system

Abstract

Provide a system. [Solution] means of collecting testimonies and documentary data about specific historical events; A means of preprocessing the collected data to remove noise and standardize the format; a means of tagging and adding metadata to the preprocessed data; A means to train generative AI using tagged data; and a means for users to request specific historical stories; A means for generative AI to generate stories based on user requests; a means for providing the generated story to a user; A system including:

Description

[Technical Field]

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

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

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

[0004] In recent years, it has become increasingly difficult to convey to future generations the raw voices and emotions of people who experienced historically significant events such as the Great Kanto Earthquake, the end of the war, and airplane crashes. Accordingly, there is a demand for a means to convey to future generations the real experiences and emotions that cannot be captured by mere documents or records. However, there is no efficient method for collecting, organizing, and reproducing these valuable testimonies and episodes. Therefore, the objective of this invention is to provide an efficient and effective system for realistically recreating historical events and passing them on to future generations. [Means for solving the problem]

[0005] The present invention is a system that includes a means for collecting testimonies and documentary data related to specific historical events, a means for preprocessing the collected data to remove noise and standardize the format, a means for tagging the preprocessed data to add metadata, a means for a generative AI to learn using the tagged data, a means for a user to request a specific historical story, a means for the generative AI to generate a story based on the user request, and a means for providing the generated story to the user. This makes it possible to provide users with content that realistically reproduces their desired historical experiences and emotions, thereby effectively passing on historical events to future generations.

[0006] A "historical event" is an important event or group of events that occurred in the past and had a major social, cultural, and economic impact.

[0007] "Testimony" refers to the words or writings of those involved or witnesses regarding an individual incident or experience, and is information that serves to convey facts and emotions.

[0008] "Document data" refers to information stored in written form, such as books, magazines, reports, and electronic documents, and includes records of historical events.

[0009] "Collection methods" is a general term for methods, techniques, and devices used to locate and collect specific information or data.

[0010] "Preprocessing" refers to a series of processing steps carried out to prepare data for analysis and use, and mainly involves removing noise and standardizing data formats.

[0011] "Noise removal" is the process of removing unnecessary or erroneous information, leaving only accurate and useful data.

[0012] "Unification of format" refers to aligning data that exists in different forms and formats into a consistent, standard format.

[0013] "Tagging" is a technique for assigning specific attributes or keywords to data, making it easier to identify the characteristics and content of the data.

[0014] "Metadata" is data that contains information about the data itself, and supplements the content and structure of the data through tagging.

[0015] "Generative AI" is an artificial intelligence technology that generates new text and content based on given data.

[0016] "Training" is the process of feeding data into an AI model and making it perform better through repeated training.

[0017] A "user request" is a request or input from a user to a system for specific information or services.

[0018] "Generating a story" means that AI creates a new narrative or explanation based on provided data or user requests.

[0019] "Means of providing" refers to the methods and technologies by which the system transmits the content generated by the system to the user and enables viewing and listening. [Brief explanation of the drawings]

[0020] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5]FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0021] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0022] First, the terms used in the following description will be explained.

[0023] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0024] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0025] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0026] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0027] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0028] [First embodiment]

[0029] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0030] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0031] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0032] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0033] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0034] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0035] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0036] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0037] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0038] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0039] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0040] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0041] This invention is a storytelling system that collects testimonies and documentary data about historical events and trains generative AI to convey real experiences and emotions to the next generation. The system's program provides users with a real historical experience through a series of processes: data collection, preprocessing, tagging, AI learning, request processing, content generation, and provision.

[0042] System action

[0043] 1. Data Collection

[0044] The server collects testimonies and documentary data related to historical events from various data sources, including public digital archives, university and research databases, public information on the Internet, etc. For example, the server collects audio files of interviews, written transcripts, video content, etc.

[0045] 2. Data Preprocessing

[0046] The server preprocesses the collected data. Specifically, it transcribes the audio data, removes noise from the text data, and standardizes notation. For example, it transcribes an audio interview about the Great Kanto Earthquake and corrects noisy parts and typos.

[0047] 3. Data tagging

[0048] The server assigns tags to the preprocessed data. These tags include metadata about the content, emotions, and specific themes of the data. For example, testimonies from immediately after the end of the war might be tagged with "war," "end of the war," "hardship," and "hope."

[0049] 4. AI Learning

[0050] The server then trains a generative AI model (e.g., GPT-4 (registered trademark)) on the tagged data. The AI ​​model uses this data to learn people's emotions and experiences regarding specific historical events, improving the quality of the stories it generates.

[0051] 5. User request input

[0052] The terminal provides an interface for users to request specific historical stories. Users input the events, periods, and specific experiences they are interested in. For example, a user might input, "I would like to hear in detail about the experiences of victims of the Great Kanto Earthquake."

[0053] 6. Content Generation

[0054] The server receives user requests and issues instructions to the generative AI, which then generates a realistic story based on relevant data. This story includes realistic emotions and specific episodes based on the training data. For example, the AI ​​recreates the experiences and emotions of victims of the Great Kanto Earthquake and generates a detailed story.

[0055] 7. Content Provision

[0056] The device provides the generated content to the user. The content can be in the form of text, audio, or video, and the user can read, listen, or watch it. For example, a story recreating the experiences of victims of the Great Kanto Earthquake can be provided to the user in audio format.

[0057] This system makes it possible to effectively convey the real experiences and emotions surrounding historical events to future generations, providing a deeper understanding that cannot be captured through records or documents alone.

[0058] The processing flow will be explained below.

[0059] Step 1:

[0060] The server collects testimonies and documentary data about specific historical events, including audio interviews, written records, and video content from public digital archives, university and research databases, and publicly available online sources.

[0061] Step 2:

[0062] The server converts the collected audio data into text. Specifically, it uses voice recognition software to transcribe the interview audio and save it as text data.

[0063] Step 3:

[0064] The server preprocesses the collected text data by removing noise and unnecessary information from the text data and standardizing the format, for example, correcting typos and consolidating data that exists in multiple formats into a standard format.

[0065] Step 4:

[0066] The server then tags the preprocessed text data. Specifically, it assigns metadata tags related to the content, emotions, and specific themes of the data. For example, testimonies from immediately after the end of the war might be tagged with "war," "end of the war," "hardship," and "hope."

[0067] Step 5:

[0068] The server uses the tagged data to train a generative AI model (e.g., GPT-4) over multiple epochs to learn the content and sentiment of the data.

[0069] Step 6:

[0070] The device provides an interface that allows users to request specific historical stories, providing text boxes and options for users to input events, dates, and specific experiences that interest them.

[0071] Step 7:

[0072] The terminal sends the user's request to the server, which then transmits the information entered by the user about the specific historical event or experience.

[0073] Step 8:

[0074] The server analyzes the user request and prepares the appropriate data, specifically filtering the relevant data based on the request content and extracting the required information.

[0075] Step 9:

[0076] The server uses generative AI to generate a story based on the user's request, specifically, a story that recreates a specific historical event or experience requested based on the learned data.

[0077] Step 10:

[0078] The device provides the generated story to the user. Specifically, the device displays or plays the story in the form of text, audio, or video, making it accessible to the user. For example, the device plays a story recreating the experience of the Great Kanto Earthquake in audio format.

[0079] Example 1

[0080] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0081] Testimonies and documentary data about historical events are often fragmented, making continuous preservation and transmission difficult. Furthermore, in order to pass on this information to future generations, it is necessary to express it as a story that includes real experiences and emotions, but current technology makes this difficult. Furthermore, in order to obtain detailed information about a specific historical event that interests a user, they must search for the appropriate information from within a vast amount of data, which is time-consuming and laborious.

[0082] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0083] In this invention, the server includes means for acquiring testimonies and document data related to specific historical events, means for preprocessing the acquired data to remove noise and standardize formats, means for tagging the preprocessed data and adding metadata, means for training a generative artificial intelligence using the tagged data, means for a user to request a specific historical story, means for the generative artificial intelligence to generate a story based on the user request, and means for providing the generated story to the user. This makes it possible to efficiently collect and process information related to historical events and generate realistic stories that meet the user's interests.

[0084] "Testimony and documentary data" is a general term for records and information left by people, either verbally or in writing, about a particular historical event.

[0085] "Means of acquisition" are the techniques and methods used to collect testimonial or documentary data from a particular data source.

[0086] "Preprocessing" is a process for improving the quality of acquired data by removing noise and standardizing the format.

[0087] "Noise removal" is the process of removing unnecessary information and errors contained in data.

[0088] "Format unification" is the process of converting data notation and formats into a unified form.

[0089] "Tagging" refers to adding labels to data that indicate its content or characteristics.

[0090] "Metadata" is supplementary information that describes the content and characteristics of data.

[0091] "Generative AI" is AI that has the ability to generate new information and content based on input data.

[0092] "Means of learning" refers to methods and techniques for training generative artificial intelligence using data.

[0093] A "user request" is an action or content of a user requesting a specific historical story or information from the system.

[0094] "Means for generating a story" refers to the techniques and methods by which generative artificial intelligence creates a new story based on a user request.

[0095] "Means of providing" refers to the technology or method for delivering the generated story to the user in an appropriate format.

[0096] "Content analysis" refers to the process of understanding a user request and preparing the appropriate data in response.

[0097] This invention is a storytelling system that collects testimonies and documentary data related to historical events and trains generative AI to convey real experiences and emotions to the next generation. This system includes a series of processes centered on servers, terminals, and users.

[0098] System Overview

[0099] Data collection

[0100] The server collects testimonies and documentary data related to historical events from various sources, including public digital archives, university and research institute databases, and public information on the Internet. For example, the server uses APIs to download data from public digital archives and web scraping technology to collect public information on the Internet. This includes audio files, written records, and video content.

[0101] Data Preprocessing

[0102] The server preprocesses the collected data. Specifically, it converts the audio data into text (using a speech recognition API, e.g., Google® Cloud Speech-to-Text API), removes noise from the text data, and standardizes the format. For example, it transcribes an audio interview about the Great Kanto Earthquake, corrects typos, and standardizes the text to standard Japanese expressions. It also uses noise filtering technology.

[0103] Data Tagging

[0104] The server then tags the preprocessed data. Using natural language processing (NLP) technology, it assigns content and sentiment tags, such as "war," "end of the war," "hardship," and "hope," to the data. It also uses entity recognition technology to extract specific historical events and people's names and automatically add associated tags.

[0105] AI Learning

[0106] The server trains a generative AI model (e.g., GPT-4) using the tagged data. Using machine learning libraries such as PyTorch and TENSORFLOW (registered trademark), the data is divided into training and validation sets to train the AI ​​model. During this process, hyperparameters are adjusted and data is shuffled to create a high-quality generative model.

[0107] User request input

[0108] The device provides an interface for users to request specific historical stories. Through a web or mobile application, users can input specific events, dates, and specific experiences. An example prompt might be, "I would like to hear in detail about the experiences of victims of the Great Kanto Earthquake of 1923. Please focus in particular on the emotions and specific events that occurred at the time."

[0109] Content Generation

[0110] The server instructs the generative AI based on the user's request, generating a realistic story based on relevant data. The generated story includes emotions and specific episodes based on the training data. For example, the AI ​​might generate a detailed account of an experience such as, "On September 1, 1923, we were hit by a huge earthquake..."

[0111] Content provider

[0112] The device then provides the generated content to the user. The content can be in the form of text, audio, or video. For example, if the generated story is provided in audio format, the device uses the Google Cloud Text-to-Speech API to convert the text into audio, which the user can read, listen to, or watch.

[0113] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0114] Step 1:

[0115] Data collection

[0116] The server collects testimonies and documentary data about historical events from various sources, including public digital archives, university and research institute databases, and publicly available information on the Internet. Specifically, it uses APIs to download data from digital archives and web scraping technology to collect public information on the Internet, such as audio files, written records, and video content.

[0117] Input: Public digital archives, databases, public information on the internet

[0118] Output: Collected testimonies and literature data

[0119] Step 2:

[0120] Data Preprocessing

[0121] The server performs preprocessing on the collected data. First, it uses a speech recognition API (e.g., Google Cloud Speech-to-Text API) to convert the audio data into text. Next, it standardizes the format of the text data and removes noise. This includes correcting typos, converting to standard Japanese expressions, and removing unnecessary information. For example, it transcribes an interview about the Great Kanto Earthquake and uses noise filtering technology to improve the quality of the audio data.

[0122] Input: Collected testimonies and literature data

[0123] Output: Preprocessed character data

[0124] Step 3:

[0125] Data Tagging

[0126] The server assigns tags to the preprocessed data that indicate its content and characteristics. Using natural language processing (NLP) technology, content and sentiment tags such as "war," "end of the war," "hardship," and "hope" are added to the data. Entity recognition technology is used for tagging, extracting specific historical events and people's names and assigning relevant tags.

[0127] Input: Preprocessed character data

[0128] Output: Tagged data

[0129] Step 4:

[0130] AI Learning

[0131] The server trains a generative AI model (e.g., GPT-4) on the tagged data. It uses machine learning libraries such as PyTorch and TensorFlow to split the data into a training set and a validation set, adjusting hyperparameters and shuffling the data to build a high-quality generative model.

[0132] Input: tagged data

[0133] Output: Trained generative AI model

[0134] Step 5:

[0135] User request input

[0136] The device provides an interface for users to request specific historical stories. Users can input specific events, dates, and specific experiences through a web or mobile application. For example, a user might request, "I'd like to hear in detail about the experiences of victims of the Great Kanto Earthquake of 1923."

[0137] Input: User request

[0138] Output: Request data

[0139] Step 6:

[0140] Content Generation

[0141] The server receives a user request and issues instructions to the generative AI, which then generates a realistic story based on the relevant data. The generated story includes realistic emotions and specific episodes based on the training data. For example, a user can input a prompt such as "Recreate the experiences of victims of the Great Kanto Earthquake of 1923," and a detailed story will be generated.

[0142] Input: Request data, trained generative AI model

[0143] Output: Generated stories

[0144] Step 7:

[0145] Content provider

[0146] The device provides the generated content to the user. The content can be in the form of text, audio, or video, and the user can read, listen, or watch it. For example, if the content is in audio format, the device converts the text to audio using the Google Cloud Text-to-Speech API and provides it to the user.

[0147] Input: Generated stories

[0148] Output: The content that is served to the user

[0149] (Application example 1)

[0150] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0151] In the fields of history education and research, it is difficult to convey to future generations the deep understanding and real experiences that cannot be captured through mere documentary materials and testimonies. Furthermore, the lack of diversity in information and realistic expressions of emotions makes it difficult for learners and researchers to form concrete images. Furthermore, there is a need to provide this information in an easily accessible format.

[0152] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0153] In this invention, the server includes means for collecting testimonies and document data related to specific historical events, means for preprocessing the collected data to remove noise and standardize the format, means for tagging the preprocessed data to add metadata, means for training a generative AI using the tagged data, means for a user to request a specific historical story, means for the generative AI to generate a story based on the user request, means for providing the generated story to the user, and means for providing an API for distributing the generated story to a smartphone app. This allows users to easily access real stories related to historical events in the form of text, audio, and video through the smartphone app, which can be used to deepen their learning and research.

[0154] Below are definitions of important terms included in the claims according to application examples.

[0155] "Specific historical events" refer to historical events associated with specific years or events, such as the Great Kanto Earthquake or war.

[0156] "Testimony and documentary data" refers to people's oral and written testimonies, as well as information recorded in books, reports, etc.

[0157] "Preprocessing" refers to the process of removing noise from collected data and standardizing its format.

[0158] A "tag" refers to identifying information that is added as metadata about the content, emotion, or specific theme of data.

[0159] "Metadata" refers to additional information about the attributes and characteristics of data.

[0160] "Generative AI" refers to artificial intelligence that learns from collected data and generates new stories.

[0161] A "request" refers to a user-input request for a particular historical story.

[0162] "API" stands for Application Program Interface and refers to a mechanism that allows software to communicate with each other.

[0163] "Smartphone app" refers to application software that runs on a smartphone.

[0164] This invention is a storytelling system that collects testimonies and documentary data about specific historical events and trains a generative AI to convey real experiences and emotions to future generations. This system is implemented through the following series of processing steps.

[0165] The main components of the system are data collection, preprocessing, tagging, AI learning, user request, content generation, and content provision. These functions run on the server and on the smartphone app used by users.

[0166] Data collection

[0167] The server collects testimonies and documentary data about specific historical events from publicly available information on the Internet, digital archives, and databases of universities and research institutions. For example, the server collects audio files, written records, and video content from interviews about the Great Kanto Earthquake. This allows information from a variety of data sources to be managed in a unified manner.

[0168] Data Preprocessing

[0169] The collected data is preprocessed on the server. During this process, the audio data is transcribed, noise is removed from the text data, and the notation is standardized. Specifically, an audio interview about the "Great Kanto Earthquake" is transcribed, and noisy parts and typos are corrected.

[0170] Data Tagging

[0171] The preprocessed data is then tagged on the server. These tags include metadata about the content, emotions, and specific themes of the data. For example, testimonies from immediately after the end of the war are tagged with "war," "end of the war," "hardship," and "hope." This allows for efficient data search and management.

[0172] AI Learning

[0173] Based on the tagged data, the server trains a generative AI model (e.g., GPT-4). This AI model uses the collected data to learn people's emotions and experiences regarding specific historical events, improving the quality of the stories it generates. This allows it to provide users with realistic and emotionally rich stories.

[0174] User request input and content generation

[0175] Through a smartphone app, users have an interface that allows them to request a specific historical story, for example, by typing, "I'd like to hear in detail about the experiences of victims of the Great Kanto Earthquake." Based on this request, the server instructs a generative AI model to generate a realistic story based on relevant data.

[0176] Content provider

[0177] The generated story is provided to the user in the form of text, audio, or video. For example, the server generates an audio story recreating the experiences of victims of the Great Kanto Earthquake and delivers it to the user via a smartphone app. The generated content is also delivered to the smartphone app via an API.

[0178] Examples of concrete examples and prompts

[0179] As a specific example, if a user opens the app and requests to listen to an audio account of the Great Kanto Earthquake, the AI ​​will generate a realistic story, display it in the app, convert it into audio, and play it back.

[0180] Examples of prompts include:

[0181] "What was your experience as a victim of the Great Kanto Earthquake?"

[0182] How did life change after the war ended?

[0183] "Tell me about the hope and despair you felt immediately after the end of the war."

[0184] This allows users to gain a deeper emotional connection to historical events, making it a useful tool for education and research.

[0185] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0186] Step 1:

[0187] The server collects testimonies and documentary data related to a specific historical event. In this process, data is collected from publicly available information on the Internet, digital archives, and databases of universities and research institutions. For example, it collects audio files of interviews, written records, and video content related to the "Great Kanto Earthquake." The input is a search keyword (e.g., "Great Kanto Earthquake"), and the output is the collected data.

[0188] Step 2:

[0189] The server preprocesses the collected data. This involves transcribing the audio data, removing noise from the text data, and standardizing the notation. Specifically, it converts the audio data into text using speech recognition software (e.g., Google Speech-to-Text API) and cleans the text. The input is the collected data, and the output is the preprocessed data.

[0190] Step 3:

[0191] The server then tags the preprocessed data. During this process, it adds metadata about the content, emotions, and specific themes of the data. For example, tags such as "war," "end of the war," "hardship," and "hope" are added to testimonies from immediately after the end of the war. The input is the preprocessed data, and the output is the tagged data.

