receives feature data as input and generates personalized responses.

[0839] "Response" refers to the answers and recommendation information provided to the user by a generative AI.

[0840] A "server" refers to a computer system that is responsible for processing tasks such as analyzing audio data, acquiring interaction data, and executing generative AI.

[0841] "Restaurant and food recommendations" refers to a function that suggests appropriate restaurants and dishes based on the user's preferences and interests.

[0842] "Speech recognition" refers to the technology that analyzes speech data and converts it into text data.

[0843] "Food delivery" refers to a service that delivers food and beverages to the user's location.

[0844] This invention is a system that acquires user voice data, analyzes it, and provides personalized responses. This system mainly consists of a smartphone, a server, and a generative AI.

[0845] Acquisition of audio data

[0846] When a user speaks a question or command into their smartphone, voice data is captured. For example, they might say, "What restaurant would you recommend for dinner tonight?" This voice data is captured by the device and sent to the server in real time.

[0847] Audio data conversion

[0848] The server passes the received audio data to a natural language processing (NLP) module, which converts the audio data into text data. The converted text data will be in the form of "What restaurant would you recommend for dinner tonight?"

[0849] Acquisition of interaction data

[0850] The server uses the user's identification information to retrieve the user's past interaction data from relevant databases. This data includes the applications the user has used and their past order history. For example, it might confirm that the user has a preference for ordering Japanese food in the past.

[0851] Data analysis and feature data generation

[0852] The server analyzes text data and the user's past interaction data to extract user preferences and interests. This generates characteristic data about the user. For example, it might be identified that the user tends to prefer Japanese food. This characteristic data is managed in a way that does not identify individuals.

[0853] Response generation by generative AI

[0854] The server trains a generative AI model based on the generated feature data. The generative AI uses the user's feature data to generate personalized responses. For example, it might create a response that reflects the user's interests, such as, "There's a Japanese restaurant called XYZ nearby. It has good reviews, and is especially known for its delicious sashimi."

[0855] Providing a response

[0856] The device receives personalized responses generated by generative AI models and provides them to the user in voice or text format. For example, if a user asks, "What restaurants would you recommend for dinner tonight?", the system will recommend a Japanese restaurant based on the user's preferences and provide details about its location.

[0857] Specific example

[0858] The following outlines the process when a user uses their smartphone to request restaurant recommendations.

[0859] 1. User: "What restaurants would you recommend for dinner tonight?"

[0860] 2. The device captures the audio and sends the data to the server.

[0861] 3. The server converts the audio data into text. The result of the conversion will be, "What restaurant would you recommend for dinner tonight?"

[0862] 4. The server uses the identification information to retrieve the user's past interaction data from the relevant database. For example, it might confirm that the user prefers Japanese food.

[0863] 5. The server analyzes the data and generates feature data based on user preferences.

[0864] 6. The server uses a generative AI model to generate personalized responses that reflect the user's characteristics. For example, it might respond, "There's a Japanese restaurant called XYZ nearby. It has good reviews, especially for its sashimi."

[0865] 7. The terminal provides the generated response to the user.

[0866] Example of a prompt:

[0867] The user likes Japanese food and is looking for highly-rated restaurants. Text: Tell me about nearby Japanese restaurants. Please recommend restaurants that reflect these criteria.

[0868] In this way, the system can analyze the user's voice and recommend the most suitable restaurants and dishes based on the user's preferences. This improves the user's food delivery experience.

[0869] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0870] Step 1:

[0871] The user verbally asks a question into their smartphone. For example, they might ask, "What restaurants would you recommend for dinner tonight?" This audio data is captured by the device. The input data is an audio signal, and the output is audio data.

[0872] Step 2:

[0873] The device sends the captured audio data to the server. Since the transmitted audio data cannot be analyzed directly, speech recognition is required.

[0874] Step 3:

[0875] The server uses a speech recognition module to convert speech data into text data. This process is performed using natural language processing (NLP) techniques. The input data is speech data, and the output is text data such as "What restaurant would you recommend for dinner tonight?"

[0876] Step 4:

[0877] The server uses the user's identification information to retrieve the user's past interaction data from the relevant database. This data includes past order history and search history. The input data is the user's identification information, and the output is interaction data.

[0878] Step 5:

[0879] The server analyzes text data and the user's past interaction data to generate feature data. This process involves data analysis to extract the user's preferences and interests. The input data consists of text data and interaction data, and the output is feature data that indicates the user's preferences and interests.

[0880] Step 6:

[0881] The server trains a generative AI model using the generated feature data. The generative AI model is used to generate personalized responses based on the feature data. The input data is the feature data, and the output is a personalized response based on the generated prompt sentences.

[0882] Step 7:

[0883] The server sends the generated personalized response to the terminal. This response is then ready to be provided to the user. The input data is the generated response, and the output is the response data directed to the terminal.

[0884] Step 8:

[0885] The terminal provides the user with personalized responses received from the server. These responses are delivered in either voice or text format. Input data is the response data, and output is the information provided to the user. For example, the user might be notified with something like, "There's a Japanese restaurant called XYZ nearby. It has high ratings in online reviews, and is especially known for its delicious sashimi."

[0886] Example of a prompt:

[0887] The user likes Japanese food and is looking for highly-rated restaurants. Text: Tell me about nearby Japanese restaurants. Please recommend restaurants that reflect these criteria.

[0888] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0889] The system for carrying out the present invention has the function of acquiring user voice data, analyzing it, and providing a personalized response. The system mainly includes a terminal, a server, a generative AI, and an emotion engine.

[0890] Acquisition and conversion of audio data

[0891] The user speaks a question or command into their smartphone. For example, they might say, "What's the weather like today?" This voice data is captured by the device and sent to a server in real time. The server passes the received voice data to a natural language processing (NLP) module, which converts the voice data into text data. The converted text data will be in the form of "What's the weather like today?"

[0892] Analysis of emotional data

[0893] After receiving text data, the server passes it to the emotion engine. The emotion engine analyzes the user's emotions from the voice and text data and generates emotion data such as joy, surprise, anger, and sadness. For example, the statement "What's the weather like today?" might be analyzed as containing the user's anxiety or worry.

[0894] Acquisition of customer data

[0895] The server uses the user's identification information to retrieve the user's past interaction data from relevant customer databases (e.g., databases of telecommunications companies or related services). This data includes the applications the user has used and their past search history. For example, it might be confirmed that the user frequently uses a particular weather app.

[0896] Data analysis and feature data generation

[0897] The server uses machine learning algorithms to analyze text data, sentiment data, and the user's past interaction data to generate characteristic data based on the user's preferences and interests. For example, it might identify that the user tends to go out on sunny days or that their current speech contains anxiety. This characteristic data is managed in a way that does not identify individuals.

[0898] Response generation by generative AI

[0899] The server trains a generative AI model based on the generated feature and sentiment data. The generative AI uses the user's feature and sentiment data to generate personalized responses. For example, it creates responses that also take the user's emotions into account, such as, "It's sunny today. It might get a little cloudy in the afternoon, but don't worry. You can also check with your favorite weather app."

[0900] Providing a response

[0901] The device receives personalized responses generated by generative AI models and provides them to the user in voice or text format. For example, if a user asks about cafe locations, the system recommends cafes that suit the user's preferences and current mood, and provides details about the locations. If the user is tired, it might recommend a quiet cafe.

[0902] Specific example

[0903] Situation

[0904] The following outlines the process when a user uses their smartphone to ask for the location of a cafe.

[0905] 1. User: "Can you recommend a nearby cafe?"

[0906] 2. The device captures the audio and sends the data to the server.

[0907] 3. The server converts the voice data into text. The result of the conversion is, "Please tell me about a nearby cafe."

[0908] 4. The server uses the identification information to retrieve the user's past interaction data from the relevant customer database. For example, it might be confirmed that the user prefers quiet cafes.

[0909] 5. The server uses an emotion engine to generate emotion data based on the user's utterances. For example, it might analyze that the user is seeking relaxation.

[0910] 6. The server analyzes the data to generate feature data and sentiment data based on user preferences.