[0192] Step 4:

[0193] The server trains a generative AI model using the tagged data. During this process, the dataset is input into a generative AI (e.g., GPT-4) to train the model. This allows the AI ​​to learn people's emotions and experiences regarding specific historical events and improve its story-generating capabilities. The input is the tagged data, and the output is the trained AI model.

[0194] Step 5:

[0195] Users request a specific historical story through a smartphone app. This process involves inputting search keywords and specific inquiries into the on-screen interface. For example, they might input, "I'd like to hear in detail about the experiences of victims of the Great Kanto Earthquake." The input is the user request, and the output is a prompt to the generative AI.

[0196] Step 6:

[0197] The server issues instructions to the generative AI based on the user's request and generates a story. During this process, the AI ​​model is instructed to generate a story based on the prompt. The generated story contains realistic emotions and specific episodes based on the training data. The input is the user's request prompt, and the output is the generated story.

[0198] Step 7:

[0199] The server provides the generated story to the user using an API to distribute it to a smartphone app. In this process, content is provided in the form of text, audio, or video. For example, the generated story can be sent to the app in audio format, allowing the user to listen to it. The input is the generated story, and the output is the content provided to the user.

[0200] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0201] This invention combines a system that uses generative AI to learn testimonies and document data related to historical events, and generates and provides realistic stories based on user requests, with an emotion engine that recognizes the user's emotions and reflects them in the story generation.

[0202] System action

[0203] 1. Data Collection

[0204] The server collects testimonies and documentary data about specific historical events, including audio interviews, written records, and video content from public digital archives, university and research databases, and publicly available online sources.

[0205] 2. Data Preprocessing

[0206] The server preprocesses the collected data. Specifically, it converts the audio data into text, transcribes it, and saves it as text data. It also removes noise from the text data and standardizes its format.

[0207] 3. Data tagging

[0208] The server then tags the preprocessed data. Specifically, it assigns metadata tags related to the content, emotions, and specific themes of the data. For example, testimonies from immediately after the end of the war might be tagged with "war," "end of the war," "hardship," and "hope."

[0209] 4. AI Learning

[0210] The server uses the tagged data to train a generative AI model (e.g., GPT-4) over multiple epochs to learn the content and sentiment of the data.

[0211] 5. User request input

[0212] The device provides an interface that allows users to request specific historical stories, providing text boxes and options for users to input events, dates, and specific experiences that interest them.

[0213] 6. Incorporating an Emotional Engine

[0214] The device incorporates an emotion engine to recognize the user's emotions. Specifically, it recognizes emotions from the user's input, as well as from biometric sensors, cameras, microphones, and other sensors, based on the user's facial expressions, tone of voice, heart rate, and other information.

[0215] 7. User Request Submission

[0216] The device sends the user's request and the recognized emotion data to the server. For example, if a user inputs "I would like to hear in detail about the experiences of victims of the Great Kanto Earthquake," the emotion engine will simultaneously recognize the user's emotions, such as sadness or excitement.

[0217] 8. Request Parsing and Data Preparation

[0218] The server analyzes the user request and emotion data and prepares appropriate data, which involves filtering relevant data based on the request content and extracting necessary information taking emotion data into account.

[0219] 9. Content Generation

[0220] The server uses generative AI to generate stories based on user requests. Specifically, it generates stories that recreate specific historical events or experiences that reflect the user's emotions based on the learned data. For example, it depicts the experiences of disaster victims by emphasizing sadness and hope based on emotional data.

[0221] 10. Content Provision

[0222] The device then provides the generated story to the user. The content is provided in the form of text, audio, and video, and the user can read, listen, or watch it. For example, a story recreating the experience of the Great Kanto Earthquake is played in audio format, with narration tailored to the user's emotions.

[0223] This system not only allows users to share their real experiences and emotions about historical events with future generations, but also provides customized stories based on the user's emotions, enabling efficient and effective historical transmission that leads to deeper understanding and emotional empathy.

[0224] The processing flow will be explained below.

[0225] Step 1:

[0226] The server collects testimonies and documentary data about specific historical events, using public digital archives, university and research databases, and publicly available information on the internet to retrieve interview audio files, written transcripts, video content, and more.

[0227] Step 2:

[0228] The server converts the collected audio data into text using speech recognition software, transcribing the interview audio and saving it as text data.

[0229] Step 3:

[0230] The server preprocesses the collected text data by removing noise and standardizing notation, for example, correcting typos and consolidating data in multiple formats into a standard format.

[0231] Step 4:

[0232] The server then tags the preprocessed text data, adding metadata tags related to the content, emotions, and specific themes of the data. For example, an interview about the Great Kanto Earthquake might be tagged with words like "earthquake," "damage," "reconstruction," and "hope."

[0233] Step 5:

[0234] The server uses the tagged data to train a generative AI model (e.g., GPT-4) over multiple epochs to learn the content and sentiment of the data.

[0235] Step 6:

[0236] The device provides an interface that allows users to request specific historical stories, providing text boxes and options for users to input events, dates, and specific experiences that interest them.

[0237] Step 7:

[0238] The device incorporates an emotion engine to recognize the user's emotions. It utilizes biometric sensors, cameras, and microphones to analyze the user's facial expressions, tone of voice, heart rate, etc., and recognizes the user's emotional state in real time.

[0239] Step 8:

[0240] The device sends the user's request and the recognized emotion data to the server. For example, if the user inputs "I would like to know more about the experiences of victims of the Great Kanto Earthquake," the emotion engine will recognize the user's sadness or excitement.

[0241] Step 9:

[0242] The server analyzes the user request and emotion data and prepares appropriate data, for example, filtering relevant data based on the request content and extracting necessary information taking emotion data into consideration.

[0243] Step 10:

[0244] The server uses generative AI to generate stories based on user requests. Based on the data the server has learned, it generates stories that recreate specific historical events or experiences that reflect the user's emotions. For example, based on emotional data, it could portray the experiences of disaster victims by emphasizing sadness and hope.

[0245] Step 11:

[0246] The device then provides the generated story to the user. The content is provided in the form of text, audio, and video, and the user can read, listen, or watch it. For example, a story recreating the experience of the Great Kanto Earthquake is played in audio format, with narration tailored to the user's emotions.

[0247] Example 2

[0248] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0249] In systems that use generative AI to generate realistic stories using testimonies and document data about historical events and provide them to users, there is a demand for systems that can generate stories that reflect the user's emotions, thereby gaining deeper empathy and understanding.

[0250] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0251] In this invention, the server includes means for collecting testimonies and document data related to specific historical events, means for preprocessing the collected data to remove noise and standardize the format, means for tagging the preprocessed data and adding metadata, means for training a generative AI using the tagged data, means for a user to request a specific historical story, means for recognizing the user's emotions and acquiring the emotion data, means for the generative AI to generate a story based on the user request and the emotion data, and means for providing the generated story to the user. This enables the generation of realistic and emotionally rich stories based on the user's emotions.

[0252] A "specific historical event" refers to an important occurrence or phenomenon in a particular era or region, and has historical significance.

[0253] A "testimony" is a statement or record made by an individual about a particular event or experience, and may be recorded in the form of an audio file, written document, video, or other format.

[0254] "Documentary data" refers to written data such as books, papers, articles, and official records related to a particular historical event.

[0255] "Means of collection" refers to the means of obtaining specific information using the Internet, digital archives, databases, etc.

[0256] "Preprocessing" refers to data formatting, including removing noise from the data and standardizing the format.

[0257] "Means of adding metadata by tagging" refers to a means of adding keywords and attribute information that indicate the content and characteristics of the data to the data, making subsequent searches and analysis easier.

[0258] "Generative AI" refers to artificial intelligence technology that generates natural language based on large datasets, and includes models such as GPT-4.

[0259] "Means for a user to request a particular historical story" refers to the means by which a user inputs historical stories of interest through the interface.

[0260] "Means for recognizing the user's emotions and acquiring emotional data" refers to means for detecting emotions by analyzing the user's facial expressions, tone of voice, heart rate, etc., and acquiring that data.

[0261] "Means for generating stories" refers to a means for generating stories in text, audio, or video format using generative AI based on user requests and emotional data.

[0262] The "means for providing" refers to a means for delivering the generated story to the user in a predetermined format.

[0263] This invention is a system that uses generative AI to generate realistic stories based on testimonies and document data about historical events, and provides these to users. Furthermore, it aims to gain deeper empathy and understanding by recognizing the user's emotions and generating stories based on them.

[0264] Data collection

[0265] The server collects testimonies and documentary data about specific historical events, including interview audio files, written records, and video content, from public digital archives, university and research institution databases, and publicly available information on the Internet, and stores them in dedicated storage.

[0266] Data Preprocessing

[0267] The server preprocesses the collected data. Audio data is converted to text using ASR (automatic speech recognition) software, and written and video data is converted to text using OCR (optical character recognition) tools. This text data is then denoised and standardized, and saved as a standardized dataset.

[0268] Data Tagging

[0269] The server then assigns tags to the preprocessed data, using NLP (natural language processing) tools to extract keywords based on the content and emotions of the data, and assigns tags such as "Great Kanto Earthquake," "experiences of victims," ​​"hope," and "despair."

[0270] AI Learning

[0271] The server trains a generative AI (e.g., GPT-4) on the tagged data. The training process is divided into multiple epochs, and inputs text data and emotional information into a large database so that the AI ​​can understand and apply them to generation.

[0272] User request input

[0273] The terminal provides a user input interface. Using options and text boxes, the user can enter details of historical events or experiences that interest them. For example, they can enter text such as "I would like to hear about experiences of the Great Kanto Earthquake."

[0274] Incorporating an emotion engine

[0275] The device is equipped with a device for recognizing the user's emotions using an emotion engine. It uses biometric sensors, a camera, and a microphone to analyze the user's facial expressions, tone of voice, heart rate, etc. in real time to detect whether the user is excited or sad.

[0276] User request submission

[0277] The device sends the user's request and the recognized emotional data to the server. For example, a request such as "I would like to hear in detail about the experiences of victims of the Great Kanto Earthquake" is sent to the server along with emotional data such as "sadness" or "excitement."

[0278] Request Parsing and Data Preparation

[0279] The server analyzes the user's request and emotional data to select the appropriate data, searching for testimonies and documents related to the requested historical event and filtering the most suitable ones based on the emotional data.

[0280] Content Generation

[0281] The server uses generative AI to generate content based on the selected data. Based on the emotional data, it can generate a story detailing the experiences of disaster victims, reflecting emotions such as sadness and hope.

[0282] Content provider

[0283] The device provides the generated story to the user, who can listen to it in text, audio, and video formats. Specifically, audio content recreating the experience of the Great Kanto Earthquake is provided, with narration based on emotions.

[0284] Prompt Sentence Examples

[0285] A user can send a request to the system using a prompt such as:

[0286] "I would like to hear in detail about the experiences of the victims of the Great Kanto Earthquake. I would like to know what difficulties they faced and how they overcame them, including the emotional aspects. I would also be happy if you could touch on their feelings of hope and courage."

[0287] As a result, this system enables the generation of realistic and emotionally rich stories based on the user's emotions, providing a deep understanding and empathy for historical events.

[0288] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0289] Step 1:

[0290] Data collection

[0291] The server collects testimonies and documentary data related to specific historical events from public digital archives, university databases, and research institution databases. Specifically, it downloads interview audio files, written records, video content, etc. and stores them in dedicated storage. The input is the URL or API of the digital archive or database, and the output is the collected data such as audio files, text files, and video files.

[0292] Step 2:

[0293] Data Preprocessing

[0294] The server preprocesses the collected data. Audio data is converted to text using ASR software, and written and video data is converted to text using OCR tools. Furthermore, the server denoises the text data and standardizes its format. The input is the collected audio, text, and video files, and the output is denoised and standardized formatted text data.

[0295] Step 3:

[0296] Data Tagging

[0297] The server assigns tags to the preprocessed text data. Using NLP tools, it extracts keywords and metadata based on the content and emotions of the data and assigns tags such as "Great Kanto Earthquake," "experiences of victims," ​​"hope," and "despair." The input is the preprocessed text data, and the output is tagged text data.

[0298] Step 4:

[0299] AI Learning

[0300] The server trains a generative AI (e.g., GPT-4) on tagged text data. The training process is divided into multiple epochs, and inputs text data and emotional information so that the AI ​​can understand the content and emotions of the data and apply them to generation. The input is tagged text data, and the output is a trained AI model.

[0301] Step 5:

[0302] User request input

[0303] The terminal provides an interface for users to request specific historical stories. Users use options and text boxes to input events, dates, and specific experiences of interest. The input is the request text entered by the user, and the output is the data as the user request.

[0304] Step 6:

[0305] Incorporating an emotion engine

[0306] The device uses an emotion engine to recognize the user's emotions. It uses biometric sensors, cameras, and microphones to analyze the user's facial expressions, tone of voice, heart rate, etc. in real time. The input is the user's facial expression, voice data, and heart rate data, and the output is analyzed emotion data.

[0307] Step 7:

[0308] User request submission

[0309] The terminal transmits the user's request and the recognized emotion data to the server. The input is the user's request and the emotion data, and the output is the request data transmitted to the server.

[0310] Step 8:

[0311] Request Parsing and Data Preparation

[0312] The server analyzes the user's request and emotional data and selects appropriate data. It filters relevant testimonials and materials according to the request content and extracts the most appropriate information based on the emotional data. The input is the user's request and emotional data, and the output is appropriate data for use in generation.

[0313] Step 9:

[0314] Content Generation

[0315] The server generates content using generative AI based on the selected data. Based on user requests and emotional data, it generates a story that recreates a specific historical event or experience. The input is appropriate data and emotional data, and the output is the generated story content.

[0316] Step 10:

[0317] Content provider

[0318] The device provides the generated story to the user. The content is provided in the form of text, audio, or video, and the user can read, listen, or watch it. The input is the generated story content, and the output is the content in a format that the user can view.

[0319] (Application example 2)

[0320] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0321] Conventional information systems for historical events have struggled to provide personalized stories tailored to the user's individual emotions and interests. As a result, users only receive general information, making it difficult to gain a deep understanding or empathy. Furthermore, the lack of emotion-based customization limits the user experience.

[0322] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting testimonies and document data related to a specific historical event, means for preprocessing the collected data to remove noise and standardize formats, means for tagging the preprocessed data and adding metadata, means for training a generative AI using the tagged data, means for a user to request a specific historical story, means for incorporating an emotion engine for recognizing the user's emotions, means for transmitting the recognized emotion data and the user request to the server, means for the generative AI to generate a story based on the user request and emotion data, and means for providing the generated story to the user. This makes it possible to provide a personalized, realistic story that reflects the user's emotions.

[0323] A "specific historical event" refers to an event or series of events of historical significance for which detailed records or testimonies exist.

[0324] "Testimony and documentary data" refers to eyewitness accounts, personal experiences, written records, documents, and other relevant materials relating to contemporary events and circumstances.

[0325] "Preprocessing" refers to the process of removing noise from collected data, standardizing the format, and processing data in a consistent manner to handle data of different formats.

[0326] "Tagging" refers to the process of adding metadata to data that indicates its content or sentiment.

[0327] "Generative AI" refers to an artificial intelligence model that can generate new content based on input data.

[0328] "Request" means a request or question entered by a User seeking specific information or a story.

[0329] An "emotion engine" refers to a system that analyzes biometric data such as a user's facial expression, tone of voice, and heart rate to recognize the user's current emotional state.

[0330] "Providing" refers to delivering the generated story to the user in the form of text, audio, video, etc.

[0331] This invention is a system that collects testimonies and document data related to specific historical events and generates personalized stories that reflect the user's emotions based on them. The system is configured as follows.

[0332] The server collects testimonials and documentary data about specific historical events from public digital archives, academic databases, and publicly available information on the internet, and includes various formats such as interview audio files, written records, and video content.

[0333] The collected data is pre-processed on the server, which includes converting speech to text, removing noise, and standardizing the format, allowing data of various formats to be handled consistently.

[0334] Next, the preprocessed data is given metadata tags based on content, emotion, and theme. For example, testimonies from immediately after the end of the war are given tags such as "war," "end of the war," "hardship," and "hope."

[0335] The tagged data is then fed to a generative AI (e.g., GPT-4) as training data. The AI ​​model is trained over multiple epochs to learn the content and sentiment of the data.

[0336] The user's device provides an interface for the user to request a specific historical story, with text boxes and options for the user to input events, dates, and specific experiences of interest.

[0337] The device also has a built-in emotion engine that recognizes the user's emotions using user input, biometric sensors, cameras, microphones, etc. The emotion engine analyzes the user's emotions in real time based on the user's facial expressions, tone of voice, heart rate, etc.

[0338] A user requests a specific historical story, and the device sends the request and the recognized emotion data to the server. For example, if a user requests "I want to know more about the experiences of victims of the Great Kanto Earthquake," and the emotion engine recognizes the user's sadness or excitement, this information is sent to the server.

[0339] The server prepares the necessary data based on the user request and emotion data, filters the relevant data based on the request content, and extracts the necessary information taking the emotion data into consideration.

[0340] The server uses generative AI to generate stories based on user requests. For example, it generates a story that recreates a specific historical event or experience that reflects the user's emotions based on the learned data. In this case, it depicts the experiences of disaster victims by emphasizing sadness and hope based on the emotional data.

[0341] Finally, the device provides the generated story to the user. The content is provided in the form of text, audio, and video, and the user can read, listen, or watch it. For example, a story recreating the experience of the Great Kanto Earthquake is played in audio format, with narration responding to the user's emotions.

[0342] As a concrete example, the following is an example of a prompt sentence to be input to a generative AI model:

[0343] "Generate a historical story about the Great Kanto Earthquake. Please reflect the user's emotion: sadness and shock."

[0344] This makes it possible to provide users with a personalized and realistic story that responds to their emotions, leading to a deeper understanding and emotional empathy for historical events.

[0345] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0346] Step 1:

[0347] The server collects testimonials and documentary data about a particular historical event. As input, it uses public digital archives, academic databases, and publicly available information on the Internet, and gathers data from these in the form of interview audio files, written records, video content, etc. The output is the original data stored in a variety of formats.

[0348] Step 2:

[0349] The server preprocesses the collected data. Specifically, it converts the audio data into text, transcribes it, and saves it as text data. It also removes noise from the collected text data and standardizes its format. The input is the collected audio data and text data, and the output is preprocessed text data in a unified format.

[0350] Step 3:

[0351] The server assigns tags to the preprocessed data and adds metadata. Specifically, it assigns tags related to the content, emotions, and specific themes of the data. For example, tags such as "war," "end of the war," "hardship," and "hope" are assigned to the data. The input is preprocessed text data, and the output is tagged text data.

[0352] Step 4:

[0353] The server trains a generative AI on the tagged data. Specifically, it trains a generative AI model, such as GPT-4, over multiple epochs to learn the content and sentiment of the data. The input is tagged text data, and the output is a trained generative AI model.

[0354] Step 5:

[0355] The terminal provides an interface through which users can request specific historical stories. Users use text boxes and options to input events, dates, and specific experiences of interest. The input is the user's request, which results in the generation of the requested data. The output is the user's requested data.

[0356] Step 6:

[0357] The device incorporates an emotion engine to recognize the user's emotions. Specifically, it uses biometric sensors, a camera, and a microphone to recognize emotions in real time from the user's facial expressions, tone of voice, and heart rate. The input is the user's biometric data, and the output is analyzed emotional data.

[0358] Step 7:

[0359] The device transmits the user's request and the recognized emotion data to the server. For example, if a user requests, "I want to know about the experiences of victims of the Great Kanto Earthquake," and the emotion engine recognizes the user's sadness or excitement, this information is transmitted to the server. The input is the user's request data and emotion data, and the output is data transmitted to the server.

[0360] Step 8:

[0361] The server analyzes the user request and emotion data and prepares appropriate data. It filters relevant data based on the request content and extracts necessary information taking emotion data into consideration. The input is the user's request information and emotion data, and the output is the filtered and extracted data.

[0362] Step 9:

[0363] The server uses generative AI to generate a story based on the user request. The trained AI model generates a story that reflects the user request and emotions based on the provided data. The inputs are the filtered and extracted data, user request data, and emotion data, and the output is the generated story.

[0364] Step 10:

[0365] The terminal provides the generated story to the user. The content is provided in the form of text, audio, and video, and the user can read, listen to, or watch it. For example, a story recreating the experience of the Great Kanto Earthquake is played in audio format, with narration tailored to the user's emotions. The input is the generated story, and the output is the story provided to the user.

[0366] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0367] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0368] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0369] [Second embodiment]

[0370] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0371] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0372] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0373] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0374] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0375] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0376] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0377] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0378] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0379] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0380] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0381] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0382] This invention is a storytelling system that collects testimonies and documentary data about historical events and trains generative AI to convey real experiences and emotions to the next generation. The system's program provides users with a real historical experience through a series of processes: data collection, preprocessing, tagging, AI learning, request processing, content generation, and provision.

[0383] System action

[0384] 1. Data Collection

[0385] The server collects testimonies and documentary data related to historical events from various data sources, including public digital archives, university and research databases, public information on the Internet, etc. For example, the server collects audio files of interviews, written transcripts, video content, etc.

[0386] 2. Data Preprocessing

[0387] The server preprocesses the collected data. Specifically, it transcribes the audio data, removes noise from the text data, and standardizes notation. For example, it transcribes an audio interview about the Great Kanto Earthquake and corrects noisy parts and typos.

[0388] 3. Data tagging

[0389] The server assigns tags to the preprocessed data. These tags include metadata about the content, emotions, and specific themes of the data. For example, testimonies from immediately after the end of the war might be tagged with "war," "end of the war," "hardship," and "hope."

[0390] 4. AI Learning

[0391] The server then uses the tagged data to train a generative AI model (e.g., GPT-4) that uses this data to learn about people's emotions and experiences regarding specific historical events, improving the quality of the stories it generates.

[0392] 5. User request input

[0393] The terminal provides an interface for users to request specific historical stories. Users input the events, periods, and specific experiences they are interested in. For example, a user might input, "I would like to hear in detail about the experiences of victims of the Great Kanto Earthquake."

[0394] 6. Content Generation

[0395] The server receives user requests and issues instructions to the generative AI, which then generates a realistic story based on relevant data. This story includes realistic emotions and specific episodes based on the training data. For example, the AI ​​recreates the experiences and emotions of victims of the Great Kanto Earthquake and generates a detailed story.

[0396] 7. Content Provision

[0397] The device provides the generated content to the user. The content can be in the form of text, audio, or video, and the user can read, listen, or watch it. For example, a story recreating the experiences of victims of the Great Kanto Earthquake can be provided to the user in audio format.

[0398] This system makes it possible to effectively convey the real experiences and emotions surrounding historical events to future generations, providing a deeper understanding that cannot be captured through records or documents alone.

[0399] The processing flow will be explained below.

[0400] Step 1:

[0401] The server collects testimonies and documentary data about specific historical events, including audio interviews, written records, and video content from public digital archives, university and research databases, and publicly available online sources.

[0402] Step 2:

[0403] The server converts the collected audio data into text. Specifically, it uses voice recognition software to transcribe the interview audio and save it as text data.

[0404] Step 3:

[0405] The server preprocesses the collected text data by removing noise and unnecessary information from the text data and standardizing the format, for example, correcting typos and consolidating data that exists in multiple formats into a standard format.

[0406] Step 4:

[0407] The server then tags the preprocessed text data. Specifically, it assigns metadata tags related to the content, emotions, and specific themes of the data. For example, testimonies from immediately after the end of the war might be tagged with "war," "end of the war," "hardship," and "hope."

[0408] Step 5:

[0409] The server uses the tagged data to train a generative AI model (e.g., GPT-4) over multiple epochs to learn the content and sentiment of the data.

[0410] Step 6:

[0411] The device provides an interface that allows users to request specific historical stories, providing text boxes and options for users to input events, dates, and specific experiences that interest them.

[0412] Step 7:

[0413] The terminal sends the user's request to the server, which then transmits the information entered by the user about the specific historical event or experience.

[0414] Step 8:

[0415] The server analyzes the user request and prepares the appropriate data, specifically filtering the relevant data based on the request content and extracting the required information.

[0416] Step 9:

[0417] The server uses generative AI to generate a story based on the user's request, specifically, a story that recreates a specific historical event or experience requested based on the learned data.

[0418] Step 10:

[0419] The device provides the generated story to the user. Specifically, the device displays or plays the story in the form of text, audio, or video, making it accessible to the user. For example, the device plays a story recreating the experience of the Great Kanto Earthquake in audio format.

[0420] Example 1

[0421] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0422] Testimonies and documentary data about historical events are often fragmented, making continuous preservation and transmission difficult. Furthermore, in order to pass on this information to future generations, it is necessary to express it as a story that includes real experiences and emotions, but current technology makes this difficult. Furthermore, in order to obtain detailed information about a specific historical event that interests a user, they must search for the appropriate information from within a vast amount of data, which is time-consuming and laborious.

[0423] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0424] In this invention, the server includes means for acquiring testimonies and document data related to specific historical events, means for preprocessing the acquired data to remove noise and standardize formats, means for tagging the preprocessed data and adding metadata, means for training a generative artificial intelligence using the tagged data, means for a user to request a specific historical story, means for the generative artificial intelligence to generate a story based on the user request, and means for providing the generated story to the user. This makes it possible to efficiently collect and process information related to historical events and generate realistic stories that meet the user's interests.

[0425] "Testimony and documentary data" is a general term for records and information left by people, either verbally or in writing, about a particular historical event.

[0426] "Means of acquisition" are the techniques and methods used to collect testimonial or documentary data from a particular data source.

[0427] "Preprocessing" is a process for improving the quality of acquired data by removing noise and standardizing the format.

[0428] "Noise removal" is the process of removing unnecessary information and errors contained in data.

[0429] "Format unification" is the process of converting data notation and formats into a unified form.

[0430] "Tagging" refers to adding labels to data that indicate its content or characteristics.

[0431] "Metadata" is supplementary information that describes the content and characteristics of data.

[0432] "Generative AI" is AI that has the ability to generate new information and content based on input data.

[0433] "Means of learning" refers to methods and techniques for training generative artificial intelligence using data.

[0434] A "user request" is an action or content of a user requesting a specific historical story or information from the system.

[0435] "Means for generating a story" refers to the techniques and methods by which generative artificial intelligence creates a new story based on a user request.

[0436] "Means of providing" refers to the technology or method for delivering the generated story to the user in an appropriate format.

[0437] "Content analysis" refers to the process of understanding a user request and preparing the appropriate data in response.

[0438] This invention is a storytelling system that collects testimonies and documentary data related to historical events and trains generative AI to convey real experiences and emotions to the next generation. This system includes a series of processes centered on servers, terminals, and users.

[0439] System Overview

[0440] Data collection

[0441] The server collects testimonies and documentary data related to historical events from various sources, including public digital archives, university and research institute databases, and public information on the Internet. For example, the server uses APIs to download data from public digital archives and web scraping technology to collect public information on the Internet. This includes audio files, written records, and video content.

[0442] Data Preprocessing

[0443] The server preprocesses the collected data. Specifically, it converts the audio data into text (using a speech recognition API, e.g., Google Cloud Speech-to-Text API), removes noise from the text data, and standardizes the format. For example, it transcribes an audio interview about the Great Kanto Earthquake, corrects typos, and standardizes the text to standard Japanese expressions. It also uses noise filtering technology.

[0444] Data Tagging

[0445] The server then tags the preprocessed data. Using natural language processing (NLP) technology, it assigns content and sentiment tags, such as "war," "end of the war," "hardship," and "hope," to the data. It also uses entity recognition technology to extract specific historical events and people's names and automatically add associated tags.

[0446] AI Learning

[0447] The server trains a generative AI model (e.g., GPT-4) on the tagged data. Using machine learning libraries such as PyTorch and TensorFlow, the data is divided into training and validation sets to train the AI ​​model. During this process, hyperparameters are adjusted and the data is shuffled to create a high-quality generative model.

[0448] User request input

[0449] The device provides an interface for users to request specific historical stories. Through a web or mobile application, users can input specific events, dates, and specific experiences. An example prompt might be, "I would like to hear in detail about the experiences of victims of the Great Kanto Earthquake of 1923. Please focus in particular on the emotions and specific events that occurred at the time."

[0450] Content Generation

[0451] The server instructs the generative AI based on the user's request, generating a realistic story based on relevant data. The generated story includes emotions and specific episodes based on the training data. For example, the AI ​​might generate a detailed account of an experience such as, "On September 1, 1923, we were hit by a huge earthquake..."

[0452] Content provider

[0453] The device then provides the generated content to the user. The content can be in the form of text, audio, or video. For example, if the generated story is provided in audio format, the device uses the Google Cloud Text-to-Speech API to convert the text into audio, which the user can read, listen to, or watch.

[0454] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0455] Step 1:

[0456] Data collection

[0457] The server collects testimonies and documentary data about historical events from various sources, including public digital archives, university and research institute databases, and publicly available information on the Internet. Specifically, it uses APIs to download data from digital archives and web scraping technology to collect public information on the Internet, such as audio files, written records, and video content.

[0458] Input: Public digital archives, databases, public information on the internet

[0459] Output: Collected testimonies and literature data

[0460] Step 2:

[0461] Data Preprocessing

[0462] The server performs preprocessing on the collected data. First, it uses a speech recognition API (e.g., Google Cloud Speech-to-Text API) to convert the audio data into text. Next, it standardizes the format of the text data and removes noise. This includes correcting typos, converting to standard Japanese expressions, and removing unnecessary information. For example, it transcribes an interview about the Great Kanto Earthquake and uses noise filtering technology to improve the quality of the audio data.

[0463] Input: Collected testimonies and literature data

[0464] Output: Preprocessed character data

[0465] Step 3:

[0466] Data Tagging

[0467] The server assigns tags to the preprocessed data that indicate its content and characteristics. Using natural language processing (NLP) technology, content and sentiment tags such as "war," "end of the war," "hardship," and "hope" are added to the data. Entity recognition technology is used for tagging, extracting specific historical events and people's names and assigning relevant tags.

[0468] Input: Preprocessed character data

[0469] Output: Tagged data

[0470] Step 4:

[0471] AI Learning

[0472] The server trains a generative AI model (e.g., GPT-4) on the tagged data. It uses machine learning libraries such as PyTorch and TensorFlow to split the data into a training set and a validation set, adjusting hyperparameters and shuffling the data to build a high-quality generative model.

[0473] Input: tagged data

[0474] Output: Trained generative AI model

[0475] Step 5:

[0476] User request input

[0477] The device provides an interface for users to request specific historical stories. Users can input specific events, dates, and specific experiences through a web or mobile application. For example, a user might request, "I'd like to hear in detail about the experiences of victims of the Great Kanto Earthquake of 1923."

[0478] Input: User request

[0479] Output: Request data

[0480] Step 6:

[0481] Content Generation

[0482] The server receives a user request and issues instructions to the generative AI, which then generates a realistic story based on the relevant data. The generated story includes realistic emotions and specific episodes based on the training data. For example, a user can input a prompt such as "Recreate the experiences of victims of the Great Kanto Earthquake of 1923," and a detailed story will be generated.

[0483] Input: Request data, trained generative AI model

[0484] Output: Generated stories

[0485] Step 7:

[0486] Content provider

[0487] The device provides the generated content to the user. The content can be in the form of text, audio, or video, and the user can read, listen, or watch it. For example, if the content is in audio format, the device converts the text to audio using the Google Cloud Text-to-Speech API and provides it to the user.

[0488] Input: Generated stories

[0489] Output: The content that is served to the user

[0490] (Application example 1)

[0491] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0492] In the fields of history education and research, it is difficult to convey to future generations the deep understanding and real experiences that cannot be captured through mere documentary materials and testimonies. Furthermore, the lack of diversity in information and realistic expressions of emotions makes it difficult for learners and researchers to form concrete images. Furthermore, there is a need to provide this information in an easily accessible format.

[0493] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0494] In this invention, the server includes means for collecting testimonies and document data related to specific historical events, means for preprocessing the collected data to remove noise and standardize the format, means for tagging the preprocessed data to add metadata, means for training a generative AI using the tagged data, means for a user to request a specific historical story, means for the generative AI to generate a story based on the user request, means for providing the generated story to the user, and means for providing an API for distributing the generated story to a smartphone app. This allows users to easily access real stories related to historical events in the form of text, audio, and video through the smartphone app, which can be used to deepen their learning and research.

[0495] Below are definitions of important terms included in the claims according to application examples.

[0496] "Specific historical events" refer to historical events associated with specific years or events, such as the Great Kanto Earthquake or war.

[0497] "Testimony and documentary data" refers to people's oral and written testimonies, as well as information recorded in books, reports, etc.

[0498] "Preprocessing" refers to the process of removing noise from collected data and standardizing its format.

[0499] A "tag" refers to identifying information that is added as metadata about the content, emotion, or specific theme of data.

[0500] "Metadata" refers to additional information about the attributes and characteristics of data.

[0501] "Generative AI" refers to artificial intelligence that learns from collected data and generates new stories.

[0502] A "request" refers to a user-input request for a particular historical story.

[0503] "API" stands for Application Program Interface and refers to a mechanism that allows software to communicate with each other.

[0504] "Smartphone app" refers to application software that runs on a smartphone.

[0505] This invention is a storytelling system that collects testimonies and documentary data about specific historical events and trains a generative AI to convey real experiences and emotions to future generations. This system is implemented through the following series of processing steps.

[0506] The main components of the system are data collection, preprocessing, tagging, AI learning, user request, content generation, and content provision. These functions run on the server and on the smartphone app used by users.

[0507] Data collection

[0508] The server collects testimonies and documentary data about specific historical events from publicly available information on the Internet, digital archives, and databases of universities and research institutions. For example, the server collects audio files, written records, and video content from interviews about the Great Kanto Earthquake. This allows information from a variety of data sources to be managed in a unified manner.

[0509] Data Preprocessing

[0510] The collected data is preprocessed on the server. During this process, the audio data is transcribed, noise is removed from the text data, and the notation is standardized. Specifically, an audio interview about the "Great Kanto Earthquake" is transcribed, and noisy parts and typos are corrected.

[0511] Data Tagging

[0512] The preprocessed data is then tagged on the server. These tags include metadata about the content, emotions, and specific themes of the data. For example, testimonies from immediately after the end of the war are tagged with "war," "end of the war," "hardship," and "hope." This allows for efficient data search and management.

[0513] AI Learning

[0514] Based on the tagged data, the server trains a generative AI model (e.g., GPT-4). This AI model uses the collected data to learn people's emotions and experiences regarding specific historical events, improving the quality of the stories it generates. This allows it to provide users with realistic and emotionally rich stories.

[0515] User request input and content generation

[0516] Through a smartphone app, users have an interface that allows them to request a specific historical story, for example, by typing, "I'd like to hear in detail about the experiences of victims of the Great Kanto Earthquake." Based on this request, the server instructs a generative AI model to generate a realistic story based on relevant data.

[0517] Content provider

[0518] The generated story is provided to the user in the form of text, audio, or video. For example, the server generates an audio story recreating the experiences of victims of the Great Kanto Earthquake and delivers it to the user via a smartphone app. The generated content is also delivered to the smartphone app via an API.

[0519] Examples of concrete examples and prompts

[0520] As a specific example, if a user opens the app and requests to listen to an audio account of the Great Kanto Earthquake, the AI ​​will generate a realistic story, display it in the app, convert it into audio, and play it back.

[0521] Examples of prompts include:

[0522] "What was your experience as a victim of the Great Kanto Earthquake?"

[0523] How did life change after the war ended?

[0524] "Tell me about the hope and despair you felt immediately after the end of the war."

[0525] This allows users to gain a deeper emotional connection to historical events, making it a useful tool for education and research.

[0526] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0527] Step 1:

[0528] The server collects testimonies and documentary data related to a specific historical event. In this process, data is collected from publicly available information on the Internet, digital archives, and databases of universities and research institutions. For example, it collects audio files of interviews, written records, and video content related to the "Great Kanto Earthquake." The input is a search keyword (e.g., "Great Kanto Earthquake"), and the output is the collected data.

[0529] Step 2:

[0530] The server preprocesses the collected data. This involves transcribing the audio data, removing noise from the text data, and standardizing the notation. Specifically, it converts the audio data into text using speech recognition software (e.g., Google Speech-to-Text API) and cleans the text. The input is the collected data, and the output is the preprocessed data.

[0531] Step 3:

[0532] The server then tags the preprocessed data. During this process, it adds metadata about the content, emotions, and specific themes of the data. For example, tags such as "war," "end of the war," "hardship," and "hope" are added to testimonies from immediately after the end of the war. The input is the preprocessed data, and the output is the tagged data.

[0533] Step 4:

[0534] The server trains a generative AI model using the tagged data. During this process, the dataset is input into a generative AI (e.g., GPT-4) to train the model. This allows the AI ​​to learn people's emotions and experiences regarding specific historical events and improve its story-generating capabilities. The input is the tagged data, and the output is the trained AI model.

[0535] Step 5:

[0536] Users request a specific historical story through a smartphone app. This process involves inputting search keywords and specific inquiries into the on-screen interface. For example, they might input, "I'd like to hear in detail about the experiences of victims of the Great Kanto Earthquake." The input is the user request, and the output is a prompt to the generative AI.

[0537] Step 6:

[0538] The server issues instructions to the generative AI based on the user's request and generates a story. During this process, the AI ​​model is instructed to generate a story based on the prompt. The generated story contains realistic emotions and specific episodes based on the training data. The input is the user's request prompt, and the output is the generated story.

[0539] Step 7:

[0540] The server provides the generated story to the user using an API to distribute it to a smartphone app. In this process, content is provided in the form of text, audio, or video. For example, the generated story can be sent to the app in audio format, allowing the user to listen to it. The input is the generated story, and the output is the content provided to the user.

[0541] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0542] This invention combines a system that uses generative AI to learn testimonies and document data related to historical events, and generates and provides realistic stories based on user requests, with an emotion engine that recognizes the user's emotions and reflects them in the story generation.

[0543] System action

[0544] 1. Data Collection

[0545] The server collects testimonies and documentary data about specific historical events, including audio interviews, written records, and video content from public digital archives, university and research databases, and publicly available online sources.

[0546] 2. Data Preprocessing

[0547] The server preprocesses the collected data. Specifically, it converts the audio data into text, transcribes it, and saves it as text data. It also removes noise from the text data and standardizes its format.

[0548] 3. Data tagging

[0549] The server then tags the preprocessed data. Specifically, it assigns metadata tags related to the content, emotions, and specific themes of the data. For example, testimonies from immediately after the end of the war might be tagged with "war," "end of the war," "hardship," and "hope."