[0911] 7. The server uses generative AI models to generate personalized responses that reflect the user's characteristics and emotions.

[0912] 8. The terminal provides the user with a generated response. For example, it might respond, "There is a quiet, relaxing cafe nearby that you might like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[0913] The present invention, through the process described above, provides information optimized for the user by personalizing responses based on the user's speech content, past data, and even emotional data.

[0914] The following describes the processing flow.

[0915] Step 1:

[0916] The user speaks to their smartphone. For example, they might say, "What's the weather like today?"

[0917] Step 2:

[0918] The device captures the user's speech using its microphone and sends that audio data to the server in real time.

[0919] Step 3:

[0920] The server receives the audio data and passes it to a natural language processing (NLP) module, which converts the audio data into text data. The result of this conversion is "What's the weather like today?"

[0921] Step 4:

[0922] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. This data includes the applications the user has used and their past search history. For example, it might be confirmed that the user frequently uses a particular weather app.

[0923] Step 5:

[0924] The server passes text data to the emotion engine, which analyzes the user's emotions. The emotion engine identifies whether the user is feeling emotions such as joy, surprise, anger, or anxiety based on the tone and content of the voice. For example, the statement "What's the weather like today?" might be analyzed as containing anxiety.

[0925] Step 6:

[0926] The server analyzes sentiment data, text data, and acquired interaction data using machine learning algorithms to generate feature data. This feature data includes things like whether a user tends to go out on sunny days, whether they prefer a particular weather app, and their current emotional state.

[0927] Step 7:

[0928] The server trains a generative AI model based on the generated feature and sentiment data. The generative AI model is configured to generate personalized responses that reflect the user's feature and sentiment data.

[0929] Step 8:

[0930] The device uses a generative AI model to generate personalized responses. These responses take into account the user's preferences and current mood. For example, it might say something like, "It's sunny today. It might get a little cloudy in the afternoon, but don't worry. You can also check your favorite weather app."

[0931] Step 9:

[0932] The device provides the user with a generated, personalized response via voice or text. The user can then review the response and decide on their next action. For example, if the user asks about cafe locations, the system recommends cafes that suit the user's preferences and current mood, and provides details about those locations.

[0933] (Example 2)

[0934] Next, we will describe Example 2. 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."

[0935] Conventional speech recognition systems only convert user speech into text and are unable to provide responses that take into account the user's emotions or past interactions. As a result, they tend to provide generic answers without offering truly personalized and useful information to the user. Furthermore, there is a lack of means to generate and securely manage feature data based on individual user preferences and emotions. This invention aims to solve these problems.

[0936] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0937] In this invention, the server includes means for acquiring user voice data, means for converting voice data into text data, means for analyzing user emotion data, means for acquiring user past interaction data, means for generating feature data by analyzing text data, emotion data and the user's past interaction data, generative AI means for generating personalized responses based on the feature data and emotion data, and means for providing the generated personalized responses to the user. This makes it possible to provide more appropriate and personalized responses that take into account the user's emotions and past interactions.

[0938] "User" refers to an individual or group that uses this system.

[0939] "Voice data" refers to data that records voice information spoken by a user in digital format.

[0940] "Text data" refers to string information obtained by converting audio data using natural language processing technology.

[0941] "Emotional data" refers to data that indicates the emotional state of a user, analyzed and extracted from their voice and text.

[0942] "Past interaction data" refers to information such as the user's past actions, usage history, and search history.

[0943] "Feature data" refers to characteristic information obtained by analyzing a user's preferences, interests, emotions, etc.

[0944] "Generative AI methods" refer to artificial intelligence models that generate optimal responses for users based on feature data and sentiment data.

[0945] "Encryption" refers to the process of transforming information to protect it from unauthorized access.

[0946] "Analysis" refers to the process of thoroughly examining data and extracting its meaning.

[0947] A "speech recognition algorithm" refers to an algorithm that converts speech data into text data.

[0948] A "natural language processing (NLP) module" refers to a software module that implements technology to analyze audio data and convert it into corresponding text data.

[0949] A "personalized response" refers to a response created with the individual user's characteristics and emotions in mind.

[0950] The system for carrying out the present invention has the function of acquiring user voice data, analyzing it, and providing a personalized response. This system mainly includes a user, a terminal, a server, a generative AI model, and an emotion engine.

[0951] First, the user speaks a question or command into a device such as a smartphone. For example, voice input such as "What's the weather like today?" or "Can you tell me about a nearby cafe?" is performed. This voice data is captured by the device and then sent to the server in real time. The hardware used by the device includes a microphone and a communication module, and the transmitted data is encrypted to ensure reliability.

[0952] The server passes the received audio data to a natural language processing (NLP) module, which converts the audio data into text data. This conversion process uses a speech recognition algorithm, so that, for example, an audio statement like "Please tell me about a nearby cafe" is converted into similar text data.

[0953] Next, the server sends the converted text data to the emotion engine for analysis. The emotion engine analyzes the user's emotions from the audio and text data and generates emotion data such as joy, surprise, anger, and sadness. For example, it extracts emotions such as tension and relaxation from the user's tone and way of speaking.

[0954] User interaction data is also a crucial element. Based on user identification information, the server retrieves data such as past operation history, usage history, and search history from relevant databases. This allows for a detailed understanding of what the user likes and what their tendencies are.

[0955] The server combines and analyzes the obtained text data, sentiment data, and past interaction data to generate feature data. This feature data includes user preferences and interests, and machine learning algorithms are used for analysis.

[0956] Based on the generated feature and sentiment data, the server uses a generative AI model to create personalized responses. This allows for the creation of appropriate responses tailored to the user's specific situation and emotions. For example, it could respond with something like, "There's a quiet, relaxing cafe nearby that you'll like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[0957] Ultimately, the generated personalized response is delivered to the user via the device. The response is communicated to the user in voice or text format, and relevant information is specifically presented.

[0958] Specific example

[0959] For example, consider the following scenario where a user asks their smartphone, "Please tell me about nearby cafes."

[0960] 1. The user asks, "Can you tell me about a nearby cafe?"

[0961] 2. The device captures the audio and sends the data to the server.

[0962] 3. The server converts the voice data into text. The result of the conversion is "Please tell me about a nearby cafe."

[0963] 4. The server uses an emotion engine to generate emotion data based on the user's utterances. For example, it might analyze that the user is seeking relaxation.

[0964] 5. The server retrieves the user's past interaction data from the relevant database. For example, it might confirm that the user prefers quiet cafes.

[0965] 6. The server analyzes text data, sentiment data, and past interaction data to generate feature data.

[0966] 7. The server uses a generative AI model to generate personalized responses that reflect the user's characteristics and emotions.

[0967] 8. The terminal provides the user with a generated response. It responds, "There is a quiet, relaxing cafe nearby that you might like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[0968] In this way, the system can personalize its responses based on the user's utterances, past data, and even emotional data, providing information optimized for the user.

[0969] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0970] Step 1: User voice input

[0971] The user speaks into their smartphone. For example, they might say, "Please tell me about a nearby cafe." The device uses its microphone to capture the audio.

[0972] Input: User's voice data

[0973] Output: Captured audio data

[0974] Step 2: Sending the audio data

[0975] The device transmits the captured audio data to the server in real time. During this process, the data is encrypted and securely transferred over the internet.

[0976] Input: Captured audio data

[0977] Output: Audio data sent to the server

[0978] Step 3: Convert speech to text

[0979] The server passes the received audio data to a natural language processing (NLP) module, which converts the audio data into text data. A speech recognition algorithm is used for this process. For example, the audio "Please tell me about a nearby cafe" is converted into text.

[0980] Input: Sent audio data

[0981] Output: Text data "Please tell me about nearby cafes."

[0982] Step 4: Analysis of emotional data

[0983] The server passes the converted text data to the emotion engine. The emotion engine analyzes the user's emotions from the audio and text data and generates emotion data. For example, it might analyze that the user is seeking relaxation.