[0550] 4. AI Learning

[0551] The server uses the tagged data to train a generative AI model (e.g., GPT-4) over multiple epochs to learn the content and sentiment of the data.

[0552] 5. User request input

[0553] The device provides an interface that allows users to request specific historical stories, providing text boxes and options for users to input events, dates, and specific experiences that interest them.

[0554] 6. Incorporating an Emotional Engine

[0555] The device incorporates an emotion engine to recognize the user's emotions. Specifically, it recognizes emotions from the user's input, as well as from biometric sensors, cameras, microphones, and other sensors, based on the user's facial expressions, tone of voice, heart rate, and other information.

[0556] 7. User Request Submission

[0557] The device sends the user's request and the recognized emotion data to the server. For example, if a user inputs "I would like to hear in detail about the experiences of victims of the Great Kanto Earthquake," the emotion engine will simultaneously recognize the user's emotions, such as sadness or excitement.

[0558] 8. Request Parsing and Data Preparation

[0559] The server analyzes the user request and emotion data and prepares appropriate data, which involves filtering relevant data based on the request content and extracting necessary information taking emotion data into account.

[0560] 9. Content Generation

[0561] The server uses generative AI to generate stories based on user requests. Specifically, it generates stories that recreate specific historical events or experiences that reflect the user's emotions based on the learned data. For example, it depicts the experiences of disaster victims by emphasizing sadness and hope based on emotional data.

[0562] 10. Content Provision

[0563] The device then provides the generated story to the user. The content is provided in the form of text, audio, and video, and the user can read, listen, or watch it. For example, a story recreating the experience of the Great Kanto Earthquake is played in audio format, with narration tailored to the user's emotions.

[0564] This system not only allows users to share their real experiences and emotions about historical events with future generations, but also provides customized stories based on the user's emotions, enabling efficient and effective historical transmission that leads to deeper understanding and emotional empathy.

[0565] The processing flow will be explained below.

[0566] Step 1:

[0567] The server collects testimonies and documentary data about specific historical events, using public digital archives, university and research databases, and publicly available information on the internet to retrieve interview audio files, written transcripts, video content, and more.

[0568] Step 2:

[0569] The server converts the collected audio data into text using speech recognition software, transcribing the interview audio and saving it as text data.

[0570] Step 3:

[0571] The server preprocesses the collected text data by removing noise and standardizing notation, for example, correcting typos and consolidating data in multiple formats into a standard format.

[0572] Step 4:

[0573] The server then tags the preprocessed text data, adding metadata tags related to the content, emotions, and specific themes of the data. For example, an interview about the Great Kanto Earthquake might be tagged with words like "earthquake," "damage," "reconstruction," and "hope."

[0574] Step 5:

[0575] The server uses the tagged data to train a generative AI model (e.g., GPT-4) over multiple epochs to learn the content and sentiment of the data.

[0576] Step 6:

[0577] The device provides an interface that allows users to request specific historical stories, providing text boxes and options for users to input events, dates, and specific experiences that interest them.

[0578] Step 7:

[0579] The device incorporates an emotion engine to recognize the user's emotions. It utilizes biometric sensors, cameras, and microphones to analyze the user's facial expressions, tone of voice, heart rate, etc., and recognizes the user's emotional state in real time.

[0580] Step 8:

[0581] The device sends the user's request and the recognized emotion data to the server. For example, if the user inputs "I would like to know more about the experiences of victims of the Great Kanto Earthquake," the emotion engine will recognize the user's sadness or excitement.

[0582] Step 9:

[0583] The server analyzes the user request and emotion data and prepares appropriate data, for example, filtering relevant data based on the request content and extracting necessary information taking emotion data into consideration.

[0584] Step 10:

[0585] The server uses generative AI to generate stories based on user requests. Based on the data the server has learned, it generates stories that recreate specific historical events or experiences that reflect the user's emotions. For example, based on emotional data, it could portray the experiences of disaster victims by emphasizing sadness and hope.

[0586] Step 11:

[0587] The device then provides the generated story to the user. The content is provided in the form of text, audio, and video, and the user can read, listen, or watch it. For example, a story recreating the experience of the Great Kanto Earthquake is played in audio format, with narration tailored to the user's emotions.

[0588] Example 2

[0589] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0590] In systems that use generative AI to generate realistic stories using testimonies and document data about historical events and provide them to users, there is a demand for systems that can generate stories that reflect the user's emotions, thereby gaining deeper empathy and understanding.

[0591] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0592] In this invention, the server includes means for collecting testimonies and document data related to specific historical events, means for preprocessing the collected data to remove noise and standardize the format, means for tagging the preprocessed data and adding metadata, means for training a generative AI using the tagged data, means for a user to request a specific historical story, means for recognizing the user's emotions and acquiring the emotion data, means for the generative AI to generate a story based on the user request and the emotion data, and means for providing the generated story to the user. This enables the generation of realistic and emotionally rich stories based on the user's emotions.

[0593] A "specific historical event" refers to an important occurrence or phenomenon in a particular era or region, and has historical significance.

[0594] A "testimony" is a statement or record made by an individual about a particular event or experience, and may be recorded in the form of an audio file, written document, video, or other format.

[0595] "Documentary data" refers to written data such as books, papers, articles, and official records related to a particular historical event.

[0596] "Means of collection" refers to the means of obtaining specific information using the Internet, digital archives, databases, etc.

[0597] "Preprocessing" refers to data formatting, including removing noise from the data and standardizing the format.

[0598] "Means of adding metadata by tagging" refers to a means of adding keywords and attribute information that indicate the content and characteristics of the data to the data, making subsequent searches and analysis easier.

[0599] "Generative AI" refers to artificial intelligence technology that generates natural language based on large datasets, and includes models such as GPT-4.

[0600] "Means for a user to request a particular historical story" refers to the means by which a user inputs historical stories of interest through the interface.

[0601] "Means for recognizing the user's emotions and acquiring emotional data" refers to means for detecting emotions by analyzing the user's facial expressions, tone of voice, heart rate, etc., and acquiring that data.

[0602] "Means for generating stories" refers to a means for generating stories in text, audio, or video format using generative AI based on user requests and emotional data.

[0603] The "means for providing" refers to a means for delivering the generated story to the user in a predetermined format.

[0604] This invention is a system that uses generative AI to generate realistic stories based on testimonies and document data about historical events, and provides these to users. Furthermore, it aims to gain deeper empathy and understanding by recognizing the user's emotions and generating stories based on them.

[0605] Data collection

[0606] The server collects testimonies and documentary data about specific historical events, including interview audio files, written records, and video content, from public digital archives, university and research institution databases, and publicly available information on the Internet, and stores them in dedicated storage.

[0607] Data Preprocessing

[0608] The server preprocesses the collected data. Audio data is converted to text using ASR (automatic speech recognition) software, and written and video data is converted to text using OCR (optical character recognition) tools. This text data is then denoised and standardized, and saved as a standardized dataset.

[0609] Data Tagging

[0610] The server then assigns tags to the preprocessed data, using NLP (natural language processing) tools to extract keywords based on the content and emotions of the data, and assigns tags such as "Great Kanto Earthquake," "experiences of victims," ​​"hope," and "despair."

[0611] AI Learning

[0612] The server trains a generative AI (e.g., GPT-4) on the tagged data. The training process is divided into multiple epochs, and inputs text data and emotional information into a large database so that the AI ​​can understand and apply them to generation.

[0613] User request input

[0614] The terminal provides a user input interface. Using options and text boxes, the user can enter details of historical events or experiences that interest them. For example, they can enter text such as "I would like to hear about experiences of the Great Kanto Earthquake."

[0615] Incorporating an emotion engine

[0616] The device is equipped with a device for recognizing the user's emotions using an emotion engine. It uses biometric sensors, a camera, and a microphone to analyze the user's facial expressions, tone of voice, heart rate, etc. in real time to detect whether the user is excited or sad.

[0617] User request submission

[0618] The device sends the user's request and the recognized emotional data to the server. For example, a request such as "I would like to hear in detail about the experiences of victims of the Great Kanto Earthquake" is sent to the server along with emotional data such as "sadness" or "excitement."

[0619] Request Parsing and Data Preparation

[0620] The server analyzes the user's request and emotional data to select the appropriate data, searching for testimonies and documents related to the requested historical event and filtering the most suitable ones based on the emotional data.

[0621] Content Generation

[0622] The server uses generative AI to generate content based on the selected data. Based on the emotional data, it can generate a story detailing the experiences of disaster victims, reflecting emotions such as sadness and hope.

[0623] Content provider

[0624] The device provides the generated story to the user, who can listen to it in text, audio, and video formats. Specifically, audio content recreating the experience of the Great Kanto Earthquake is provided, with narration based on emotions.

[0625] Prompt Sentence Examples

[0626] A user can send a request to the system using a prompt such as:

[0627] "I would like to hear in detail about the experiences of the victims of the Great Kanto Earthquake. I would like to know what difficulties they faced and how they overcame them, including the emotional aspects. I would also be happy if you could touch on their feelings of hope and courage."

[0628] As a result, this system enables the generation of realistic and emotionally rich stories based on the user's emotions, providing a deep understanding and empathy for historical events.

[0629] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0630] Step 1:

[0631] Data collection

[0632] The server collects testimonies and documentary data related to specific historical events from public digital archives, university databases, and research institution databases. Specifically, it downloads interview audio files, written records, video content, etc. and stores them in dedicated storage. The input is the URL or API of the digital archive or database, and the output is the collected data such as audio files, text files, and video files.

[0633] Step 2:

[0634] Data Preprocessing

[0635] The server preprocesses the collected data. Audio data is converted to text using ASR software, and written and video data is converted to text using OCR tools. Furthermore, the server denoises the text data and standardizes its format. The input is the collected audio, text, and video files, and the output is denoised and standardized formatted text data.

[0636] Step 3:

[0637] Data Tagging

[0638] The server assigns tags to the preprocessed text data. Using NLP tools, it extracts keywords and metadata based on the content and emotions of the data and assigns tags such as "Great Kanto Earthquake," "experiences of victims," ​​"hope," and "despair." The input is the preprocessed text data, and the output is tagged text data.

[0639] Step 4:

[0640] AI Learning

[0641] The server trains a generative AI (e.g., GPT-4) on tagged text data. The training process is divided into multiple epochs, and inputs text data and emotional information so that the AI ​​can understand the content and emotions of the data and apply them to generation. The input is tagged text data, and the output is a trained AI model.

[0642] Step 5:

[0643] User request input

[0644] The terminal provides an interface for users to request specific historical stories. Users use options and text boxes to input events, dates, and specific experiences of interest. The input is the request text entered by the user, and the output is the data as the user request.

[0645] Step 6:

[0646] Incorporating an emotion engine

[0647] The device uses an emotion engine to recognize the user's emotions. It uses biometric sensors, cameras, and microphones to analyze the user's facial expressions, tone of voice, heart rate, etc. in real time. The input is the user's facial expression, voice data, and heart rate data, and the output is analyzed emotion data.

[0648] Step 7:

[0649] User request submission

[0650] The terminal transmits the user's request and the recognized emotion data to the server. The input is the user's request and the emotion data, and the output is the request data transmitted to the server.

[0651] Step 8:

[0652] Request Parsing and Data Preparation

[0653] The server analyzes the user's request and emotional data and selects appropriate data. It filters relevant testimonials and materials according to the request content and extracts the most appropriate information based on the emotional data. The input is the user's request and emotional data, and the output is appropriate data for use in generation.

[0654] Step 9:

[0655] Content Generation

[0656] The server generates content using generative AI based on the selected data. Based on user requests and emotional data, it generates a story that recreates a specific historical event or experience. The input is appropriate data and emotional data, and the output is the generated story content.

[0657] Step 10:

[0658] Content provider

[0659] The device provides the generated story to the user. The content is provided in the form of text, audio, or video, and the user can read, listen, or watch it. The input is the generated story content, and the output is the content in a format that the user can view.

[0660] (Application example 2)

[0661] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0662] Conventional information systems for historical events have struggled to provide personalized stories tailored to the user's individual emotions and interests. As a result, users only receive general information, making it difficult to gain a deep understanding or empathy. Furthermore, the lack of emotion-based customization limits the user experience.

[0663] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting testimonies and document data related to a specific historical event, means for preprocessing the collected data to remove noise and standardize formats, means for tagging the preprocessed data and adding metadata, means for training a generative AI using the tagged data, means for a user to request a specific historical story, means for incorporating an emotion engine for recognizing the user's emotions, means for transmitting the recognized emotion data and the user request to the server, means for the generative AI to generate a story based on the user request and emotion data, and means for providing the generated story to the user. This makes it possible to provide a personalized, realistic story that reflects the user's emotions.

[0664] A "specific historical event" refers to an event or series of events of historical significance for which detailed records or testimonies exist.

[0665] "Testimony and documentary data" refers to eyewitness accounts, personal experiences, written records, documents, and other relevant materials relating to contemporary events and circumstances.

[0666] "Preprocessing" refers to the process of removing noise from collected data, standardizing the format, and processing data in a consistent manner to handle data of different formats.

[0667] "Tagging" refers to the process of adding metadata to data that indicates its content or sentiment.

[0668] "Generative AI" refers to an artificial intelligence model that can generate new content based on input data.

[0669] "Request" means a request or question entered by a User seeking specific information or a story.

[0670] An "emotion engine" refers to a system that analyzes biometric data such as a user's facial expression, tone of voice, and heart rate to recognize the user's current emotional state.

[0671] "Providing" refers to delivering the generated story to the user in the form of text, audio, video, etc.

[0672] This invention is a system that collects testimonies and document data related to specific historical events and generates personalized stories that reflect the user's emotions based on them. The system is configured as follows.

[0673] The server collects testimonials and documentary data about specific historical events from public digital archives, academic databases, and publicly available information on the internet, and includes various formats such as interview audio files, written records, and video content.

[0674] The collected data is pre-processed on the server, which includes converting speech to text, removing noise, and standardizing the format, allowing data of various formats to be handled consistently.

[0675] Next, the preprocessed data is given metadata tags based on content, emotion, and theme. For example, testimonies from immediately after the end of the war are given tags such as "war," "end of the war," "hardship," and "hope."

[0676] The tagged data is then fed to a generative AI (e.g., GPT-4) as training data. The AI ​​model is trained over multiple epochs to learn the content and sentiment of the data.

[0677] The user's device provides an interface for the user to request a specific historical story, with text boxes and options for the user to input events, dates, and specific experiences of interest.

[0678] The device also has a built-in emotion engine that recognizes the user's emotions using user input, biometric sensors, cameras, microphones, etc. The emotion engine analyzes the user's emotions in real time based on the user's facial expressions, tone of voice, heart rate, etc.

[0679] A user requests a specific historical story, and the device sends the request and the recognized emotion data to the server. For example, if a user requests "I want to know more about the experiences of victims of the Great Kanto Earthquake," and the emotion engine recognizes the user's sadness or excitement, this information is sent to the server.

[0680] The server prepares the necessary data based on the user request and emotion data, filters the relevant data based on the request content, and extracts the necessary information taking the emotion data into consideration.

[0681] The server uses generative AI to generate stories based on user requests. For example, it generates a story that recreates a specific historical event or experience that reflects the user's emotions based on the learned data. In this case, it depicts the experiences of disaster victims by emphasizing sadness and hope based on the emotional data.

[0682] Finally, the device provides the generated story to the user. The content is provided in the form of text, audio, and video, and the user can read, listen, or watch it. For example, a story recreating the experience of the Great Kanto Earthquake is played in audio format, with narration responding to the user's emotions.

[0683] As a concrete example, the following is an example of a prompt sentence to be input to a generative AI model:

[0684] "Generate a historical story about the Great Kanto Earthquake. Please reflect the user's emotion: sadness and shock."

[0685] This makes it possible to provide users with a personalized and realistic story that responds to their emotions, leading to a deeper understanding and emotional empathy for historical events.

[0686] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0687] Step 1:

[0688] The server collects testimonials and documentary data about a particular historical event. As input, it uses public digital archives, academic databases, and publicly available information on the Internet, and gathers data from these in the form of interview audio files, written records, video content, etc. The output is the original data stored in a variety of formats.

[0689] Step 2:

[0690] The server preprocesses the collected data. Specifically, it converts the audio data into text, transcribes it, and saves it as text data. It also removes noise from the collected text data and standardizes its format. The input is the collected audio data and text data, and the output is preprocessed text data in a unified format.

[0691] Step 3:

[0692] The server assigns tags to the preprocessed data and adds metadata. Specifically, it assigns tags related to the content, emotions, and specific themes of the data. For example, tags such as "war," "end of the war," "hardship," and "hope" are assigned to the data. The input is preprocessed text data, and the output is tagged text data.

[0693] Step 4:

[0694] The server trains a generative AI on the tagged data. Specifically, it trains a generative AI model, such as GPT-4, over multiple epochs to learn the content and sentiment of the data. The input is tagged text data, and the output is a trained generative AI model.

[0695] Step 5:

[0696] The terminal provides an interface through which users can request specific historical stories. Users use text boxes and options to input events, dates, and specific experiences of interest. The input is the user's request, which results in the generation of the requested data. The output is the user's requested data.

[0697] Step 6:

[0698] The device incorporates an emotion engine to recognize the user's emotions. Specifically, it uses biometric sensors, a camera, and a microphone to recognize emotions in real time from the user's facial expressions, tone of voice, and heart rate. The input is the user's biometric data, and the output is analyzed emotional data.

[0699] Step 7:

[0700] The device transmits the user's request and the recognized emotion data to the server. For example, if a user requests, "I want to know about the experiences of victims of the Great Kanto Earthquake," and the emotion engine recognizes the user's sadness or excitement, this information is transmitted to the server. The input is the user's request data and emotion data, and the output is data transmitted to the server.

[0701] Step 8:

[0702] The server analyzes the user request and emotion data and prepares appropriate data. It filters relevant data based on the request content and extracts necessary information taking emotion data into consideration. The input is the user's request information and emotion data, and the output is the filtered and extracted data.

[0703] Step 9:

[0704] The server uses generative AI to generate a story based on the user request. The trained AI model generates a story that reflects the user request and emotions based on the provided data. The inputs are the filtered and extracted data, user request data, and emotion data, and the output is the generated story.

[0705] Step 10:

[0706] The terminal provides the generated story to the user. The content is provided in the form of text, audio, and video, and the user can read, listen to, or watch it. For example, a story recreating the experience of the Great Kanto Earthquake is played in audio format, with narration tailored to the user's emotions. The input is the generated story, and the output is the story provided to the user.

[0707] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0708] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0709] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0710] [Third embodiment]

[0711] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0712] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0713] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0714] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0715] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0716] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0717] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0718] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0719] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0720] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0721] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0722] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0723] This invention is a storytelling system that collects testimonies and documentary data about historical events and trains generative AI to convey real experiences and emotions to the next generation. The system's program provides users with a real historical experience through a series of processes: data collection, preprocessing, tagging, AI learning, request processing, content generation, and provision.

[0724] System action

[0725] 1. Data Collection

[0726] The server collects testimonies and documentary data related to historical events from various data sources, including public digital archives, university and research databases, public information on the Internet, etc. For example, the server collects audio files of interviews, written transcripts, video content, etc.

[0727] 2. Data Preprocessing

[0728] The server preprocesses the collected data. Specifically, it transcribes the audio data, removes noise from the text data, and standardizes notation. For example, it transcribes an audio interview about the Great Kanto Earthquake and corrects noisy parts and typos.

[0729] 3. Data tagging

[0730] The server assigns tags to the preprocessed data. These tags include metadata about the content, emotions, and specific themes of the data. For example, testimonies from immediately after the end of the war might be tagged with "war," "end of the war," "hardship," and "hope."

[0731] 4. AI Learning

[0732] The server then uses the tagged data to train a generative AI model (e.g., GPT-4) that uses this data to learn about people's emotions and experiences regarding specific historical events, improving the quality of the stories it generates.

[0733] 5. User request input

[0734] The terminal provides an interface for users to request specific historical stories. Users input the events, periods, and specific experiences they are interested in. For example, a user might input, "I would like to hear in detail about the experiences of victims of the Great Kanto Earthquake."

[0735] 6. Content Generation

[0736] The server receives user requests and issues instructions to the generative AI, which then generates a realistic story based on relevant data. This story includes realistic emotions and specific episodes based on the training data. For example, the AI ​​recreates the experiences and emotions of victims of the Great Kanto Earthquake and generates a detailed story.

[0737] 7. Content Provision

[0738] The device provides the generated content to the user. The content can be in the form of text, audio, or video, and the user can read, listen, or watch it. For example, a story recreating the experiences of victims of the Great Kanto Earthquake can be provided to the user in audio format.

[0739] This system makes it possible to effectively convey the real experiences and emotions surrounding historical events to future generations, providing a deeper understanding that cannot be captured through records or documents alone.

[0740] The processing flow will be explained below.

[0741] Step 1:

[0742] The server collects testimonies and documentary data about specific historical events, including audio interviews, written records, and video content from public digital archives, university and research databases, and publicly available online sources.

[0743] Step 2:

[0744] The server converts the collected audio data into text. Specifically, it uses voice recognition software to transcribe the interview audio and save it as text data.

[0745] Step 3:

[0746] The server preprocesses the collected text data by removing noise and unnecessary information from the text data and standardizing the format, for example, correcting typos and consolidating data that exists in multiple formats into a standard format.

[0747] Step 4:

[0748] The server then tags the preprocessed text data. Specifically, it assigns metadata tags related to the content, emotions, and specific themes of the data. For example, testimonies from immediately after the end of the war might be tagged with "war," "end of the war," "hardship," and "hope."

[0749] Step 5:

[0750] The server uses the tagged data to train a generative AI model (e.g., GPT-4) over multiple epochs to learn the content and sentiment of the data.

[0751] Step 6:

[0752] The device provides an interface that allows users to request specific historical stories, providing text boxes and options for users to input events, dates, and specific experiences that interest them.

[0753] Step 7:

[0754] The terminal sends the user's request to the server, which then transmits the information entered by the user about the specific historical event or experience.

[0755] Step 8:

[0756] The server analyzes the user request and prepares the appropriate data, specifically filtering the relevant data based on the request content and extracting the required information.

[0757] Step 9:

[0758] The server uses generative AI to generate a story based on the user's request, specifically, a story that recreates a specific historical event or experience requested based on the learned data.

[0759] Step 10:

[0760] The device provides the generated story to the user. Specifically, the device displays or plays the story in the form of text, audio, or video, making it accessible to the user. For example, the device plays a story recreating the experience of the Great Kanto Earthquake in audio format.

[0761] Example 1

[0762] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0763] Testimonies and documentary data about historical events are often fragmented, making continuous preservation and transmission difficult. Furthermore, in order to pass on this information to future generations, it is necessary to express it as a story that includes real experiences and emotions, but current technology makes this difficult. Furthermore, in order to obtain detailed information about a specific historical event that interests a user, they must search for the appropriate information from within a vast amount of data, which is time-consuming and laborious.

[0764] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0765] In this invention, the server includes means for acquiring testimonies and document data related to specific historical events, means for preprocessing the acquired data to remove noise and standardize formats, means for tagging the preprocessed data and adding metadata, means for training a generative artificial intelligence using the tagged data, means for a user to request a specific historical story, means for the generative artificial intelligence to generate a story based on the user request, and means for providing the generated story to the user. This makes it possible to efficiently collect and process information related to historical events and generate realistic stories that meet the user's interests.

[0766] "Testimony and documentary data" is a general term for records and information left by people, either verbally or in writing, about a particular historical event.

[0767] "Means of acquisition" are the techniques and methods used to collect testimonial or documentary data from a particular data source.

[0768] "Preprocessing" is a process for improving the quality of acquired data by removing noise and standardizing the format.

[0769] "Noise removal" is the process of removing unnecessary information and errors contained in data.

[0770] "Format unification" is the process of converting data notation and formats into a unified form.

[0771] "Tagging" refers to adding labels to data that indicate its content or characteristics.

[0772] "Metadata" is supplementary information that describes the content and characteristics of data.

[0773] "Generative AI" is AI that has the ability to generate new information and content based on input data.

[0774] "Means of learning" refers to methods and techniques for training generative artificial intelligence using data.

[0775] A "user request" is an action or content of a user requesting a specific historical story or information from the system.

[0776] "Means for generating a story" refers to the techniques and methods by which generative artificial intelligence creates a new story based on a user request.

[0777] "Means of providing" refers to the technology or method for delivering the generated story to the user in an appropriate format.

[0778] "Content analysis" refers to the process of understanding a user request and preparing the appropriate data in response.

[0779] This invention is a storytelling system that collects testimonies and documentary data related to historical events and trains generative AI to convey real experiences and emotions to the next generation. This system includes a series of processes centered on servers, terminals, and users.

[0780] System Overview

[0781] Data collection

[0782] The server collects testimonies and documentary data related to historical events from various sources, including public digital archives, university and research institute databases, and public information on the Internet. For example, the server uses APIs to download data from public digital archives and web scraping technology to collect public information on the Internet. This includes audio files, written records, and video content.

[0783] Data Preprocessing

[0784] The server preprocesses the collected data. Specifically, it converts the audio data into text (using a speech recognition API, e.g., Google Cloud Speech-to-Text API), removes noise from the text data, and standardizes the format. For example, it transcribes an audio interview about the Great Kanto Earthquake, corrects typos, and standardizes the text to standard Japanese expressions. It also uses noise filtering technology.

[0785] Data Tagging

[0786] The server then tags the preprocessed data. Using natural language processing (NLP) technology, it assigns content and sentiment tags, such as "war," "end of the war," "hardship," and "hope," to the data. It also uses entity recognition technology to extract specific historical events and people's names and automatically add associated tags.

[0787] AI Learning

[0788] The server trains a generative AI model (e.g., GPT-4) on the tagged data. Using machine learning libraries such as PyTorch and TensorFlow, the data is divided into training and validation sets to train the AI ​​model. During this process, hyperparameters are adjusted and the data is shuffled to create a high-quality generative model.

[0789] User request input

[0790] The device provides an interface for users to request specific historical stories. Through a web or mobile application, users can input specific events, dates, and specific experiences. An example prompt might be, "I would like to hear in detail about the experiences of victims of the Great Kanto Earthquake of 1923. Please focus in particular on the emotions and specific events that occurred at the time."

[0791] Content Generation

[0792] The server instructs the generative AI based on the user's request, generating a realistic story based on relevant data. The generated story includes emotions and specific episodes based on the training data. For example, the AI ​​might generate a detailed account of an experience such as, "On September 1, 1923, we were hit by a huge earthquake..."

[0793] Content provider

[0794] The device then provides the generated content to the user. The content can be in the form of text, audio, or video. For example, if the generated story is provided in audio format, the device uses the Google Cloud Text-to-Speech API to convert the text into audio, which the user can read, listen to, or watch.

[0795] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0796] Step 1:

[0797] Data collection

[0798] The server collects testimonies and documentary data about historical events from various sources, including public digital archives, university and research institute databases, and publicly available information on the Internet. Specifically, it uses APIs to download data from digital archives and web scraping technology to collect public information on the Internet, such as audio files, written records, and video content.

[0799] Input: Public digital archives, databases, public information on the internet

[0800] Output: Collected testimonies and literature data

[0801] Step 2:

[0802] Data Preprocessing

[0803] The server performs preprocessing on the collected data. First, it uses a speech recognition API (e.g., Google Cloud Speech-to-Text API) to convert the audio data into text. Next, it standardizes the format of the text data and removes noise. This includes correcting typos, converting to standard Japanese expressions, and removing unnecessary information. For example, it transcribes an interview about the Great Kanto Earthquake and uses noise filtering technology to improve the quality of the audio data.

[0804] Input: Collected testimonies and literature data

[0805] Output: Preprocessed character data

[0806] Step 3:

[0807] Data Tagging

[0808] The server assigns tags to the preprocessed data that indicate its content and characteristics. Using natural language processing (NLP) technology, content and sentiment tags such as "war," "end of the war," "hardship," and "hope" are added to the data. Entity recognition technology is used for tagging, extracting specific historical events and people's names and assigning relevant tags.

[0809] Input: Preprocessed character data

[0810] Output: Tagged data

[0811] Step 4:

[0812] AI Learning

[0813] The server trains a generative AI model (e.g., GPT-4) on the tagged data. It uses machine learning libraries such as PyTorch and TensorFlow to split the data into a training set and a validation set, adjusting hyperparameters and shuffling the data to build a high-quality generative model.

[0814] Input: tagged data

[0815] Output: Trained generative AI model

[0816] Step 5:

[0817] User request input

[0818] The device provides an interface for users to request specific historical stories. Users can input specific events, dates, and specific experiences through a web or mobile application. For example, a user might request, "I'd like to hear in detail about the experiences of victims of the Great Kanto Earthquake of 1923."

[0819] Input: User request

[0820] Output: Request data

[0821] Step 6:

[0822] Content Generation

[0823] The server receives a user request and issues instructions to the generative AI, which then generates a realistic story based on the relevant data. The generated story includes realistic emotions and specific episodes based on the training data. For example, a user can input a prompt such as "Recreate the experiences of victims of the Great Kanto Earthquake of 1923," and a detailed story will be generated.

[0824] Input: Request data, trained generative AI model

[0825] Output: Generated stories

[0826] Step 7:

[0827] Content provider

[0828] The device provides the generated content to the user. The content can be in the form of text, audio, or video, and the user can read, listen, or watch it. For example, if the content is in audio format, the device converts the text to audio using the Google Cloud Text-to-Speech API and provides it to the user.

[0829] Input: Generated stories

[0830] Output: The content that is served to the user

[0831] (Application example 1)

[0832] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0833] In the fields of history education and research, it is difficult to convey to future generations the deep understanding and real experiences that cannot be captured through mere documentary materials and testimonies. Furthermore, the lack of diversity in information and realistic expressions of emotions makes it difficult for learners and researchers to form concrete images. Furthermore, there is a need to provide this information in an easily accessible format.

[0834] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0835] In this invention, the server includes means for collecting testimonies and document data related to specific historical events, means for preprocessing the collected data to remove noise and standardize the format, means for tagging the preprocessed data to add metadata, means for training a generative AI using the tagged data, means for a user to request a specific historical story, means for the generative AI to generate a story based on the user request, means for providing the generated story to the user, and means for providing an API for distributing the generated story to a smartphone app. This allows users to easily access real stories related to historical events in the form of text, audio, and video through the smartphone app, which can be used to deepen their learning and research.

[0836] Below are definitions of important terms included in the claims according to application examples.

[0837] "Specific historical events" refer to historical events associated with specific years or events, such as the Great Kanto Earthquake or war.

[0838] "Testimony and documentary data" refers to people's oral and written testimonies, as well as information recorded in books, reports, etc.

[0839] "Preprocessing" refers to the process of removing noise from collected data and standardizing its format.

[0840] A "tag" refers to identifying information that is added as metadata about the content, emotion, or specific theme of data.

[0841] "Metadata" refers to additional information about the attributes and characteristics of data.

[0842] "Generative AI" refers to artificial intelligence that learns from collected data and generates new stories.

[0843] A "request" refers to a user-input request for a particular historical story.

[0844] "API" stands for Application Program Interface and refers to a mechanism that allows software to communicate with each other.

[0845] "Smartphone app" refers to application software that runs on a smartphone.

[0846] This invention is a storytelling system that collects testimonies and documentary data about specific historical events and trains a generative AI to convey real experiences and emotions to future generations. This system is implemented through the following series of processing steps.

[0847] The main components of the system are data collection, preprocessing, tagging, AI learning, user request, content generation, and content provision. These functions run on the server and on the smartphone app used by users.

[0848] Data collection

[0849] The server collects testimonies and documentary data about specific historical events from publicly available information on the Internet, digital archives, and databases of universities and research institutions. For example, the server collects audio files, written records, and video content from interviews about the Great Kanto Earthquake. This allows information from a variety of data sources to be managed in a unified manner.

[0850] Data Preprocessing

[0851] The collected data is preprocessed on the server. During this process, the audio data is transcribed, noise is removed from the text data, and the notation is standardized. Specifically, an audio interview about the "Great Kanto Earthquake" is transcribed, and noisy parts and typos are corrected.

[0852] Data Tagging

[0853] The preprocessed data is then tagged on the server. These tags include metadata about the content, emotions, and specific themes of the data. For example, testimonies from immediately after the end of the war are tagged with "war," "end of the war," "hardship," and "hope." This allows for efficient data search and management.

[0854] AI Learning

[0855] Based on the tagged data, the server trains a generative AI model (e.g., GPT-4). This AI model uses the collected data to learn people's emotions and experiences regarding specific historical events, improving the quality of the stories it generates. This allows it to provide users with realistic and emotionally rich stories.

[0856] User request input and content generation

[0857] Through a smartphone app, users have an interface that allows them to request a specific historical story, for example, by typing, "I'd like to hear in detail about the experiences of victims of the Great Kanto Earthquake." Based on this request, the server instructs a generative AI model to generate a realistic story based on relevant data.

[0858] Content provider

[0859] The generated story is provided to the user in the form of text, audio, or video. For example, the server generates an audio story recreating the experiences of victims of the Great Kanto Earthquake and delivers it to the user via a smartphone app. The generated content is also delivered to the smartphone app via an API.

[0860] Examples of concrete examples and prompts

[0861] As a specific example, if a user opens the app and requests to listen to an audio account of the Great Kanto Earthquake, the AI ​​will generate a realistic story, display it in the app, convert it into audio, and play it back.

[0862] Examples of prompts include:

[0863] "What was your experience as a victim of the Great Kanto Earthquake?"

[0864] How did life change after the war ended?

[0865] "Tell me about the hope and despair you felt immediately after the end of the war."

[0866] This allows users to gain a deeper emotional connection to historical events, making it a useful tool for education and research.

[0867] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0868] Step 1:

[0869] The server collects testimonies and documentary data related to a specific historical event. In this process, data is collected from publicly available information on the Internet, digital archives, and databases of universities and research institutions. For example, it collects audio files of interviews, written records, and video content related to the "Great Kanto Earthquake." The input is a search keyword (e.g., "Great Kanto Earthquake"), and the output is the collected data.

[0870] Step 2:

[0871] The server preprocesses the collected data. This involves transcribing the audio data, removing noise from the text data, and standardizing the notation. Specifically, it converts the audio data into text using speech recognition software (e.g., Google Speech-to-Text API) and cleans the text. The input is the collected data, and the output is the preprocessed data.

[0872] Step 3:

[0873] The server then tags the preprocessed data. During this process, it adds metadata about the content, emotions, and specific themes of the data. For example, tags such as "war," "end of the war," "hardship," and "hope" are added to testimonies from immediately after the end of the war. The input is the preprocessed data, and the output is the tagged data.

[0874] Step 4:

[0875] The server trains a generative AI model using the tagged data. During this process, the dataset is input into a generative AI (e.g., GPT-4) to train the model. This allows the AI ​​to learn people's emotions and experiences regarding specific historical events and improve its story-generating capabilities. The input is the tagged data, and the output is the trained AI model.

[0876] Step 5:

[0877] Users request a specific historical story through a smartphone app. This process involves inputting search keywords and specific inquiries into the on-screen interface. For example, they might input, "I'd like to hear in detail about the experiences of victims of the Great Kanto Earthquake." The input is the user request, and the output is a prompt to the generative AI.

[0878] Step 6:

[0879] The server issues instructions to the generative AI based on the user's request and generates a story. During this process, the AI ​​model is instructed to generate a story based on the prompt. The generated story contains realistic emotions and specific episodes based on the training data. The input is the user's request prompt, and the output is the generated story.

[0880] Step 7:

[0881] The server provides the generated story to the user using an API to distribute it to a smartphone app. In this process, content is provided in the form of text, audio, or video. For example, the generated story can be sent to the app in audio format, allowing the user to listen to it. The input is the generated story, and the output is the content provided to the user.

[0882] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0883] This invention combines a system that uses generative AI to learn testimonies and document data related to historical events, and generates and provides realistic stories based on user requests, with an emotion engine that recognizes the user's emotions and reflects them in the story generation.

[0884] System action

[0885] 1. Data Collection

[0886] The server collects testimonies and documentary data about specific historical events, including audio interviews, written records, and video content from public digital archives, university and research databases, and publicly available online sources.

[0887] 2. Data Preprocessing

[0888] The server preprocesses the collected data. Specifically, it converts the audio data into text, transcribes it, and saves it as text data. It also removes noise from the text data and standardizes its format.

[0889] 3. Data tagging

[0890] The server then tags the preprocessed data. Specifically, it assigns metadata tags related to the content, emotions, and specific themes of the data. For example, testimonies from immediately after the end of the war might be tagged with "war," "end of the war," "hardship," and "hope."

[0891] 4. AI Learning

[0892] The server uses the tagged data to train a generative AI model (e.g., GPT-4) over multiple epochs to learn the content and sentiment of the data.

[0893] 5. User request input

[0894] The device provides an interface that allows users to request specific historical stories, providing text boxes and options for users to input events, dates, and specific experiences that interest them.

[0895] 6. Incorporating an Emotional Engine

[0896] The device incorporates an emotion engine to recognize the user's emotions. Specifically, it recognizes emotions from the user's input, as well as from biometric sensors, cameras, microphones, and other sensors, based on the user's facial expressions, tone of voice, heart rate, and other information.

[0897] 7. User Request Submission

[0898] The device sends the user's request and the recognized emotion data to the server. For example, if a user inputs "I would like to hear in detail about the experiences of victims of the Great Kanto Earthquake," the emotion engine will simultaneously recognize the user's emotions, such as sadness or excitement.

[0899] 8. Request Parsing and Data Preparation

[0900] The server analyzes the user request and emotion data and prepares appropriate data, which involves filtering relevant data based on the request content and extracting necessary information taking emotion data into account.

[0901] 9. Content Generation

[0902] The server uses generative AI to generate stories based on user requests. Specifically, it generates stories that recreate specific historical events or experiences that reflect the user's emotions based on the learned data. For example, it depicts the experiences of disaster victims by emphasizing sadness and hope based on emotional data.

[0903] 10. Content Provision

[0904] The device then provides the generated story to the user. The content is provided in the form of text, audio, and video, and the user can read, listen, or watch it. For example, a story recreating the experience of the Great Kanto Earthquake is played in audio format, with narration tailored to the user's emotions.

[0905] This system not only allows users to share their real experiences and emotions about historical events with future generations, but also provides customized stories based on the user's emotions, enabling efficient and effective historical transmission that leads to deeper understanding and emotional empathy.

[0906] The processing flow will be explained below.

[0907] Step 1:

[0908] The server collects testimonies and documentary data about specific historical events, using public digital archives, university and research databases, and publicly available information on the internet to retrieve interview audio files, written transcripts, video content, and more.