[0984] Input: Converted text data

[0985] Output: User emotion data (e.g., relaxed)

[0986] Step 5: Obtain customer data

[0987] Based on the user's identification information, the server retrieves customer data, such as past operation history and usage history, from related databases. This includes the user's frequently searched topics and application usage history.

[0988] Input: User identification information

[0989] Output: User's past interaction data

[0990] Step 6: Data analysis and feature data generation

[0991] The server performs analysis based on the obtained text data, sentiment data, and past interaction data. Using machine learning algorithms, it generates feature data representing the user's preferences and interests. For example, the feature that the user prefers quiet cafes might be extracted.

[0992] Input: Text data, sentiment data, past interaction data

[0993] Output: User characteristic data

[0994] Step 7: Generating responses using generative AI

[0995] The server operates a generative AI model based on the generated feature and sentiment data. The generative AI model generates personalized responses tailored to the user's characteristics. For example, it might generate a response like, "There's a quiet, relaxing cafe nearby that you'll like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[0996] Input: Feature data, sentiment data

[0997] Output: Personalized response

[0998] Step 8: Provide a response

[0999] The device receives personalized responses sent from the server and provides them to the user in voice or text format. Users can receive these responses through their smartphone screen or voice assistant.

[1000] Input: Personalized response

[1001] Output: Provides a response to the user.

[1002] (Application Example 2)

[1003] Next, we will explain application example 2. In the following explanation, 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."

[1004] In today's shopping environment, personalized services that cater to the diverse needs and emotions of consumers are in demand. However, conventional systems struggle to effectively utilize user voice and emotional data, combining them with past interaction history to provide optimal responses. This makes it difficult to deliver a truly satisfying shopping experience. Furthermore, the lack of technology to generate personalized responses in real time is also a challenge.

[1005] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1006] In this invention, the server includes means for acquiring user voice data, means for converting voice data into text data, and means for acquiring the user's past interaction data. This makes it possible to analyze user voice-based questions and commands in a shopping environment in real time, and to provide personalized responses to the user by combining past interaction data and emotion data. Furthermore, by using generative AI means, it is possible to realize the functionality of an optimal shopping assistant that reflects the user's current emotions and past purchase history.

[1007] "User voice data" refers to audio signals that include words and commands spoken by the user.

[1008] "Text data" refers to data obtained by converting audio data into text or written words.

[1009] "User's past interaction data" refers to data related to actions and activities the user has performed in the past, including, for example, past purchase history and logs of applications used.

[1010] "Feature data" refers to data that indicates user preferences and emotions, extracted by analyzing text data and past interaction data.

[1011] "Emotional data" refers to data that indicates the emotional state of a user, analyzed from their speech content and tone of voice, and includes emotions such as joy, surprise, anger, sadness, and anxiety.

[1012] "Generative AI methods" are means of generating personalized responses based on user characteristic data and sentiment data, using machine learning algorithms and generative models.

[1013] A "personalized response" refers to an optimized, individualized answer or recommendation that takes into account each user's preferences and emotions.

[1014] "Means of delivery" refers to the means of communicating the generated personalized response to the user in the form of voice or text.

[1015] The system for carrying out the present invention has the function of acquiring user voice data, analyzing it, and providing a personalized response. The system mainly includes a terminal, a server, a generative AI, and an emotion engine. The configuration and processing of each are described in detail below.

[1016] Acquisition and conversion of audio data

[1017] The user speaks a question or command into their smartphone or smart glasses. For example, they might say, "Which coffee maker do you recommend nearby?" This voice data is captured by the device and sent to a server in real time. The server passes the received voice data to a natural language processing (NLP) module, which converts the voice data into text data. The converted text data will be in the form of "Which coffee maker do you recommend nearby?"

[1018] Analysis of emotional data

[1019] After the server receives the text data, it passes it to the emotion engine. The emotion engine analyzes the user's emotions from the voice and text data and generates emotion data such as joy, relief, surprise, and excitement. For example, if the user says, "What's a good coffee maker nearby?", the emotion engine might analyze that the user is excited.

[1020] Acquisition of customer data

[1021] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. This data includes items the user has previously purchased and services they have used. For example, a record of the user having previously purchased a coffee maker may be found.

[1022] Data analysis and feature data generation

[1023] The server uses machine learning algorithms to analyze text data, sentiment data, and the user's past interaction data to generate characteristic data based on the user's preferences and interests. For example, it might identify that the user has a history of liking espresso or that their current statement contains excitement. This characteristic data is managed in a way that does not identify individuals.

[1024] Response generation by generative AI

[1025] The server generates personalized responses using a generative AI model based on the generated feature and sentiment data. The generative AI uses the user's feature and sentiment data to create an optimal, individually tailored response. For example, it might generate a response that takes the user's emotions into account, such as, "You've always liked espresso, so we recommend the latest espresso maker A. It's on sale right now, so please take this opportunity to try it."

[1026] Providing a response

[1027] The device receives personalized responses generated by a generative AI model and provides them to the user in voice or text format. For example, it might respond, "Nearby, we have coffee maker A, which is perfect for espresso, and it's currently on sale."

[1028] Explanation of specific examples

[1029] For example, if a user asks, "Which coffee maker do you recommend near me?", the system converts the voice into text, performs sentiment analysis, and then, referencing past purchase history and preferences, generates a response such as, "Since you've always liked espresso, we recommend the latest espresso maker A. It's currently on sale, so please take this opportunity to try it."

[1030] Example of a prompt

[1031] User comment: "What coffee maker would you recommend near me?"

[1032] Emotion: {"neg": 0.0, "neu": 0.5, "pos": 0.5, "compound": 0.5}

[1033] Customer information: {"name": "User", "preferences": "Espresso", "history": ["Coffee beans", "Coffee machine"]}

[1034] response:"

[1035] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1036] Step 1:

[1037] Acquisition of user voice data

[1038] The user speaks questions or commands into their smartphone or smart glasses. For example, they might say, "Which coffee maker do you recommend nearby?" This voice data is captured by the device and sent to the server in real time.

[1039] Input: Audio data

[1040] Output: Audio data

[1041] Step 2:

[1042] Converting audio data to text

[1043] The server passes the received audio data to a natural language processing (NLP) module, which converts the audio data into text data. The converted text data will be in the form of "What is the best coffee maker nearby?"

[1044] Input: Audio data

[1045] Output: Text data

[1046] Step 3:

[1047] Analysis of emotional data

[1048] After the server receives the text data, it passes it to the emotion engine. The emotion engine analyzes the user's emotions from the voice and text data and generates emotion data such as joy, relief, surprise, and excitement. For example, if the user says, "What's a good coffee maker nearby?", the emotion engine might analyze that the user is excited.

[1049] Input: Audio data, text data

[1050] Output: Sentiment data

[1051] Step 4:

[1052] Acquisition of customer data

[1053] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. This data includes items the user has previously purchased and services they have used. For example, a record of the user having previously purchased a coffee maker may be found.

[1054] Input: User identification information

[1055] Output: Past interaction data

[1056] Step 5:

[1057] Data analysis and feature data generation

[1058] The server uses machine learning algorithms to analyze text data, sentiment data, and past user interaction data to generate characteristic data based on the user's preferences and interests. For example, it might reveal that the user tends to like espresso and identify if their current statement contains excitement. This characteristic data is managed in a way that does not identify individuals.

[1059] Input: Text data, sentiment data, historical interaction data

[1060] Output: Feature data

[1061] Step 6:

[1062] Response generation by generative AI

[1063] The server uses a generative AI model to generate personalized responses based on the generated feature data and sentiment data. The generative AI utilizes the user's feature data and sentiment data to create appropriate responses tailored to each individual, such as, "You've always liked espresso, so we recommend the latest espresso maker A. It's on sale right now, so please take this opportunity to try it."

[1064] Input: Feature data, sentiment data

[1065] Output: Personalized response

[1066] Step 7:

[1067] Providing a response

[1068] The device receives personalized responses generated by a generative AI model and provides them to the user in voice or text format. For example, it might respond, "Nearby, we have coffee maker A, which is perfect for espresso, and it's currently on sale."