[0909] Step 2:

[0910] The server converts the collected audio data into text using speech recognition software, transcribing the interview audio and saving it as text data.

[0911] Step 3:

[0912] The server preprocesses the collected text data by removing noise and standardizing notation, for example, correcting typos and consolidating data in multiple formats into a standard format.

[0913] Step 4:

[0914] The server then tags the preprocessed text data, adding metadata tags related to the content, emotions, and specific themes of the data. For example, an interview about the Great Kanto Earthquake might be tagged with words like "earthquake," "damage," "reconstruction," and "hope."

[0915] Step 5:

[0916] The server uses the tagged data to train a generative AI model (e.g., GPT-4) over multiple epochs to learn the content and sentiment of the data.

[0917] Step 6:

[0918] The device provides an interface that allows users to request specific historical stories, providing text boxes and options for users to input events, dates, and specific experiences that interest them.

[0919] Step 7:

[0920] The device incorporates an emotion engine to recognize the user's emotions. It utilizes biometric sensors, cameras, and microphones to analyze the user's facial expressions, tone of voice, heart rate, etc., and recognizes the user's emotional state in real time.

[0921] Step 8:

[0922] The device sends the user's request and the recognized emotion data to the server. For example, if the user inputs "I would like to know more about the experiences of victims of the Great Kanto Earthquake," the emotion engine will recognize the user's sadness or excitement.

[0923] Step 9:

[0924] The server analyzes the user request and emotion data and prepares appropriate data, for example, filtering relevant data based on the request content and extracting necessary information taking emotion data into consideration.

[0925] Step 10:

[0926] The server uses generative AI to generate stories based on user requests. Based on the data the server has learned, it generates stories that recreate specific historical events or experiences that reflect the user's emotions. For example, based on emotional data, it could portray the experiences of disaster victims by emphasizing sadness and hope.

[0927] Step 11:

[0928] The device then provides the generated story to the user. The content is provided in the form of text, audio, and video, and the user can read, listen, or watch it. For example, a story recreating the experience of the Great Kanto Earthquake is played in audio format, with narration tailored to the user's emotions.

[0929] Example 2

[0930] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0931] In systems that use generative AI to generate realistic stories using testimonies and document data about historical events and provide them to users, there is a demand for systems that can generate stories that reflect the user's emotions, thereby gaining deeper empathy and understanding.

[0932] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0933] In this invention, the server includes means for collecting testimonies and document data related to specific historical events, means for preprocessing the collected data to remove noise and standardize the format, means for tagging the preprocessed data and adding metadata, means for training a generative AI using the tagged data, means for a user to request a specific historical story, means for recognizing the user's emotions and acquiring the emotion data, means for the generative AI to generate a story based on the user request and the emotion data, and means for providing the generated story to the user. This enables the generation of realistic and emotionally rich stories based on the user's emotions.

[0934] A "specific historical event" refers to an important occurrence or phenomenon in a particular era or region, and has historical significance.

[0935] A "testimony" is a statement or record made by an individual about a particular event or experience, and may be recorded in the form of an audio file, written document, video, or other format.

[0936] "Documentary data" refers to written data such as books, papers, articles, and official records related to a particular historical event.

[0937] "Means of collection" refers to the means of obtaining specific information using the Internet, digital archives, databases, etc.

[0938] "Preprocessing" refers to data formatting, including removing noise from the data and standardizing the format.

[0939] "Means of adding metadata by tagging" refers to a means of adding keywords and attribute information that indicate the content and characteristics of the data to the data, making subsequent searches and analysis easier.

[0940] "Generative AI" refers to artificial intelligence technology that generates natural language based on large datasets, and includes models such as GPT-4.

[0941] "Means for a user to request a particular historical story" refers to the means by which a user inputs historical stories of interest through the interface.

[0942] "Means for recognizing the user's emotions and acquiring emotional data" refers to means for detecting emotions by analyzing the user's facial expressions, tone of voice, heart rate, etc., and acquiring that data.

[0943] "Means for generating stories" refers to a means for generating stories in text, audio, or video format using generative AI based on user requests and emotional data.

[0944] The "means for providing" refers to a means for delivering the generated story to the user in a predetermined format.

[0945] This invention is a system that uses generative AI to generate realistic stories based on testimonies and document data about historical events, and provides these to users. Furthermore, it aims to gain deeper empathy and understanding by recognizing the user's emotions and generating stories based on them.

[0946] Data collection

[0947] The server collects testimonies and documentary data about specific historical events, including interview audio files, written records, and video content, from public digital archives, university and research institution databases, and publicly available information on the Internet, and stores them in dedicated storage.

[0948] Data Preprocessing

[0949] The server preprocesses the collected data. Audio data is converted to text using ASR (automatic speech recognition) software, and written and video data is converted to text using OCR (optical character recognition) tools. This text data is then denoised and standardized, and saved as a standardized dataset.

[0950] Data Tagging

[0951] The server then assigns tags to the preprocessed data, using NLP (natural language processing) tools to extract keywords based on the content and emotions of the data, and assigns tags such as "Great Kanto Earthquake," "experiences of victims," ​​"hope," and "despair."

[0952] AI Learning

[0953] The server trains a generative AI (e.g., GPT-4) on the tagged data. The training process is divided into multiple epochs, and inputs text data and emotional information into a large database so that the AI ​​can understand and apply them to generation.

[0954] User request input

[0955] The terminal provides a user input interface. Using options and text boxes, the user can enter details of historical events or experiences that interest them. For example, they can enter text such as "I would like to hear about experiences of the Great Kanto Earthquake."

[0956] Incorporating an emotion engine

[0957] The device is equipped with a device for recognizing the user's emotions using an emotion engine. It uses biometric sensors, a camera, and a microphone to analyze the user's facial expressions, tone of voice, heart rate, etc. in real time to detect whether the user is excited or sad.

[0958] User request submission

[0959] The device sends the user's request and the recognized emotional data to the server. For example, a request such as "I would like to hear in detail about the experiences of victims of the Great Kanto Earthquake" is sent to the server along with emotional data such as "sadness" or "excitement."

[0960] Request Parsing and Data Preparation

[0961] The server analyzes the user's request and emotional data to select the appropriate data, searching for testimonies and documents related to the requested historical event and filtering the most suitable ones based on the emotional data.

[0962] Content Generation

[0963] The server uses generative AI to generate content based on the selected data. Based on the emotional data, it can generate a story detailing the experiences of disaster victims, reflecting emotions such as sadness and hope.

[0964] Content provider

[0965] The device provides the generated story to the user, who can listen to it in text, audio, and video formats. Specifically, audio content recreating the experience of the Great Kanto Earthquake is provided, with narration based on emotions.

[0966] Prompt Sentence Examples

[0967] A user can send a request to the system using a prompt such as:

[0968] "I would like to hear in detail about the experiences of the victims of the Great Kanto Earthquake. I would like to know what difficulties they faced and how they overcame them, including the emotional aspects. I would also be happy if you could touch on their feelings of hope and courage."

[0969] As a result, this system enables the generation of realistic and emotionally rich stories based on the user's emotions, providing a deep understanding and empathy for historical events.

[0970] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0971] Step 1:

[0972] Data collection

[0973] The server collects testimonies and documentary data related to specific historical events from public digital archives, university databases, and research institution databases. Specifically, it downloads interview audio files, written records, video content, etc. and stores them in dedicated storage. The input is the URL or API of the digital archive or database, and the output is the collected data such as audio files, text files, and video files.

[0974] Step 2:

[0975] Data Preprocessing

[0976] The server preprocesses the collected data. Audio data is converted to text using ASR software, and written and video data is converted to text using OCR tools. Furthermore, the server denoises the text data and standardizes its format. The input is the collected audio, text, and video files, and the output is denoised and standardized formatted text data.

[0977] Step 3:

[0978] Data Tagging

[0979] The server assigns tags to the preprocessed text data. Using NLP tools, it extracts keywords and metadata based on the content and emotions of the data and assigns tags such as "Great Kanto Earthquake," "experiences of victims," ​​"hope," and "despair." The input is the preprocessed text data, and the output is tagged text data.

[0980] Step 4:

[0981] AI Learning

[0982] The server trains a generative AI (e.g., GPT-4) on tagged text data. The training process is divided into multiple epochs, and inputs text data and emotional information so that the AI ​​can understand the content and emotions of the data and apply them to generation. The input is tagged text data, and the output is a trained AI model.

[0983] Step 5:

[0984] User request input

[0985] The terminal provides an interface for users to request specific historical stories. Users use options and text boxes to input events, dates, and specific experiences of interest. The input is the request text entered by the user, and the output is the data as the user request.

[0986] Step 6:

[0987] Incorporating an emotion engine

[0988] The device uses an emotion engine to recognize the user's emotions. It uses biometric sensors, cameras, and microphones to analyze the user's facial expressions, tone of voice, heart rate, etc. in real time. The input is the user's facial expression, voice data, and heart rate data, and the output is analyzed emotion data.

[0989] Step 7:

[0990] User request submission

[0991] The terminal transmits the user's request and the recognized emotion data to the server. The input is the user's request and the emotion data, and the output is the request data transmitted to the server.

[0992] Step 8:

[0993] Request Parsing and Data Preparation

[0994] The server analyzes the user's request and emotional data and selects appropriate data. It filters relevant testimonials and materials according to the request content and extracts the most appropriate information based on the emotional data. The input is the user's request and emotional data, and the output is appropriate data for use in generation.

[0995] Step 9:

[0996] Content Generation

[0997] The server generates content using generative AI based on the selected data. Based on user requests and emotional data, it generates a story that recreates a specific historical event or experience. The input is appropriate data and emotional data, and the output is the generated story content.

[0998] Step 10:

[0999] Content provider

[1000] The device provides the generated story to the user. The content is provided in the form of text, audio, or video, and the user can read, listen, or watch it. The input is the generated story content, and the output is the content in a format that the user can view.

[1001] (Application example 2)

[1002] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1003] Conventional information systems for historical events have struggled to provide personalized stories tailored to the user's individual emotions and interests. As a result, users only receive general information, making it difficult to gain a deep understanding or empathy. Furthermore, the lack of emotion-based customization limits the user experience.

[1004] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting testimonies and document data related to a specific historical event, means for preprocessing the collected data to remove noise and standardize formats, means for tagging the preprocessed data and adding metadata, means for training a generative AI using the tagged data, means for a user to request a specific historical story, means for incorporating an emotion engine for recognizing the user's emotions, means for transmitting the recognized emotion data and the user request to the server, means for the generative AI to generate a story based on the user request and emotion data, and means for providing the generated story to the user. This makes it possible to provide a personalized, realistic story that reflects the user's emotions.

[1005] A "specific historical event" refers to an event or series of events of historical significance for which detailed records or testimonies exist.

[1006] "Testimony and documentary data" refers to eyewitness accounts, personal experiences, written records, documents, and other relevant materials relating to contemporary events and circumstances.

[1007] "Preprocessing" refers to the process of removing noise from collected data, standardizing the format, and processing data in a consistent manner to handle data of different formats.

[1008] "Tagging" refers to the process of adding metadata to data that indicates its content or sentiment.

[1009] "Generative AI" refers to an artificial intelligence model that can generate new content based on input data.

[1010] "Request" means a request or question entered by a User seeking specific information or a story.

[1011] An "emotion engine" refers to a system that analyzes biometric data such as a user's facial expression, tone of voice, and heart rate to recognize the user's current emotional state.

[1012] "Providing" refers to delivering the generated story to the user in the form of text, audio, video, etc.

[1013] This invention is a system that collects testimonies and document data related to specific historical events and generates personalized stories that reflect the user's emotions based on them. The system is configured as follows.

[1014] The server collects testimonials and documentary data about specific historical events from public digital archives, academic databases, and publicly available information on the internet, and includes various formats such as interview audio files, written records, and video content.

[1015] The collected data is pre-processed on the server, which includes converting speech to text, removing noise, and standardizing the format, allowing data of various formats to be handled consistently.

[1016] Next, the preprocessed data is given metadata tags based on content, emotion, and theme. For example, testimonies from immediately after the end of the war are given tags such as "war," "end of the war," "hardship," and "hope."

[1017] The tagged data is then fed to a generative AI (e.g., GPT-4) as training data. The AI ​​model is trained over multiple epochs to learn the content and sentiment of the data.

[1018] The user's device provides an interface for the user to request a specific historical story, with text boxes and options for the user to input events, dates, and specific experiences of interest.

[1019] The device also has a built-in emotion engine that recognizes the user's emotions using user input, biometric sensors, cameras, microphones, etc. The emotion engine analyzes the user's emotions in real time based on the user's facial expressions, tone of voice, heart rate, etc.

[1020] A user requests a specific historical story, and the device sends the request and the recognized emotion data to the server. For example, if a user requests "I want to know more about the experiences of victims of the Great Kanto Earthquake," and the emotion engine recognizes the user's sadness or excitement, this information is sent to the server.

[1021] The server prepares the necessary data based on the user request and emotion data, filters the relevant data based on the request content, and extracts the necessary information taking the emotion data into consideration.

[1022] The server uses generative AI to generate stories based on user requests. For example, it generates a story that recreates a specific historical event or experience that reflects the user's emotions based on the learned data. In this case, it depicts the experiences of disaster victims by emphasizing sadness and hope based on the emotional data.

[1023] Finally, the device provides the generated story to the user. The content is provided in the form of text, audio, and video, and the user can read, listen, or watch it. For example, a story recreating the experience of the Great Kanto Earthquake is played in audio format, with narration responding to the user's emotions.

[1024] As a concrete example, the following is an example of a prompt sentence to be input to a generative AI model:

[1025] "Generate a historical story about the Great Kanto Earthquake. Please reflect the user's emotion: sadness and shock."

[1026] This makes it possible to provide users with a personalized and realistic story that responds to their emotions, leading to a deeper understanding and emotional empathy for historical events.

[1027] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1028] Step 1:

[1029] The server collects testimonials and documentary data about a particular historical event. As input, it uses public digital archives, academic databases, and publicly available information on the Internet, and gathers data from these in the form of interview audio files, written records, video content, etc. The output is the original data stored in a variety of formats.

[1030] Step 2:

[1031] The server preprocesses the collected data. Specifically, it converts the audio data into text, transcribes it, and saves it as text data. It also removes noise from the collected text data and standardizes its format. The input is the collected audio data and text data, and the output is preprocessed text data in a unified format.

[1032] Step 3:

[1033] The server assigns tags to the preprocessed data and adds metadata. Specifically, it assigns tags related to the content, emotions, and specific themes of the data. For example, tags such as "war," "end of the war," "hardship," and "hope" are assigned to the data. The input is preprocessed text data, and the output is tagged text data.

[1034] Step 4:

[1035] The server trains a generative AI on the tagged data. Specifically, it trains a generative AI model, such as GPT-4, over multiple epochs to learn the content and sentiment of the data. The input is tagged text data, and the output is a trained generative AI model.

[1036] Step 5:

[1037] The terminal provides an interface through which users can request specific historical stories. Users use text boxes and options to input events, dates, and specific experiences of interest. The input is the user's request, which results in the generation of the requested data. The output is the user's requested data.

[1038] Step 6:

[1039] The device incorporates an emotion engine to recognize the user's emotions. Specifically, it uses biometric sensors, a camera, and a microphone to recognize emotions in real time from the user's facial expressions, tone of voice, and heart rate. The input is the user's biometric data, and the output is analyzed emotional data.

[1040] Step 7:

[1041] The device transmits the user's request and the recognized emotion data to the server. For example, if a user requests, "I want to know about the experiences of victims of the Great Kanto Earthquake," and the emotion engine recognizes the user's sadness or excitement, this information is transmitted to the server. The input is the user's request data and emotion data, and the output is data transmitted to the server.

[1042] Step 8:

[1043] The server analyzes the user request and emotion data and prepares appropriate data. It filters relevant data based on the request content and extracts necessary information taking emotion data into consideration. The input is the user's request information and emotion data, and the output is the filtered and extracted data.

[1044] Step 9:

[1045] The server uses generative AI to generate a story based on the user request. The trained AI model generates a story that reflects the user request and emotions based on the provided data. The inputs are the filtered and extracted data, user request data, and emotion data, and the output is the generated story.

[1046] Step 10:

[1047] The terminal provides the generated story to the user. The content is provided in the form of text, audio, and video, and the user can read, listen to, or watch it. For example, a story recreating the experience of the Great Kanto Earthquake is played in audio format, with narration tailored to the user's emotions. The input is the generated story, and the output is the story provided to the user.

[1048] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1049] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1050] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1051] [Fourth embodiment]

[1052] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1053] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1054] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1055] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1056] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1057] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1058] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1059] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1060] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1061] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1062] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1063] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1064] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1065] This invention is a storytelling system that collects testimonies and documentary data about historical events and trains generative AI to convey real experiences and emotions to the next generation. The system's program provides users with a real historical experience through a series of processes: data collection, preprocessing, tagging, AI learning, request processing, content generation, and provision.

[1066] System action

[1067] 1. Data Collection

[1068] The server collects testimonies and documentary data related to historical events from various data sources, including public digital archives, university and research databases, public information on the Internet, etc. For example, the server collects audio files of interviews, written transcripts, video content, etc.

[1069] 2. Data Preprocessing

[1070] The server preprocesses the collected data. Specifically, it transcribes the audio data, removes noise from the text data, and standardizes notation. For example, it transcribes an audio interview about the Great Kanto Earthquake and corrects noisy parts and typos.

[1071] 3. Data tagging

[1072] The server assigns tags to the preprocessed data. These tags include metadata about the content, emotions, and specific themes of the data. For example, testimonies from immediately after the end of the war might be tagged with "war," "end of the war," "hardship," and "hope."

[1073] 4. AI Learning

[1074] The server then uses the tagged data to train a generative AI model (e.g., GPT-4) that uses this data to learn about people's emotions and experiences regarding specific historical events, improving the quality of the stories it generates.

[1075] 5. User request input

[1076] The terminal provides an interface for users to request specific historical stories. Users input the events, periods, and specific experiences they are interested in. For example, a user might input, "I would like to hear in detail about the experiences of victims of the Great Kanto Earthquake."

[1077] 6. Content Generation

[1078] The server receives user requests and issues instructions to the generative AI, which then generates a realistic story based on relevant data. This story includes realistic emotions and specific episodes based on the training data. For example, the AI ​​recreates the experiences and emotions of victims of the Great Kanto Earthquake and generates a detailed story.

[1079] 7. Content Provision

[1080] The device provides the generated content to the user. The content can be in the form of text, audio, or video, and the user can read, listen, or watch it. For example, a story recreating the experiences of victims of the Great Kanto Earthquake can be provided to the user in audio format.

[1081] This system makes it possible to effectively convey the real experiences and emotions surrounding historical events to future generations, providing a deeper understanding that cannot be captured through records or documents alone.

[1082] The processing flow will be explained below.

[1083] Step 1:

[1084] The server collects testimonies and documentary data about specific historical events, including audio interviews, written records, and video content from public digital archives, university and research databases, and publicly available online sources.

[1085] Step 2:

[1086] The server converts the collected audio data into text. Specifically, it uses voice recognition software to transcribe the interview audio and save it as text data.

[1087] Step 3:

[1088] The server preprocesses the collected text data by removing noise and unnecessary information from the text data and standardizing the format, for example, correcting typos and consolidating data that exists in multiple formats into a standard format.