[1069] Input: Personalized response

[1070] Output: Voice or text response

[1071] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1072] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1073] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1074] [Fourth Embodiment]

[1075] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1076] As shown in Figure 7, the 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.

[1077] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1078] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1079] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1080] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1081] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1082] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1083] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1084] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[1085] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1086] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1087] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1088] The system for carrying out the present invention has the function of acquiring user voice data, analyzing it, and providing a personalized response. The system mainly includes a terminal, a server, and a generative AI.

[1089] Acquisition and conversion of audio data

[1090] The user speaks a question or command into their smartphone. For example, they might say, "What's the weather like today?" This voice data is captured by the device and sent to a server in real time. The server passes the received voice data to a natural language processing (NLP) module, which converts the voice data into text data. The converted text data will be in the form of "What's the weather like today?"

[1091] Acquisition of customer data

[1092] The server uses the user's identification information to retrieve the user's past interaction data from relevant customer databases (e.g., databases of telecommunications companies or related services). This data includes the applications the user has used and their past search history. For example, it might be confirmed that the user frequently uses a particular weather app.

[1093] Data analysis and feature data generation

[1094] The server analyzes text data and the user's past interaction data to extract user preferences and interests. This generates characteristic data about the user. For example, it might be identified that the user tends to go out on sunny days. This characteristic data is managed in a way that does not identify individuals.

[1095] Response generation by generative AI

[1096] The server trains a generative AI model based on the generated feature data. The generative AI uses the user's feature data to generate personalized responses. For example, it creates responses that reflect the user's interests, such as, "It's sunny today. It might get a little cloudy in the afternoon, but you can also check with your favorite weather app."

[1097] Providing a response

[1098] The device receives personalized responses generated by generative AI models and provides them to the user in voice or text format. For example, if a user asks about a specific cafe, the system will recommend cafes that suit the user's preferences and provide details about their locations.

[1099] Specific example

[1100] Situation

[1101] The following outlines the process when a user uses their smartphone to ask for the location of a cafe.

[1102] 1. User: "Can you recommend a nearby cafe?"

[1103] 2. The device captures the audio and sends the data to the server.

[1104] 3. The server converts the voice data into text. The result of the conversion is, "Please tell me about a nearby cafe."

[1105] 4. The server uses the identification information to retrieve the user's past interaction data from the relevant customer database. For example, it might be confirmed that the user prefers quiet cafes.

[1106] 5. The server analyzes the data and generates feature data based on user preferences.

[1107] 6. The server uses a generative AI model to generate personalized responses that reflect the user's characteristics.

[1108] 7. The terminal provides the user with a generated response. For example, it might respond, "There's a quiet cafe nearby that you might like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[1109] As described above, this invention can provide users with optimized information by personalizing responses based on the user's utterances and past data.

[1110] The following describes the processing flow.

[1111] Step 1:

[1112] The user speaks to their smartphone. For example, they might say, "What's the weather like today?"

[1113] Step 2:

[1114] The device captures the user's speech using its microphone and sends that audio data to the server in real time.

[1115] Step 3:

[1116] The server receives the audio data and passes it to a natural language processing (NLP) module, which converts the audio data into text data. The result of this conversion is "What's the weather like today?"

[1117] Step 4:

[1118] The server uses the user's identification information to retrieve the user's past interaction data from databases of software providers and integrated services. This data includes the applications the user uses and their past search history.

[1119] Step 5:

[1120] The server analyzes text data and acquired interaction data using machine learning algorithms to generate feature data based on user preferences and interests. For example, it might identify that a user prefers to use a particular weather app or tends to go out on sunny days.

[1121] Step 6:

[1122] The server trains a generative AI model based on the analyzed feature data. The generative AI is then adapted to generate personalized responses using the user's feature data.

[1123] Step 7:

[1124] The device generates personalized responses through a generative AI model. These responses reflect the user's preferences and past behavior. For example, they might say, "It's sunny today. It might get a little cloudy this afternoon."

[1125] Step 8:

[1126] The device provides the user with a personalized response via voice or text. The user can then review the response and decide on their next course of action. In this way, the system provides the user with optimized information.

[1127] (Example 1)

[1128] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1129] Conventional voice assistant systems have struggled to generate personalized responses by fully utilizing user preferences and past interaction data. Furthermore, in order to provide information best suited to each individual user, there has been insufficient management of user-specific data and adequate measures to protect personal information. This invention aims to solve these problems and provide users with appropriate and personalized information quickly.

[1130] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1131] In this invention, the server includes means for acquiring user voice data, means for converting voice data into text data, means for acquiring the user's past interaction data, means for analyzing and generating feature data, means for training a generative AI model, means for generating personalized responses, and means for providing responses to the user. This makes it possible to provide personalized responses quickly and accurately based on the user's preferences and interests.

[1132] A "user" is someone who uses the system to input voice data and receive a personalized response.

[1133] "Voice data" refers to the audio recorded when a user speaks into their smartphone or other device.

[1134] "Text data" refers to data obtained by converting audio data into text format.

[1135] "Interaction data" refers to a record of a user's past interactions with the system.

[1136] "Feature data" refers to data that indicates characteristics such as user preferences and interests.

[1137] A "generative AI model" is an artificial intelligence model designed to generate personalized responses based on user characteristic data.

[1138] "Means" refers to the devices or methods used to achieve an objective.

[1139] A "server" is a central processing unit that performs audio data processing, text conversion, and acquisition and analysis of interaction data.

[1140] A "terminal" is a device that a user uses to input voice data and receive a personalized response.

[1141] A "personalized response" is a customized reply generated based on the user's characteristic data.

[1142] The system for carrying out the present invention acquires user voice data, analyzes it, and provides a personalized response. The system components include a terminal, a server, and a generative AI model.

[1143] overview

[1144] The user speaks a question or command into a device such as a smartphone. For example, they might say, "What's the weather like today?" This voice data is captured by the device and sent to the server in real time. The server passes the received voice data to a natural language processing (NLP) module, which converts the voice data into text data. The converted text data will be in the form of "What's the weather like today?".

[1145] Acquisition and conversion of audio data

[1146] When a user speaks a question or command to their smartphone, the device uses its built-in microphone to capture voice data. This voice data is then transmitted to a server in real time via an internet connection. Specifically, the software used is an audio capture module, and the hardware used is the smartphone's built-in microphone.

[1147] Server-based processing

[1148] The server passes the received audio data to a natural language processing module, which uses text conversion software such as the Google Speech-to-Text API or IBM Watson to convert the audio data into text. This converted text data is then used for subsequent analysis.

[1149] User data collection and analysis

[1150] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. This data includes the applications the user has used and their past search history. For example, it might confirm that the user frequently uses a particular weather app.

[1151] The server analyzes text data and the user's past interaction data to extract user preferences and interests. This generates feature data about the user. Analysis tools such as Python's Pandas and Scikit-learn are used for the analysis. For example, if it is found that a user tends to go out on sunny days, this information is saved as feature data.

[1152] Response generation using generative AI models

[1153] The server trains a generative AI model based on the generated feature data. Examples of generative AI models used include GPT-3 and BERT. The generative AI model uses the user's feature data to generate personalized responses. For example, a possible response might be, "It's sunny today. It might get a little cloudy this afternoon, but you can check this with your usual weather app."

[1154] Providing a response

[1155] Finally, the server sends the generated personalized response to the device. The device then provides this response to the user in either voice or text format. In the voice format, a Text-to-Speech (TTS) engine is used, and the response is delivered via the smartphone's notification system, for example. For instance, the user's screen might display a message such as, "It's sunny today. It might get a little cloudy in the afternoon, but you can also check your favorite weather app."

[1156] Examples of specific prompt messages

[1157] The following are examples of prompt messages in specific situations.

[1158] User: "Can you recommend a nearby cafe?"

[1159] Response: "There's a quiet cafe nearby that you might like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[1160] In this way, the present invention can provide personalized responses based on the user's voice input and analysis of past data.

[1161] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1162] Step 1:

[1163] The user speaks questions or commands into a device such as a smartphone. For example, they might say, "What's the weather like today?" The input is the user's voice data, and the output is the voice signal captured by the device's microphone. The device converts this voice into digital data and sends it to the server using a communication module.

[1164] Step 2:

[1165] The server receives audio data sent from the terminal. The input is digitized audio data, and the output is audio data passed to the NLP module. The server passes this data to the Natural Language Processing (NLP) module, which converts the audio data into text data. Software used includes Google Speech-to-Text API and IBM Watson.

[1166] Step 3:

[1167] The server parses the converted text data. The input is text data ("What's the weather like today?"), and the output is a query based on what the user asked. This query is stored for use in the next processing step.

[1168] Step 4:

[1169] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. The input is the user's identification information (user ID and registered email address), and the output is the user's past interaction data, which includes the applications the user used and their search history.

[1170] Step 5:

[1171] The server analyzes text data and the user's past interaction data. The input consists of text data and interaction data, while the output is feature data indicating the user's preferences and interests. This feature data includes information such as the user's tendency to go out on sunny days. Python libraries such as Pandas and Scikit-learn are used for the analysis.

[1172] Step 6:

[1173] The server trains a generative AI model based on the generated feature data. The input is the feature data, and the output is the trained generative AI model. Examples of generative AI models used include GPT-3 and BERT.

[1174] Step 7:

[1175] The server uses a generative AI model to generate personalized responses for the user. The input is a trained generative AI model and user feature data, and the output is a personalized response. For example, a response like, "It's sunny today. It might get a little cloudy in the afternoon, but you can also check with your favorite weather app," might be generated.

[1176] Step 8:

[1177] The server sends the generated personalized response to the terminal. The input is the personalized response, and the output is the response data sent to the terminal.

[1178] Step 9:

[1179] The device provides the user with the received response in either voice or text format. The input is the response data sent from the server, and the output is the personalized response presented to the user. In voice format, a Text-to-Speech (TTS) engine is used, and the response is delivered via the smartphone's notification system, for example. For instance, the user's screen might display a message such as, "It's sunny today. It might get a little cloudy in the afternoon, but you can also check with your favorite weather app."

[1180] In this way, the system analyzes the user's voice data and provides a response optimized for the user.

[1181] (Application Example 1)

[1182] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1183] Modern food delivery services are expected to rely heavily on voice commands for ordering and recommendations. However, current systems struggle to provide personalized recommendations based on users' past preferences and interests, and voice recognition for ordering also has accuracy issues. As a result, the user experience is not fully optimized.

[1184] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1185] In this invention, the server includes means for acquiring voice data, means for converting voice data into text data, means for acquiring the user's past interaction data, means for analyzing the text data and interaction data to generate feature data, means for generating personalized responses using generative AI, means for providing the generated responses, means for recommending restaurants and dishes, and means for placing food delivery orders using voice recognition. This enables the provision of restaurant and dish recommendations based on the user's preferences and interests, and further enables highly accurate food delivery orders using voice recognition.

[1186] "Audio data" refers to the voice information spoken by the user, and the captured acoustic signal.

[1187] "Text data" refers to information in string format generated by analyzing audio data.

[1188] "Interaction data" refers to information such as the actions, orders, and search history that a user has performed in the past.

[1189] "Feature data" refers to information about user preferences and interests, generated based on text data and interaction data.

[1190] "Generative AI" refers to artificial intelligence technology that receives feature data as input and generates personalized responses.

[1191] "Response" refers to the answers and recommendation information provided to the user by a generative AI.

[1192] A "server" refers to a computer system that is responsible for processing tasks such as analyzing audio data, acquiring interaction data, and executing generative AI.

[1193] "Restaurant and food recommendations" refers to a function that suggests appropriate restaurants and dishes based on the user's preferences and interests.

[1194] "Speech recognition" refers to the technology that analyzes speech data and converts it into text data.

[1195] "Food delivery" refers to a service that delivers food and beverages to the user's location.

[1196] This invention is a system that acquires user voice data, analyzes it, and provides personalized responses. This system mainly consists of a smartphone, a server, and a generative AI.

[1197] Acquisition of audio data

[1198] When a user speaks a question or command into their smartphone, voice data is captured. For example, they might say, "What restaurant would you recommend for dinner tonight?" This voice data is captured by the device and sent to the server in real time.

[1199] Audio data conversion

[1200] The server passes the received audio data to a natural language processing (NLP) module, which converts the audio data into text data. The converted text data will be in the form of "What restaurant would you recommend for dinner tonight?"

[1201] Acquisition of interaction data

[1202] The server uses the user's identification information to retrieve the user's past interaction data from relevant databases. This data includes the applications the user has used and their past order history. For example, it might confirm that the user has a preference for ordering Japanese food in the past.

[1203] Data analysis and feature data generation

[1204] The server analyzes text data and the user's past interaction data to extract user preferences and interests. This generates characteristic data about the user. For example, it might be identified that the user tends to prefer Japanese food. This characteristic data is managed in a way that does not identify individuals.

[1205] Response generation by generative AI

[1206] The server trains a generative AI model based on the generated feature data. The generative AI uses the user's feature data to generate personalized responses. For example, it might create a response that reflects the user's interests, such as, "There's a Japanese restaurant called XYZ nearby. It has good reviews, and is especially known for its delicious sashimi."

[1207] Providing a response

[1208] The device receives personalized responses generated by generative AI models and provides them to the user in voice or text format. For example, if a user asks, "What restaurants would you recommend for dinner tonight?", the system will recommend a Japanese restaurant based on the user's preferences and provide details about its location.

[1209] Specific example

[1210] The following outlines the process when a user uses their smartphone to request restaurant recommendations.

[1211] 1. User: "What restaurants would you recommend for dinner tonight?"

[1212] 2. The device captures the audio and sends the data to the server.

[1213] 3. The server converts the audio data into text. The result of the conversion will be, "What restaurant would you recommend for dinner tonight?"

[1214] 4. The server uses the identification information to retrieve the user's past interaction data from the relevant database. For example, it might confirm that the user prefers Japanese food.

[1215] 5. The server analyzes the data and generates feature data based on user preferences.

[1216] 6. The server uses a generative AI model to generate personalized responses that reflect the user's characteristics. For example, it might respond, "There's a Japanese restaurant called XYZ nearby. It has good reviews, especially for its sashimi."

[1217] 7. The terminal provides the generated response to the user.

[1218] Example of a prompt:

[1219] The user likes Japanese food and is looking for highly-rated restaurants. Text: Tell me about nearby Japanese restaurants. Please recommend restaurants that reflect these criteria.

[1220] In this way, the system can analyze the user's voice and recommend the most suitable restaurants and dishes based on the user's preferences. This improves the user's food delivery experience.

[1221] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1222] Step 1:

[1223] The user verbally asks a question into their smartphone. For example, they might ask, "What restaurants would you recommend for dinner tonight?" This audio data is captured by the device. The input data is an audio signal, and the output is audio data.

[1224] Step 2:

[1225] The device sends the captured audio data to the server. Since the transmitted audio data cannot be analyzed directly, speech recognition is required.

[1226] Step 3:

[1227] The server uses a speech recognition module to convert speech data into text data. This process is performed using natural language processing (NLP) techniques. The input data is speech data, and the output is text data such as "What restaurant would you recommend for dinner tonight?"

[1228] Step 4:

[1229] The server uses the user's identification information to retrieve the user's past interaction data from the relevant database. This data includes past order history and search history. The input data is the user's identification information, and the output is interaction data.

[1230] Step 5:

[1231] The server analyzes text data and the user's past interaction data to generate feature data. This process involves data analysis to extract the user's preferences and interests. The input data consists of text data and interaction data, and the output is feature data that indicates the user's preferences and interests.

[1232] Step 6:

[1233] The server trains a generative AI model using the generated feature data. The generative AI model is used to generate personalized responses based on the feature data. The input data is the feature data, and the output is a personalized response based on the generated prompt sentences.

[1234] Step 7:

[1235] The server sends the generated personalized response to the terminal. This response is then ready to be provided to the user. The input data is the generated response, and the output is the response data directed to the terminal.