[1089] Step 4:

[1090] The server then tags the preprocessed text data. Specifically, it assigns metadata tags related to the content, emotions, and specific themes of the data. For example, testimonies from immediately after the end of the war might be tagged with "war," "end of the war," "hardship," and "hope."

[1091] Step 5:

[1092] The server uses the tagged data to train a generative AI model (e.g., GPT-4) over multiple epochs to learn the content and sentiment of the data.

[1093] Step 6:

[1094] The device provides an interface that allows users to request specific historical stories, providing text boxes and options for users to input events, dates, and specific experiences that interest them.

[1095] Step 7:

[1096] The terminal sends the user's request to the server, which then transmits the information entered by the user about the specific historical event or experience.

[1097] Step 8:

[1098] The server analyzes the user request and prepares the appropriate data, specifically filtering the relevant data based on the request content and extracting the required information.

[1099] Step 9:

[1100] The server uses generative AI to generate a story based on the user's request, specifically, a story that recreates a specific historical event or experience requested based on the learned data.

[1101] Step 10:

[1102] The device provides the generated story to the user. Specifically, the device displays or plays the story in the form of text, audio, or video, making it accessible to the user. For example, the device plays a story recreating the experience of the Great Kanto Earthquake in audio format.

[1103] Example 1

[1104] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1105] Testimonies and documentary data about historical events are often fragmented, making continuous preservation and transmission difficult. Furthermore, in order to pass on this information to future generations, it is necessary to express it as a story that includes real experiences and emotions, but current technology makes this difficult. Furthermore, in order to obtain detailed information about a specific historical event that interests a user, they must search for the appropriate information from within a vast amount of data, which is time-consuming and laborious.

[1106] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1107] In this invention, the server includes means for acquiring testimonies and document data related to specific historical events, means for preprocessing the acquired data to remove noise and standardize formats, means for tagging the preprocessed data and adding metadata, means for training a generative artificial intelligence using the tagged data, means for a user to request a specific historical story, means for the generative artificial intelligence to generate a story based on the user request, and means for providing the generated story to the user. This makes it possible to efficiently collect and process information related to historical events and generate realistic stories that meet the user's interests.

[1108] "Testimony and documentary data" is a general term for records and information left by people, either verbally or in writing, about a particular historical event.

[1109] "Means of acquisition" are the techniques and methods used to collect testimonial or documentary data from a particular data source.

[1110] "Preprocessing" is a process for improving the quality of acquired data by removing noise and standardizing the format.

[1111] "Noise removal" is the process of removing unnecessary information and errors contained in data.

[1112] "Format unification" is the process of converting data notation and formats into a unified form.

[1113] "Tagging" refers to adding labels to data that indicate its content or characteristics.

[1114] "Metadata" is supplementary information that describes the content and characteristics of data.

[1115] "Generative AI" is AI that has the ability to generate new information and content based on input data.

[1116] "Means of learning" refers to methods and techniques for training generative artificial intelligence using data.

[1117] A "user request" is an action or content of a user requesting a specific historical story or information from the system.

[1118] "Means for generating a story" refers to the techniques and methods by which generative artificial intelligence creates a new story based on a user request.

[1119] "Means of providing" refers to the technology or method for delivering the generated story to the user in an appropriate format.

[1120] "Content analysis" refers to the process of understanding a user request and preparing the appropriate data in response.

[1121] This invention is a storytelling system that collects testimonies and documentary data related to historical events and trains generative AI to convey real experiences and emotions to the next generation. This system includes a series of processes centered on servers, terminals, and users.

[1122] System Overview

[1123] Data collection

[1124] The server collects testimonies and documentary data related to historical events from various sources, including public digital archives, university and research institute databases, and public information on the Internet. For example, the server uses APIs to download data from public digital archives and web scraping technology to collect public information on the Internet. This includes audio files, written records, and video content.

[1125] Data Preprocessing

[1126] The server preprocesses the collected data. Specifically, it converts the audio data into text (using a speech recognition API, e.g., Google Cloud Speech-to-Text API), removes noise from the text data, and standardizes the format. For example, it transcribes an audio interview about the Great Kanto Earthquake, corrects typos, and standardizes the text to standard Japanese expressions. It also uses noise filtering technology.

[1127] Data Tagging

[1128] The server then tags the preprocessed data. Using natural language processing (NLP) technology, it assigns content and sentiment tags, such as "war," "end of the war," "hardship," and "hope," to the data. It also uses entity recognition technology to extract specific historical events and people's names and automatically add associated tags.

[1129] AI Learning

[1130] The server trains a generative AI model (e.g., GPT-4) on the tagged data. Using machine learning libraries such as PyTorch and TensorFlow, the data is divided into training and validation sets to train the AI ​​model. During this process, hyperparameters are adjusted and the data is shuffled to create a high-quality generative model.

[1131] User request input

[1132] The device provides an interface for users to request specific historical stories. Through a web or mobile application, users can input specific events, dates, and specific experiences. An example prompt might be, "I would like to hear in detail about the experiences of victims of the Great Kanto Earthquake of 1923. Please focus in particular on the emotions and specific events that occurred at the time."

[1133] Content Generation

[1134] The server instructs the generative AI based on the user's request, generating a realistic story based on relevant data. The generated story includes emotions and specific episodes based on the training data. For example, the AI ​​might generate a detailed account of an experience such as, "On September 1, 1923, we were hit by a huge earthquake..."

[1135] Content provider

[1136] The device then provides the generated content to the user. The content can be in the form of text, audio, or video. For example, if the generated story is provided in audio format, the device uses the Google Cloud Text-to-Speech API to convert the text into audio, which the user can read, listen to, or watch.

[1137] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1138] Step 1:

[1139] Data collection

[1140] The server collects testimonies and documentary data about historical events from various sources, including public digital archives, university and research institute databases, and publicly available information on the Internet. Specifically, it uses APIs to download data from digital archives and web scraping technology to collect public information on the Internet, such as audio files, written records, and video content.

[1141] Input: Public digital archives, databases, public information on the internet

[1142] Output: Collected testimonies and literature data

[1143] Step 2:

[1144] Data Preprocessing

[1145] The server performs preprocessing on the collected data. First, it uses a speech recognition API (e.g., Google Cloud Speech-to-Text API) to convert the audio data into text. Next, it standardizes the format of the text data and removes noise. This includes correcting typos, converting to standard Japanese expressions, and removing unnecessary information. For example, it transcribes an interview about the Great Kanto Earthquake and uses noise filtering technology to improve the quality of the audio data.

[1146] Input: Collected testimonies and literature data

[1147] Output: Preprocessed character data

[1148] Step 3:

[1149] Data Tagging

[1150] The server assigns tags to the preprocessed data that indicate its content and characteristics. Using natural language processing (NLP) technology, content and sentiment tags such as "war," "end of the war," "hardship," and "hope" are added to the data. Entity recognition technology is used for tagging, extracting specific historical events and people's names and assigning relevant tags.

[1151] Input: Preprocessed character data

[1152] Output: Tagged data

[1153] Step 4:

[1154] AI Learning

[1155] The server trains a generative AI model (e.g., GPT-4) on the tagged data. It uses machine learning libraries such as PyTorch and TensorFlow to split the data into a training set and a validation set, adjusting hyperparameters and shuffling the data to build a high-quality generative model.

[1156] Input: tagged data

[1157] Output: Trained generative AI model

[1158] Step 5:

[1159] User request input

[1160] The device provides an interface for users to request specific historical stories. Users can input specific events, dates, and specific experiences through a web or mobile application. For example, a user might request, "I'd like to hear in detail about the experiences of victims of the Great Kanto Earthquake of 1923."

[1161] Input: User request

[1162] Output: Request data

[1163] Step 6:

[1164] Content Generation

[1165] The server receives a user request and issues instructions to the generative AI, which then generates a realistic story based on the relevant data. The generated story includes realistic emotions and specific episodes based on the training data. For example, a user can input a prompt such as "Recreate the experiences of victims of the Great Kanto Earthquake of 1923," and a detailed story will be generated.

[1166] Input: Request data, trained generative AI model

[1167] Output: Generated stories

[1168] Step 7:

[1169] Content provider

[1170] The device provides the generated content to the user. The content can be in the form of text, audio, or video, and the user can read, listen, or watch it. For example, if the content is in audio format, the device converts the text to audio using the Google Cloud Text-to-Speech API and provides it to the user.

[1171] Input: Generated stories

[1172] Output: The content that is served to the user

[1173] (Application example 1)

[1174] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1175] In the fields of history education and research, it is difficult to convey to future generations the deep understanding and real experiences that cannot be captured through mere documentary materials and testimonies. Furthermore, the lack of diversity in information and realistic expressions of emotions makes it difficult for learners and researchers to form concrete images. Furthermore, there is a need to provide this information in an easily accessible format.

[1176] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1177] In this invention, the server includes means for collecting testimonies and document data related to specific historical events, means for preprocessing the collected data to remove noise and standardize the format, means for tagging the preprocessed data to add metadata, means for training a generative AI using the tagged data, means for a user to request a specific historical story, means for the generative AI to generate a story based on the user request, means for providing the generated story to the user, and means for providing an API for distributing the generated story to a smartphone app. This allows users to easily access real stories related to historical events in the form of text, audio, and video through the smartphone app, which can be used to deepen their learning and research.

[1178] Below are definitions of important terms included in the claims according to application examples.

[1179] "Specific historical events" refer to historical events associated with specific years or events, such as the Great Kanto Earthquake or war.

[1180] "Testimony and documentary data" refers to people's oral and written testimonies, as well as information recorded in books, reports, etc.

[1181] "Preprocessing" refers to the process of removing noise from collected data and standardizing its format.

[1182] A "tag" refers to identifying information that is added as metadata about the content, emotion, or specific theme of data.

[1183] "Metadata" refers to additional information about the attributes and characteristics of data.

[1184] "Generative AI" refers to artificial intelligence that learns from collected data and generates new stories.

[1185] A "request" refers to a user-input request for a particular historical story.

[1186] "API" stands for Application Program Interface and refers to a mechanism that allows software to communicate with each other.

[1187] "Smartphone app" refers to application software that runs on a smartphone.

[1188] This invention is a storytelling system that collects testimonies and documentary data about specific historical events and trains a generative AI to convey real experiences and emotions to future generations. This system is implemented through the following series of processing steps.

[1189] The main components of the system are data collection, preprocessing, tagging, AI learning, user request, content generation, and content provision. These functions run on the server and on the smartphone app used by users.

[1190] Data collection

[1191] The server collects testimonies and documentary data about specific historical events from publicly available information on the Internet, digital archives, and databases of universities and research institutions. For example, the server collects audio files, written records, and video content from interviews about the Great Kanto Earthquake. This allows information from a variety of data sources to be managed in a unified manner.

[1192] Data Preprocessing

[1193] The collected data is preprocessed on the server. During this process, the audio data is transcribed, noise is removed from the text data, and the notation is standardized. Specifically, an audio interview about the "Great Kanto Earthquake" is transcribed, and noisy parts and typos are corrected.

[1194] Data Tagging

[1195] The preprocessed data is then tagged on the server. These tags include metadata about the content, emotions, and specific themes of the data. For example, testimonies from immediately after the end of the war are tagged with "war," "end of the war," "hardship," and "hope." This allows for efficient data search and management.

[1196] AI Learning

[1197] Based on the tagged data, the server trains a generative AI model (e.g., GPT-4). This AI model uses the collected data to learn people's emotions and experiences regarding specific historical events, improving the quality of the stories it generates. This allows it to provide users with realistic and emotionally rich stories.

[1198] User request input and content generation

[1199] Through a smartphone app, users have an interface that allows them to request a specific historical story, for example, by typing, "I'd like to hear in detail about the experiences of victims of the Great Kanto Earthquake." Based on this request, the server instructs a generative AI model to generate a realistic story based on relevant data.

[1200] Content provider

[1201] The generated story is provided to the user in the form of text, audio, or video. For example, the server generates an audio story recreating the experiences of victims of the Great Kanto Earthquake and delivers it to the user via a smartphone app. The generated content is also delivered to the smartphone app via an API.

[1202] Examples of concrete examples and prompts

[1203] As a specific example, if a user opens the app and requests to listen to an audio account of the Great Kanto Earthquake, the AI ​​will generate a realistic story, display it in the app, convert it into audio, and play it back.

[1204] Examples of prompts include:

[1205] "What was your experience as a victim of the Great Kanto Earthquake?"

[1206] How did life change after the war ended?

[1207] "Tell me about the hope and despair you felt immediately after the end of the war."

[1208] This allows users to gain a deeper emotional connection to historical events, making it a useful tool for education and research.

[1209] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1210] Step 1:

[1211] The server collects testimonies and documentary data related to a specific historical event. In this process, data is collected from publicly available information on the Internet, digital archives, and databases of universities and research institutions. For example, it collects audio files of interviews, written records, and video content related to the "Great Kanto Earthquake." The input is a search keyword (e.g., "Great Kanto Earthquake"), and the output is the collected data.

[1212] Step 2:

[1213] The server preprocesses the collected data. This involves transcribing the audio data, removing noise from the text data, and standardizing the notation. Specifically, it converts the audio data into text using speech recognition software (e.g., Google Speech-to-Text API) and cleans the text. The input is the collected data, and the output is the preprocessed data.

[1214] Step 3:

[1215] The server then tags the preprocessed data. During this process, it adds metadata about the content, emotions, and specific themes of the data. For example, tags such as "war," "end of the war," "hardship," and "hope" are added to testimonies from immediately after the end of the war. The input is the preprocessed data, and the output is the tagged data.

[1216] Step 4:

[1217] The server trains a generative AI model using the tagged data. During this process, the dataset is input into a generative AI (e.g., GPT-4) to train the model. This allows the AI ​​to learn people's emotions and experiences regarding specific historical events and improve its story-generating capabilities. The input is the tagged data, and the output is the trained AI model.

[1218] Step 5:

[1219] Users request a specific historical story through a smartphone app. This process involves inputting search keywords and specific inquiries into the on-screen interface. For example, they might input, "I'd like to hear in detail about the experiences of victims of the Great Kanto Earthquake." The input is the user request, and the output is a prompt to the generative AI.

[1220] Step 6:

[1221] The server issues instructions to the generative AI based on the user's request and generates a story. During this process, the AI ​​model is instructed to generate a story based on the prompt. The generated story contains realistic emotions and specific episodes based on the training data. The input is the user's request prompt, and the output is the generated story.

[1222] Step 7:

[1223] The server provides the generated story to the user using an API to distribute it to a smartphone app. In this process, content is provided in the form of text, audio, or video. For example, the generated story can be sent to the app in audio format, allowing the user to listen to it. The input is the generated story, and the output is the content provided to the user.

[1224] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1225] This invention combines a system that uses generative AI to learn testimonies and document data related to historical events, and generates and provides realistic stories based on user requests, with an emotion engine that recognizes the user's emotions and reflects them in the story generation.

[1226] System action

[1227] 1. Data Collection

[1228] The server collects testimonies and documentary data about specific historical events, including audio interviews, written records, and video content from public digital archives, university and research databases, and publicly available online sources.

[1229] 2. Data Preprocessing

[1230] The server preprocesses the collected data. Specifically, it converts the audio data into text, transcribes it, and saves it as text data. It also removes noise from the text data and standardizes its format.

[1231] 3. Data tagging

[1232] The server then tags the preprocessed data. Specifically, it assigns metadata tags related to the content, emotions, and specific themes of the data. For example, testimonies from immediately after the end of the war might be tagged with "war," "end of the war," "hardship," and "hope."

[1233] 4. AI Learning

[1234] The server uses the tagged data to train a generative AI model (e.g., GPT-4) over multiple epochs to learn the content and sentiment of the data.

[1235] 5. User request input

[1236] The device provides an interface that allows users to request specific historical stories, providing text boxes and options for users to input events, dates, and specific experiences that interest them.

[1237] 6. Incorporating an Emotional Engine

[1238] The device incorporates an emotion engine to recognize the user's emotions. Specifically, it recognizes emotions from the user's input, as well as from biometric sensors, cameras, microphones, and other sensors, based on the user's facial expressions, tone of voice, heart rate, and other information.

[1239] 7. User Request Submission

[1240] The device sends the user's request and the recognized emotion data to the server. For example, if a user inputs "I would like to hear in detail about the experiences of victims of the Great Kanto Earthquake," the emotion engine will simultaneously recognize the user's emotions, such as sadness or excitement.

[1241] 8. Request Parsing and Data Preparation

[1242] The server analyzes the user request and emotion data and prepares appropriate data, which involves filtering relevant data based on the request content and extracting necessary information taking emotion data into account.

[1243] 9. Content Generation

[1244] The server uses generative AI to generate stories based on user requests. Specifically, it generates stories that recreate specific historical events or experiences that reflect the user's emotions based on the learned data. For example, it depicts the experiences of disaster victims by emphasizing sadness and hope based on emotional data.

[1245] 10. Content Provision

[1246] The device then provides the generated story to the user. The content is provided in the form of text, audio, and video, and the user can read, listen, or watch it. For example, a story recreating the experience of the Great Kanto Earthquake is played in audio format, with narration tailored to the user's emotions.

[1247] This system not only allows users to share their real experiences and emotions about historical events with future generations, but also provides customized stories based on the user's emotions, enabling efficient and effective historical transmission that leads to deeper understanding and emotional empathy.

[1248] The processing flow will be explained below.

[1249] Step 1:

[1250] The server collects testimonies and documentary data about specific historical events, using public digital archives, university and research databases, and publicly available information on the internet to retrieve interview audio files, written transcripts, video content, and more.

[1251] Step 2:

[1252] The server converts the collected audio data into text using speech recognition software, transcribing the interview audio and saving it as text data.

[1253] Step 3:

[1254] The server preprocesses the collected text data by removing noise and standardizing notation, for example, correcting typos and consolidating data in multiple formats into a standard format.

[1255] Step 4:

[1256] The server then tags the preprocessed text data, adding metadata tags related to the content, emotions, and specific themes of the data. For example, an interview about the Great Kanto Earthquake might be tagged with words like "earthquake," "damage," "reconstruction," and "hope."

[1257] Step 5:

[1258] The server uses the tagged data to train a generative AI model (e.g., GPT-4) over multiple epochs to learn the content and sentiment of the data.

[1259] Step 6:

[1260] The device provides an interface that allows users to request specific historical stories, providing text boxes and options for users to input events, dates, and specific experiences that interest them.

[1261] Step 7:

[1262] The device incorporates an emotion engine to recognize the user's emotions. It utilizes biometric sensors, cameras, and microphones to analyze the user's facial expressions, tone of voice, heart rate, etc., and recognizes the user's emotional state in real time.

[1263] Step 8:

[1264] The device sends the user's request and the recognized emotion data to the server. For example, if the user inputs "I would like to know more about the experiences of victims of the Great Kanto Earthquake," the emotion engine will recognize the user's sadness or excitement.

[1265] Step 9:

[1266] The server analyzes the user request and emotion data and prepares appropriate data, for example, filtering relevant data based on the request content and extracting necessary information taking emotion data into consideration.

[1267] Step 10:

[1268] The server uses generative AI to generate stories based on user requests. Based on the data the server has learned, it generates stories that recreate specific historical events or experiences that reflect the user's emotions. For example, based on emotional data, it could portray the experiences of disaster victims by emphasizing sadness and hope.