[1236] Step 8:

[1237] The terminal provides the user with personalized responses received from the server. These responses are delivered in either voice or text format. Input data is the response data, and output is the information provided to the user. For example, the user might be notified with something like, "There's a Japanese restaurant called XYZ nearby. It has high ratings in online reviews, and is especially known for its delicious sashimi."

[1238] Example of a prompt:

[1239] The user likes Japanese food and is looking for highly-rated restaurants. Text: Tell me about nearby Japanese restaurants. Please recommend restaurants that reflect these criteria.

[1240] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1241] The system for carrying out the present invention has the function of acquiring user voice data, analyzing it, and providing a personalized response. The system mainly includes a terminal, a server, a generative AI, and an emotion engine.

[1242] Acquisition and conversion of audio data

[1243] The user speaks a question or command into their smartphone. For example, they might say, "What's the weather like today?" This voice data is captured by the device and sent to a server in real time. The server passes the received voice data to a natural language processing (NLP) module, which converts the voice data into text data. The converted text data will be in the form of "What's the weather like today?"

[1244] Analysis of emotional data

[1245] After receiving text data, the server passes it to the emotion engine. The emotion engine analyzes the user's emotions from the voice and text data and generates emotion data such as joy, surprise, anger, and sadness. For example, the statement "What's the weather like today?" might be analyzed as containing the user's anxiety or worry.

[1246] Acquisition of customer data

[1247] The server uses the user's identification information to retrieve the user's past interaction data from relevant customer databases (e.g., databases of telecommunications companies or related services). This data includes the applications the user has used and their past search history. For example, it might be confirmed that the user frequently uses a particular weather app.

[1248] Data analysis and feature data generation

[1249] The server uses machine learning algorithms to analyze text data, sentiment data, and the user's past interaction data to generate characteristic data based on the user's preferences and interests. For example, it might identify that the user tends to go out on sunny days or that their current speech contains anxiety. This characteristic data is managed in a way that does not identify individuals.

[1250] Response generation by generative AI

[1251] The server trains a generative AI model based on the generated feature and sentiment data. The generative AI uses the user's feature and sentiment data to generate personalized responses. For example, it creates responses that also take the user's emotions into account, such as, "It's sunny today. It might get a little cloudy in the afternoon, but don't worry. You can also check with your favorite weather app."

[1252] Providing a response

[1253] The device receives personalized responses generated by generative AI models and provides them to the user in voice or text format. For example, if a user asks about cafe locations, the system recommends cafes that suit the user's preferences and current mood, and provides details about the locations. If the user is tired, it might recommend a quiet cafe.

[1254] Specific example

[1255] Situation

[1256] The following outlines the process when a user uses their smartphone to ask for the location of a cafe.

[1257] 1. User: "Can you recommend a nearby cafe?"

[1258] 2. The device captures the audio and sends the data to the server.

[1259] 3. The server converts the voice data into text. The result of the conversion is, "Please tell me about a nearby cafe."

[1260] 4. The server uses the identification information to retrieve the user's past interaction data from the relevant customer database. For example, it might be confirmed that the user prefers quiet cafes.

[1261] 5. The server uses an emotion engine to generate emotion data based on the user's utterances. For example, it might analyze that the user is seeking relaxation.

[1262] 6. The server analyzes the data to generate feature data and sentiment data based on user preferences.

[1263] 7. The server uses generative AI models to generate personalized responses that reflect the user's characteristics and emotions.

[1264] 8. The terminal provides the user with a generated response. For example, it might respond, "There is a quiet, relaxing cafe nearby that you might like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[1265] The present invention, through the process described above, provides information optimized for the user by personalizing responses based on the user's speech content, past data, and even emotional data.

[1266] The following describes the processing flow.

[1267] Step 1:

[1268] The user speaks to their smartphone. For example, they might say, "What's the weather like today?"

[1269] Step 2:

[1270] The device captures the user's speech using its microphone and sends that audio data to the server in real time.

[1271] Step 3:

[1272] The server receives the audio data and passes it to a natural language processing (NLP) module, which converts the audio data into text data. The result of this conversion is "What's the weather like today?"

[1273] Step 4:

[1274] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. This data includes the applications the user has used and their past search history. For example, it might be confirmed that the user frequently uses a particular weather app.

[1275] Step 5:

[1276] The server passes text data to the emotion engine, which analyzes the user's emotions. The emotion engine identifies whether the user is feeling emotions such as joy, surprise, anger, or anxiety based on the tone and content of the voice. For example, the statement "What's the weather like today?" might be analyzed as containing anxiety.

[1277] Step 6:

[1278] The server analyzes sentiment data, text data, and acquired interaction data using machine learning algorithms to generate feature data. This feature data includes things like whether a user tends to go out on sunny days, whether they prefer a particular weather app, and their current emotional state.

[1279] Step 7:

[1280] The server trains a generative AI model based on the generated feature and sentiment data. The generative AI model is configured to generate personalized responses that reflect the user's feature and sentiment data.

[1281] Step 8:

[1282] The device uses a generative AI model to generate personalized responses. These responses take into account the user's preferences and current mood. For example, it might say something like, "It's sunny today. It might get a little cloudy in the afternoon, but don't worry. You can also check your favorite weather app."

[1283] Step 9:

[1284] The device provides the user with a generated, personalized response via voice or text. The user can then review the response and decide on their next action. For example, if the user asks about cafe locations, the system recommends cafes that suit the user's preferences and current mood, and provides details about those locations.

[1285] (Example 2)

[1286] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1287] Conventional speech recognition systems only convert user speech into text and are unable to provide responses that take into account the user's emotions or past interactions. As a result, they tend to provide generic answers without offering truly personalized and useful information to the user. Furthermore, there is a lack of means to generate and securely manage feature data based on individual user preferences and emotions. This invention aims to solve these problems.

[1288] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1289] In this invention, the server includes means for acquiring user voice data, means for converting voice data into text data, means for analyzing user emotion data, means for acquiring user past interaction data, means for generating feature data by analyzing text data, emotion data and the user's past interaction data, generative AI means for generating personalized responses based on the feature data and emotion data, and means for providing the generated personalized responses to the user. This makes it possible to provide more appropriate and personalized responses that take into account the user's emotions and past interactions.

[1290] "User" refers to an individual or group that uses this system.

[1291] "Voice data" refers to data that records voice information spoken by a user in digital format.

[1292] "Text data" refers to string information obtained by converting audio data using natural language processing technology.

[1293] "Emotional data" refers to data that indicates the emotional state of a user, analyzed and extracted from their voice and text.

[1294] "Past interaction data" refers to information such as the user's past actions, usage history, and search history.

[1295] "Feature data" refers to characteristic information obtained by analyzing a user's preferences, interests, emotions, etc.

[1296] "Generative AI methods" refer to artificial intelligence models that generate optimal responses for users based on feature data and sentiment data.

[1297] "Encryption" refers to the process of transforming information to protect it from unauthorized access.

[1298] "Analysis" refers to the process of thoroughly examining data and extracting its meaning.

[1299] A "speech recognition algorithm" refers to an algorithm that converts speech data into text data.

[1300] A "natural language processing (NLP) module" refers to a software module that implements technology to analyze audio data and convert it into corresponding text data.

[1301] A "personalized response" refers to a response created with the individual user's characteristics and emotions in mind.

[1302] The system for carrying out the present invention has the function of acquiring user voice data, analyzing it, and providing a personalized response. This system mainly includes a user, a terminal, a server, a generative AI model, and an emotion engine.

[1303] First, the user speaks a question or command into a device such as a smartphone. For example, voice input such as "What's the weather like today?" or "Can you tell me about a nearby cafe?" is performed. This voice data is captured by the device and then sent to the server in real time. The hardware used by the device includes a microphone and a communication module, and the transmitted data is encrypted to ensure reliability.

[1304] The server passes the received audio data to a natural language processing (NLP) module, which converts the audio data into text data. This conversion process uses a speech recognition algorithm, so that, for example, an audio statement like "Please tell me about a nearby cafe" is converted into similar text data.

[1305] Next, the server sends the converted text data to the emotion engine for analysis. The emotion engine analyzes the user's emotions from the audio and text data and generates emotion data such as joy, surprise, anger, and sadness. For example, it extracts emotions such as tension and relaxation from the user's tone and way of speaking.

[1306] User interaction data is also a crucial element. Based on user identification information, the server retrieves data such as past operation history, usage history, and search history from relevant databases. This allows for a detailed understanding of what the user likes and what their tendencies are.

[1307] The server combines and analyzes the obtained text data, sentiment data, and past interaction data to generate feature data. This feature data includes user preferences and interests, and machine learning algorithms are used for analysis.

[1308] Based on the generated feature and sentiment data, the server uses a generative AI model to create personalized responses. This allows for the creation of appropriate responses tailored to the user's specific situation and emotions. For example, it could respond with something like, "There's a quiet, relaxing cafe nearby that you'll like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[1309] Ultimately, the generated personalized response is delivered to the user via the device. The response is communicated to the user in voice or text format, and relevant information is specifically presented.

[1310] Specific example

[1311] For example, consider the following scenario where a user asks their smartphone, "Please tell me about nearby cafes."

[1312] 1. The user asks, "Can you tell me about a nearby cafe?"

[1313] 2. The device captures the audio and sends the data to the server.

[1314] 3. The server converts the voice data into text. The result of the conversion is "Please tell me about a nearby cafe."

[1315] 4. The server uses an emotion engine to generate emotion data based on the user's utterances. For example, it might analyze that the user is seeking relaxation.

[1316] 5. The server retrieves the user's past interaction data from the relevant database. For example, it might confirm that the user prefers quiet cafes.

[1317] 6. The server analyzes text data, sentiment data, and past interaction data to generate feature data.

[1318] 7. The server uses a generative AI model to generate personalized responses that reflect the user's characteristics and emotions.

[1319] 8. The terminal provides the user with a generated response. It responds, "There is a quiet, relaxing cafe nearby that you might like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[1320] In this way, the system can personalize its responses based on the user's utterances, past data, and even emotional data, providing information optimized for the user.

[1321] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1322] Step 1: User voice input

[1323] The user speaks into their smartphone. For example, they might say, "Please tell me about a nearby cafe." The device uses its microphone to capture the audio.

[1324] Input: User's voice data

[1325] Output: Captured audio data

[1326] Step 2: Sending the audio data

[1327] The device transmits the captured audio data to the server in real time. During this process, the data is encrypted and securely transferred over the internet.

[1328] Input: Captured audio data

[1329] Output: Audio data sent to the server

[1330] Step 3: Convert speech to text

[1331] The server passes the received audio data to a natural language processing (NLP) module, which converts the audio data into text data. A speech recognition algorithm is used for this process. For example, the audio "Please tell me about a nearby cafe" is converted into text.

[1332] Input: Sent audio data

[1333] Output: Text data "Please tell me about nearby cafes."

[1334] Step 4: Analysis of emotional data

[1335] The server passes the converted text data to the emotion engine. The emotion engine analyzes the user's emotions from the audio and text data and generates emotion data. For example, it might analyze that the user is seeking relaxation.

[1336] Input: Converted text data

[1337] Output: User emotion data (e.g., relaxed)

[1338] Step 5: Obtain customer data

[1339] Based on the user's identification information, the server retrieves customer data, such as past operation history and usage history, from related databases. This includes the user's frequently searched topics and application usage history.

[1340] Input: User identification information

[1341] Output: User's past interaction data

[1342] Step 6: Data analysis and feature data generation

[1343] The server performs analysis based on the obtained text data, sentiment data, and past interaction data. Using machine learning algorithms, it generates feature data representing the user's preferences and interests. For example, the feature that the user prefers quiet cafes might be extracted.

[1344] Input: Text data, sentiment data, past interaction data

[1345] Output: User characteristic data

[1346] Step 7: Generating responses using generative AI

[1347] The server operates a generative AI model based on the generated feature and sentiment data. The generative AI model generates personalized responses tailored to the user's characteristics. For example, it might generate a response like, "There's a quiet, relaxing cafe nearby that you'll like. The highest-rated one is 'Cafe XYZ,' which is known for its delicious coffee."

[1348] Input: Feature data, sentiment data

[1349] Output: Personalized response

[1350] Step 8: Provide a response

[1351] The device receives personalized responses sent from the server and provides them to the user in voice or text format. Users can receive these responses through their smartphone screen or voice assistant.

[1352] Input: Personalized response

[1353] Output: Provides a response to the user.

[1354] (Application Example 2)

[1355] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1356] In today's shopping environment, personalized services that cater to the diverse needs and emotions of consumers are in demand. However, conventional systems struggle to effectively utilize user voice and emotional data, combining them with past interaction history to provide optimal responses. This makes it difficult to deliver a truly satisfying shopping experience. Furthermore, the lack of technology to generate personalized responses in real time is also a challenge.

[1357] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1358] In this invention, the server includes means for acquiring user voice data, means for converting voice data into text data, and means for acquiring the user's past interaction data. This makes it possible to analyze user voice-based questions and commands in a shopping environment in real time, and to provide personalized responses to the user by combining past interaction data and emotion data. Furthermore, by using generative AI means, it is possible to realize the functionality of an optimal shopping assistant that reflects the user's current emotions and past purchase history.

[1359] "User voice data" refers to audio signals that include words and commands spoken by the user.

[1360] "Text data" refers to data obtained by converting audio data into text or written words.

[1361] "User's past interaction data" refers to data related to actions and activities the user has performed in the past, including, for example, past purchase history and logs of applications used.

[1362] "Feature data" refers to data that indicates user preferences and emotions, extracted by analyzing text data and past interaction data.

[1363] "Emotional data" refers to data that indicates the emotional state of a user, analyzed from their speech content and tone of voice, and includes emotions such as joy, surprise, anger, sadness, and anxiety.

[1364] "Generative AI methods" are means of generating personalized responses based on user characteristic data and sentiment data, using machine learning algorithms and generative models.

[1365] A "personalized response" refers to an optimized, individualized answer or recommendation that takes into account each user's preferences and emotions.

[1366] "Means of delivery" refers to the means of communicating the generated personalized response to the user in the form of voice or text.

[1367] The system for carrying out the present invention has the function of acquiring user voice data, analyzing it, and providing a personalized response. The system mainly includes a terminal, a server, a generative AI, and an emotion engine. The configuration and processing of each are described in detail below.

[1368] Acquisition and conversion of audio data

[1369] The user speaks a question or command into their smartphone or smart glasses. For example, they might say, "Which coffee maker do you recommend nearby?" This voice data is captured by the device and sent to a server in real time. The server passes the received voice data to a natural language processing (NLP) module, which converts the voice data into text data. The converted text data will be in the form of "Which coffee maker do you recommend nearby?"

[1370] Analysis of emotional data

[1371] After the server receives the text data, it passes it to the emotion engine. The emotion engine analyzes the user's emotions from the voice and text data and generates emotion data such as joy, relief, surprise, and excitement. For example, if the user says, "What's a good coffee maker nearby?", the emotion engine might analyze that the user is excited.

[1372] Acquisition of customer data

[1373] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. This data includes items the user has previously purchased and services they have used. For example, a record of the user having previously purchased a coffee maker may be found.

[1374] Data analysis and feature data generation

[1375] The server uses machine learning algorithms to analyze text data, sentiment data, and the user's past interaction data to generate characteristic data based on the user's preferences and interests. For example, it might identify that the user has a history of liking espresso or that their current statement contains excitement. This characteristic data is managed in a way that does not identify individuals.

[1376] Response generation by generative AI

[1377] The server generates personalized responses using a generative AI model based on the generated feature and sentiment data. The generative AI uses the user's feature and sentiment data to create an optimal, individually tailored response. For example, it might generate a response that takes the user's emotions into account, such as, "You've always liked espresso, so we recommend the latest espresso maker A. It's on sale right now, so please take this opportunity to try it."