[1269] Step 11:

[1270] The device then provides the generated story to the user. The content is provided in the form of text, audio, and video, and the user can read, listen, or watch it. For example, a story recreating the experience of the Great Kanto Earthquake is played in audio format, with narration tailored to the user's emotions.

[1271] Example 2

[1272] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1273] In systems that use generative AI to generate realistic stories using testimonies and document data about historical events and provide them to users, there is a demand for systems that can generate stories that reflect the user's emotions, thereby gaining deeper empathy and understanding.

[1274] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1275] In this invention, the server includes means for collecting testimonies and document data related to specific historical events, means for preprocessing the collected data to remove noise and standardize the format, means for tagging the preprocessed data and adding metadata, means for training a generative AI using the tagged data, means for a user to request a specific historical story, means for recognizing the user's emotions and acquiring the emotion data, means for the generative AI to generate a story based on the user request and the emotion data, and means for providing the generated story to the user. This enables the generation of realistic and emotionally rich stories based on the user's emotions.

[1276] A "specific historical event" refers to an important occurrence or phenomenon in a particular era or region, and has historical significance.

[1277] A "testimony" is a statement or record made by an individual about a particular event or experience, and may be recorded in the form of an audio file, written document, video, or other format.

[1278] "Documentary data" refers to written data such as books, papers, articles, and official records related to a particular historical event.

[1279] "Means of collection" refers to the means of obtaining specific information using the Internet, digital archives, databases, etc.

[1280] "Preprocessing" refers to data formatting, including removing noise from the data and standardizing the format.

[1281] "Means of adding metadata by tagging" refers to a means of adding keywords and attribute information that indicate the content and characteristics of the data to the data, making subsequent searches and analysis easier.

[1282] "Generative AI" refers to artificial intelligence technology that generates natural language based on large datasets, and includes models such as GPT-4.

[1283] "Means for a user to request a particular historical story" refers to the means by which a user inputs historical stories of interest through the interface.

[1284] "Means for recognizing the user's emotions and acquiring emotional data" refers to means for detecting emotions by analyzing the user's facial expressions, tone of voice, heart rate, etc., and acquiring that data.

[1285] "Means for generating stories" refers to a means for generating stories in text, audio, or video format using generative AI based on user requests and emotional data.

[1286] The "means for providing" refers to a means for delivering the generated story to the user in a predetermined format.

[1287] This invention is a system that uses generative AI to generate realistic stories based on testimonies and document data about historical events, and provides these to users. Furthermore, it aims to gain deeper empathy and understanding by recognizing the user's emotions and generating stories based on them.

[1288] Data collection

[1289] The server collects testimonies and documentary data about specific historical events, including interview audio files, written records, and video content, from public digital archives, university and research institution databases, and publicly available information on the Internet, and stores them in dedicated storage.

[1290] Data Preprocessing

[1291] The server preprocesses the collected data. Audio data is converted to text using ASR (automatic speech recognition) software, and written and video data is converted to text using OCR (optical character recognition) tools. This text data is then denoised and standardized, and saved as a standardized dataset.

[1292] Data Tagging

[1293] The server then assigns tags to the preprocessed data, using NLP (natural language processing) tools to extract keywords based on the content and emotions of the data, and assigns tags such as "Great Kanto Earthquake," "experiences of victims," ​​"hope," and "despair."

[1294] AI Learning

[1295] The server trains a generative AI (e.g., GPT-4) on the tagged data. The training process is divided into multiple epochs, and inputs text data and emotional information into a large database so that the AI ​​can understand and apply them to generation.

[1296] User request input

[1297] The terminal provides a user input interface. Using options and text boxes, the user can enter details of historical events or experiences that interest them. For example, they can enter text such as "I would like to hear about experiences of the Great Kanto Earthquake."

[1298] Incorporating an emotion engine

[1299] The device is equipped with a device for recognizing the user's emotions using an emotion engine. It uses biometric sensors, a camera, and a microphone to analyze the user's facial expressions, tone of voice, heart rate, etc. in real time to detect whether the user is excited or sad.

[1300] User request submission

[1301] The device sends the user's request and the recognized emotional data to the server. For example, a request such as "I would like to hear in detail about the experiences of victims of the Great Kanto Earthquake" is sent to the server along with emotional data such as "sadness" or "excitement."

[1302] Request Parsing and Data Preparation

[1303] The server analyzes the user's request and emotional data to select the appropriate data, searching for testimonies and documents related to the requested historical event and filtering the most suitable ones based on the emotional data.

[1304] Content Generation

[1305] The server uses generative AI to generate content based on the selected data. Based on the emotional data, it can generate a story detailing the experiences of disaster victims, reflecting emotions such as sadness and hope.

[1306] Content provider

[1307] The device provides the generated story to the user, who can listen to it in text, audio, and video formats. Specifically, audio content recreating the experience of the Great Kanto Earthquake is provided, with narration based on emotions.

[1308] Prompt Sentence Examples

[1309] A user can send a request to the system using a prompt such as:

[1310] "I would like to hear in detail about the experiences of the victims of the Great Kanto Earthquake. I would like to know what difficulties they faced and how they overcame them, including the emotional aspects. I would also be happy if you could touch on their feelings of hope and courage."

[1311] As a result, this system enables the generation of realistic and emotionally rich stories based on the user's emotions, providing a deep understanding and empathy for historical events.

[1312] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1313] Step 1:

[1314] Data collection

[1315] The server collects testimonies and documentary data related to specific historical events from public digital archives, university databases, and research institution databases. Specifically, it downloads interview audio files, written records, video content, etc. and stores them in dedicated storage. The input is the URL or API of the digital archive or database, and the output is the collected data such as audio files, text files, and video files.

[1316] Step 2:

[1317] Data Preprocessing

[1318] The server preprocesses the collected data. Audio data is converted to text using ASR software, and written and video data is converted to text using OCR tools. Furthermore, the server denoises the text data and standardizes its format. The input is the collected audio, text, and video files, and the output is denoised and standardized formatted text data.

[1319] Step 3:

[1320] Data Tagging

[1321] The server assigns tags to the preprocessed text data. Using NLP tools, it extracts keywords and metadata based on the content and emotions of the data and assigns tags such as "Great Kanto Earthquake," "experiences of victims," ​​"hope," and "despair." The input is the preprocessed text data, and the output is tagged text data.

[1322] Step 4:

[1323] AI Learning

[1324] The server trains a generative AI (e.g., GPT-4) on tagged text data. The training process is divided into multiple epochs, and inputs text data and emotional information so that the AI ​​can understand the content and emotions of the data and apply them to generation. The input is tagged text data, and the output is a trained AI model.

[1325] Step 5:

[1326] User request input

[1327] The terminal provides an interface for users to request specific historical stories. Users use options and text boxes to input events, dates, and specific experiences of interest. The input is the request text entered by the user, and the output is the data as the user request.

[1328] Step 6:

[1329] Incorporating an emotion engine

[1330] The device uses an emotion engine to recognize the user's emotions. It uses biometric sensors, cameras, and microphones to analyze the user's facial expressions, tone of voice, heart rate, etc. in real time. The input is the user's facial expression, voice data, and heart rate data, and the output is analyzed emotion data.

[1331] Step 7:

[1332] User request submission

[1333] The terminal transmits the user's request and the recognized emotion data to the server. The input is the user's request and the emotion data, and the output is the request data transmitted to the server.

[1334] Step 8:

[1335] Request Parsing and Data Preparation

[1336] The server analyzes the user's request and emotional data and selects appropriate data. It filters relevant testimonials and materials according to the request content and extracts the most appropriate information based on the emotional data. The input is the user's request and emotional data, and the output is appropriate data for use in generation.

[1337] Step 9:

[1338] Content Generation

[1339] The server generates content using generative AI based on the selected data. Based on user requests and emotional data, it generates a story that recreates a specific historical event or experience. The input is appropriate data and emotional data, and the output is the generated story content.

[1340] Step 10:

[1341] Content provider

[1342] The device provides the generated story to the user. The content is provided in the form of text, audio, or video, and the user can read, listen, or watch it. The input is the generated story content, and the output is the content in a format that the user can view.

[1343] (Application example 2)

[1344] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1345] Conventional information systems for historical events have struggled to provide personalized stories tailored to the user's individual emotions and interests. As a result, users only receive general information, making it difficult to gain a deep understanding or empathy. Furthermore, the lack of emotion-based customization limits the user experience.

[1346] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting testimonies and document data related to a specific historical event, means for preprocessing the collected data to remove noise and standardize formats, means for tagging the preprocessed data and adding metadata, means for training a generative AI using the tagged data, means for a user to request a specific historical story, means for incorporating an emotion engine for recognizing the user's emotions, means for transmitting the recognized emotion data and the user request to the server, means for the generative AI to generate a story based on the user request and emotion data, and means for providing the generated story to the user. This makes it possible to provide a personalized, realistic story that reflects the user's emotions.

[1347] A "specific historical event" refers to an event or series of events of historical significance for which detailed records or testimonies exist.

[1348] "Testimony and documentary data" refers to eyewitness accounts, personal experiences, written records, documents, and other relevant materials relating to contemporary events and circumstances.

[1349] "Preprocessing" refers to the process of removing noise from collected data, standardizing the format, and processing data in a consistent manner to handle data of different formats.

[1350] "Tagging" refers to the process of adding metadata to data that indicates its content or sentiment.

[1351] "Generative AI" refers to an artificial intelligence model that can generate new content based on input data.

[1352] "Request" means a request or question entered by a User seeking specific information or a story.

[1353] An "emotion engine" refers to a system that analyzes biometric data such as a user's facial expression, tone of voice, and heart rate to recognize the user's current emotional state.

[1354] "Providing" refers to delivering the generated story to the user in the form of text, audio, video, etc.

[1355] This invention is a system that collects testimonies and document data related to specific historical events and generates personalized stories that reflect the user's emotions based on them. The system is configured as follows.

[1356] The server collects testimonials and documentary data about specific historical events from public digital archives, academic databases, and publicly available information on the internet, and includes various formats such as interview audio files, written records, and video content.

[1357] The collected data is pre-processed on the server, which includes converting speech to text, removing noise, and standardizing the format, allowing data of various formats to be handled consistently.

[1358] Next, the preprocessed data is given metadata tags based on content, emotion, and theme. For example, testimonies from immediately after the end of the war are given tags such as "war," "end of the war," "hardship," and "hope."

[1359] The tagged data is then fed to a generative AI (e.g., GPT-4) as training data. The AI ​​model is trained over multiple epochs to learn the content and sentiment of the data.

[1360] The user's device provides an interface for the user to request a specific historical story, with text boxes and options for the user to input events, dates, and specific experiences of interest.

[1361] The device also has a built-in emotion engine that recognizes the user's emotions using user input, biometric sensors, cameras, microphones, etc. The emotion engine analyzes the user's emotions in real time based on the user's facial expressions, tone of voice, heart rate, etc.

[1362] A user requests a specific historical story, and the device sends the request and the recognized emotion data to the server. For example, if a user requests "I want to know more about the experiences of victims of the Great Kanto Earthquake," and the emotion engine recognizes the user's sadness or excitement, this information is sent to the server.

[1363] The server prepares the necessary data based on the user request and emotion data, filters the relevant data based on the request content, and extracts the necessary information taking the emotion data into consideration.

[1364] The server uses generative AI to generate stories based on user requests. For example, it generates a story that recreates a specific historical event or experience that reflects the user's emotions based on the learned data. In this case, it depicts the experiences of disaster victims by emphasizing sadness and hope based on the emotional data.

[1365] Finally, the device provides the generated story to the user. The content is provided in the form of text, audio, and video, and the user can read, listen, or watch it. For example, a story recreating the experience of the Great Kanto Earthquake is played in audio format, with narration responding to the user's emotions.

[1366] As a concrete example, the following is an example of a prompt sentence to be input to a generative AI model:

[1367] "Generate a historical story about the Great Kanto Earthquake. Please reflect the user's emotion: sadness and shock."

[1368] This makes it possible to provide users with a personalized and realistic story that responds to their emotions, leading to a deeper understanding and emotional empathy for historical events.

[1369] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1370] Step 1:

[1371] The server collects testimonials and documentary data about a particular historical event. As input, it uses public digital archives, academic databases, and publicly available information on the Internet, and gathers data from these in the form of interview audio files, written records, video content, etc. The output is the original data stored in a variety of formats.

[1372] Step 2:

[1373] The server preprocesses the collected data. Specifically, it converts the audio data into text, transcribes it, and saves it as text data. It also removes noise from the collected text data and standardizes its format. The input is the collected audio data and text data, and the output is preprocessed text data in a unified format.

[1374] Step 3:

[1375] The server assigns tags to the preprocessed data and adds metadata. Specifically, it assigns tags related to the content, emotions, and specific themes of the data. For example, tags such as "war," "end of the war," "hardship," and "hope" are assigned to the data. The input is preprocessed text data, and the output is tagged text data.

[1376] Step 4:

[1377] The server trains a generative AI on the tagged data. Specifically, it trains a generative AI model, such as GPT-4, over multiple epochs to learn the content and sentiment of the data. The input is tagged text data, and the output is a trained generative AI model.

[1378] Step 5:

[1379] The terminal provides an interface through which users can request specific historical stories. Users use text boxes and options to input events, dates, and specific experiences of interest. The input is the user's request, which results in the generation of the requested data. The output is the user's requested data.

[1380] Step 6:

[1381] The device incorporates an emotion engine to recognize the user's emotions. Specifically, it uses biometric sensors, a camera, and a microphone to recognize emotions in real time from the user's facial expressions, tone of voice, and heart rate. The input is the user's biometric data, and the output is analyzed emotional data.

[1382] Step 7:

[1383] The device transmits the user's request and the recognized emotion data to the server. For example, if a user requests, "I want to know about the experiences of victims of the Great Kanto Earthquake," and the emotion engine recognizes the user's sadness or excitement, this information is transmitted to the server. The input is the user's request data and emotion data, and the output is data transmitted to the server.

[1384] Step 8:

[1385] The server analyzes the user request and emotion data and prepares appropriate data. It filters relevant data based on the request content and extracts necessary information taking emotion data into consideration. The input is the user's request information and emotion data, and the output is the filtered and extracted data.

[1386] Step 9:

[1387] The server uses generative AI to generate a story based on the user request. The trained AI model generates a story that reflects the user request and emotions based on the provided data. The inputs are the filtered and extracted data, user request data, and emotion data, and the output is the generated story.

[1388] Step 10:

[1389] The terminal provides the generated story to the user. The content is provided in the form of text, audio, and video, and the user can read, listen to, or watch it. For example, a story recreating the experience of the Great Kanto Earthquake is played in audio format, with narration tailored to the user's emotions. The input is the generated story, and the output is the story provided to the user.

[1390] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1391] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1392] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1393] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1394] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1395] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1396] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1397] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1398] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1399] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1400] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1401] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1402] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1403] 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.

[1404] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1405] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1406] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1407] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1408] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1409] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1410] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1411] The following is further disclosed regarding the above embodiment.

[1412] (Claim 1)

[1413] means of collecting testimonies and documentary data about specific historical events;

[1414] A means of preprocessing the collected data to remove noise and standardize the format;

[1415] a means of tagging and adding metadata to the preprocessed data;

[1416] A means to train generative AI using tagged data; and

[1417] a means for users to request specific historical stories;

[1418] A means for generative AI to generate stories based on user requests;

[1419] a means for providing the generated story to a user;

[1420] A system including:

[1421] (Claim 2)

[1422] 10. The system of claim 1, further comprising means for analyzing the content of the user request and preparing appropriate data.

[1423] (Claim 3)

[1424] 10. The system of claim 1, further comprising means for providing the generated story in one of text, audio, and video formats.

[1425] "Example 1"

[1426] (Claim 1)

[1427] means of obtaining testimonials and documentary data on specific historical events;

[1428] means for preprocessing the acquired data to remove noise and standardize the format;

[1429] a means of tagging and adding metadata to the preprocessed data;

[1430] A means for training a generative artificial intelligence using the tagged data; and

[1431] a means for users to request specific historical stories;

[1432] A means for a generative artificial intelligence to generate a story based on a user request;

[1433] a means for providing the generated story to a user;

[1434] A system including:

[1435] (Claim 2)

[1436] 10. The system of claim 1, further comprising means for analyzing the content of the user request and preparing appropriate data.

[1437] (Claim 3)

[1438] 10. The system of claim 1, further comprising means for providing the generated story in one of text, audio, and video formats.

[1439] "Application Example 1"

[1440] (Claim 1)

[1441] means of collecting testimonies and documentary data about specific historical events;

[1442] A means of preprocessing the collected data to remove noise and standardize the format;

[1443] a means of tagging and adding metadata to the preprocessed data;

[1444] A means to train generative AI using tagged data; and

[1445] a means for users to request specific historical stories;

[1446] A means for generative AI to generate stories based on user requests;

[1447] a means for providing the generated story to a user;

[1448] A means of providing an API for distributing the generated stories to smartphone apps;

[1449] A system including:

[1450] (Claim 2)

[1451] 10. The system of claim 1, further comprising means for analyzing the content of the user request and preparing appropriate data.

[1452] (Claim 3)

[1453] 10. The system of claim 1, further comprising means for providing the generated story in one of text, audio, and video formats.

[1454] "Example 2: Combining Emotion Engines"

[1455] (Claim 1)

[1456] means of collecting testimonies and documentary data about specific historical events;

[1457] A means of preprocessing the collected data to remove noise and standardize the format;

[1458] a means of tagging and adding metadata to the preprocessed data;

[1459] A means to train generative AI using tagged data; and

[1460] a means for users to request specific historical stories;

[1461] means for recognizing a user's emotion and acquiring emotion data;

[1462] A means for the generative AI to generate a story based on user requests and emotional data; and

[1463] a means for providing the generated story to a user;

[1464] A system including:

[1465] (Claim 2)

[1466] 10. The system of claim 1, further comprising means for analyzing the content of the user request and the emotion data to prepare appropriate data.

[1467] (Claim 3)

[1468] 10. The system of claim 1, further comprising means for providing the generated story in one of text, audio, and video formats.

[1469] "Application example 2 when combining emotion engines"

[1470] (Claim 1)

[1471] means of collecting testimonies and documentary data about specific historical events;

[1472] A means of preprocessing the collected data to remove noise and standardize the format;

[1473] a means of tagging and adding metadata to the preprocessed data;

[1474] A means to train generative AI using tagged data; and

[1475] a means for users to request specific historical stories;

[1476] means for incorporating an emotion engine for recognizing user emotions;

[1477] means for transmitting the recognized emotion data and the user request to a server;

[1478] A means for generative AI to generate stories based on user requests and emotional data;

[1479] a means for providing the generated story to a user;

[1480] A system including:

[1481] (Claim 2)

[1482] 10. The system of claim 1, further comprising means for analyzing the content of the user request and preparing appropriate data.

[1483] (Claim 3)

[1484] 10. The system of claim 1, further comprising means for providing the generated story in one of text, audio, and video formats. [Explanation of symbols]

[1485] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means of collecting testimonies and documentary data about specific historical events; A means of preprocessing the collected data to remove noise and standardize the format; a means of tagging and adding metadata to the preprocessed data; A means to train generative AI using tagged data; and a means for users to request specific historical stories; A means for generative AI to generate stories based on user requests; a means for providing the generated story to a user; A system including:

2. 10. The system of claim 1, further comprising means for analyzing the content of the user request and preparing appropriate data.

3. 10. The system of claim 1, further comprising means for providing the generated story in one of the following formats: text, audio, and video.

Images