[1378] Providing a response

[1379] The device receives personalized responses generated by a generative AI model and provides them to the user in voice or text format. For example, it might respond, "Nearby, we have coffee maker A, which is perfect for espresso, and it's currently on sale."

[1380] Explanation of specific examples

[1381] For example, if a user asks, "Which coffee maker do you recommend near me?", the system converts the voice into text, performs sentiment analysis, and then, referencing past purchase history and preferences, generates a response such as, "Since you've always liked espresso, we recommend the latest espresso maker A. It's currently on sale, so please take this opportunity to try it."

[1382] Example of a prompt

[1383] User comment: "What coffee maker would you recommend near me?"

[1384] Emotion: {"neg": 0.0, "neu": 0.5, "pos": 0.5, "compound": 0.5}

[1385] Customer information: {"name": "User", "preferences": "Espresso", "history": ["Coffee beans", "Coffee machine"]}

[1386] response:"

[1387] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1388] Step 1:

[1389] Acquisition of user voice data

[1390] The user speaks questions or commands into their smartphone or smart glasses. For example, they might say, "Which coffee maker do you recommend nearby?" This voice data is captured by the device and sent to the server in real time.

[1391] Input: Audio data

[1392] Output: Audio data

[1393] Step 2:

[1394] Converting audio data to text

[1395] The server passes the received audio data to a natural language processing (NLP) module, which converts the audio data into text data. The converted text data will be in the form of "What is the best coffee maker nearby?"

[1396] Input: Audio data

[1397] Output: Text data

[1398] Step 3:

[1399] Analysis of emotional data

[1400] After the server receives the text data, it passes it to the emotion engine. The emotion engine analyzes the user's emotions from the voice and text data and generates emotion data such as joy, relief, surprise, and excitement. For example, if the user says, "What's a good coffee maker nearby?", the emotion engine might analyze that the user is excited.

[1401] Input: Audio data, text data

[1402] Output: Sentiment data

[1403] Step 4:

[1404] Acquisition of customer data

[1405] The server uses the user's identification information to retrieve the user's past interaction data from the relevant customer database. This data includes items the user has previously purchased and services they have used. For example, a record of the user having previously purchased a coffee maker may be found.

[1406] Input: User identification information

[1407] Output: Past interaction data

[1408] Step 5:

[1409] Data analysis and feature data generation

[1410] The server uses machine learning algorithms to analyze text data, sentiment data, and past user interaction data to generate characteristic data based on the user's preferences and interests. For example, it might reveal that the user tends to like espresso and identify if their current statement contains excitement. This characteristic data is managed in a way that does not identify individuals.

[1411] Input: Text data, sentiment data, historical interaction data

[1412] Output: Feature data

[1413] Step 6:

[1414] Response generation by generative AI

[1415] The server uses a generative AI model to generate personalized responses based on the generated feature data and sentiment data. The generative AI utilizes the user's feature data and sentiment data to create appropriate responses tailored to each individual, such as, "You've always liked espresso, so we recommend the latest espresso maker A. It's on sale right now, so please take this opportunity to try it."

[1416] Input: Feature data, sentiment data

[1417] Output: Personalized response

[1418] Step 7:

[1419] Providing a response

[1420] The device receives personalized responses generated by a generative AI model and provides them to the user in voice or text format. For example, it might respond, "Nearby, we have coffee maker A, which is perfect for espresso, and it's currently on sale."

[1421] Input: Personalized response

[1422] Output: Voice or text response

[1423] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1424] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1425] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1426] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1427] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1428] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1429] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1430] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1431] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1432] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1433] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1434] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1435] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1436] 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.

[1437] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1438] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1439] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1440] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1441] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1442] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1443] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[1444] The following is further disclosed regarding the embodiments described above.

[1445] (Claim 1)

[1446] A means of acquiring user voice data,

[1447] A means for converting the above audio data into text data,

[1448] A means of obtaining the user's past interaction data,

[1449] A means for generating feature data by analyzing the above text data and the user's past interaction data,

[1450] A generative AI means that generates personalized responses based on the above feature data,

[1451] A means for providing the user with a personalized response generated by the above-mentioned generative AI means,

[1452] A system that includes this.

[1453] (Claim 2)

[1454] The system according to claim 1, wherein the above-mentioned feature data includes information about the user's preferences and interests.

[1455] (Claim 3)

[1456] The system according to claim 1, which includes means for managing characteristic data in a manner that does not identify individuals.

[1457] "Example 1"

[1458] (Claim 1)

[1459] A means of acquiring user voice data,

[1460] A means for converting the above audio data into text data,

[1461] A means of obtaining the user's past interaction data,

[1462] A means for generating feature data by analyzing the above text data and the user's past interaction data,

[1463] A method for training a generative AI model based on the above feature data,

[1464] A means for generating personalized responses using the above-mentioned generative AI model,

[1465] Means for providing the user with the generated personalized response,

[1466] A system that includes this.

[1467] (Claim 2)

[1468] The system according to claim 1, wherein the above-mentioned feature data includes information about the user's preferences and interests.

[1469] (Claim 3)

[1470] The system according to claim 1, which includes means for managing characteristic data in a manner that does not identify individuals.

[1471] "Application Example 1"

[1472] (Claim 1)

[1473] A means of acquiring user voice data,

[1474] A means for converting the above audio data into text data,

[1475] A means of obtaining the user's past interaction data,

[1476] A means for generating feature data by analyzing the above text data and the user's past interaction data,

[1477] A generative AI means that generates personalized responses based on the above feature data,

[1478] A means for providing the user with a personalized response generated by the above-mentioned generative AI means,

[1479] Applications for recommending restaurants and dishes,

[1480] A method for ordering food delivery using voice recognition,

[1481] A system that includes this.

[1482] (Claim 2)

[1483] The system according to claim 1, wherein the above feature data includes information about the user's preferences and interests, and includes restaurant and food recommendation information.

[1484] (Claim 3)

[1485] The system according to claim 1, which includes means for managing characteristic data in a manner that does not identify individuals, and which includes food preference data.

[1486] "Example 2 of combining an emotion engine"

[1487] (Claim 1)

[1488] A means of acquiring user voice data,

[1489] A means for converting the above audio data into text data,

[1490] A means of analyzing user sentiment data,

[1491] A means of obtaining the user's past interaction data,

[1492] A means for generating feature data by analyzing the above text data, sentiment data, and past user interaction data,

[1493] A generative AI means for generating personalized responses based on the above feature data and emotion data,

[1494] A means for providing the user with a personalized response generated by the above-mentioned generative AI means,

[1495] A system that includes this.

[1496] (Claim 2)

[1497] The system according to claim 1, wherein the above-mentioned feature data includes information about the user's preferences and interests.

[1498] (Claim 3)

[1499] The system according to claim 1, which includes means for managing characteristic data in a manner that does not identify individuals.

[1500] "Application example 2 of combining emotional engines"

[1501] (Claim 1)

[1502] A means of acquiring user voice data,

[1503] A means for converting the above audio data into text data,

[1504] A means of obtaining the user's past interaction data,

[1505] A means for generating feature data by analyzing the above text data and the user's past interaction data,

[1506] A generative AI means for generating personalized responses based on the above feature data and emotion data,

[1507] Means for providing the user with a generated personalized response,

[1508] A system that includes this.

[1509] (Claim 2)

[1510] The system according to claim 1, wherein the above-mentioned feature data includes information about the user's preferences and interests.

[1511] (Claim 3)

[1512] The system according to claim 1, comprising means for analyzing the above-mentioned emotional data. [Explanation of symbols]

[1513] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of acquiring user voice data, A means for converting the above audio data into text data, A means of obtaining the user's past interaction data, A means for generating feature data by analyzing the above text data and the user's past interaction data, A generative AI means that generates personalized responses based on the above feature data, A means for providing the user with a personalized response generated by the above-mentioned generative AI means, A system that includes this.

2. The system according to claim 1, wherein the above-mentioned feature data includes information about the user's preferences and interests.

3. The system according to claim 1, which includes means for managing characteristic data in a manner that does not identify individuals.

Images