System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving input information of a user; means for analyzing an intent of the user from the received input information; means for automatically generating an appropriate prompt based on the analyzed intent; means for sending the generated prompt to a generative model; means for receiving a response from the generative model; and means for displaying the received response to the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] To obtain the desired output using conventional generative models, users must enter detailed and appropriate prompts. However, creating these prompts is time-consuming and requires technical skills. This prevents many users from fully utilizing the benefits of generative models. Furthermore, when input information comes in various formats, such as audio and images, the system must be able to accurately interpret them, but conventional technologies often fall short in this area. [Means for solving the problem]

[0005] The present invention provides a system including means for receiving user input information, means for analyzing the user's intention from the received input information, means for automatically generating appropriate prompts based on the analyzed intention, means for sending the generated prompts to a generative model, means for receiving responses from the generative model, and means for displaying the received responses to the user. This system eliminates the need for a user to manually input detailed prompts, allowing the user to effectively utilize the generative model through dialogue and icon manipulation alone. The system can also appropriately process input information in a variety of formats, such as voice, text, and icon manipulation, and provide the user with responses in text, image, and video data from the generative model.

[0006] "User input information" refers to information that a user provides to the system, and includes voice, text, icon operations, and the like.

[0007] "Means for receiving" refers to a function for acquiring information input by a user and incorporating it into the system.

[0008] "Means for analyzing" refers to a function that uses natural language processing or other analytical techniques to understand the user's purpose and intent from received input information.

[0009] "Means for automatic generation" refers to a function that automatically generates appropriate instructions (prompts) for obtaining the output desired by the user based on the analyzed intent.

[0010] The "means for sending" refers to a function for transferring the generated prompt to provide it as input to the Generative Model.

[0011] A "generative model" refers to an artificial intelligence model that generates output such as text, images, and videos based on user intent.

[0012] "Means for receiving (response from generative model)" refers to the function of obtaining output data returned from the generative model.

[0013] "Display means" refers to a function for presenting the response from the generative model to the user in an easy-to-understand manner. [Brief explanation of the drawings]

[0014] [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

[0015] 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.

[0016] First, the terms used in the following description will be explained.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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."

[0022] [First embodiment]

[0023] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0024] 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.

[0025] 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).

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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.

[0030] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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."

[0035] The present invention relates to a system that utilizes a generative model using user input information, and a specific embodiment thereof will be described below. The system consists of three elements: a server, a terminal, and a user.

[0036] System configuration and operation

[0037] User Input

[0038] A user accesses the system using a device (e.g., a smartphone, tablet, or PC). At this stage, the user can select from a variety of input methods, such as voice recognition, text input, and icon manipulation, to input requests to the generative model. For example, a user may input in text format, "I would like to create a summary of the planning meeting," or make a similar request by voice.

[0039] Providing an interface

[0040] The terminal provides an interface for the user to provide input information, which may include a user interface including icons, microphone access permission for voice input, text input fields, etc. When the user inputs information through these interfaces, the information is transmitted to the server in real time.

[0041] Analyzing user intent

[0042] The server analyzes the user's input information received from the device. Using natural language processing (NLP) technology, it analyzes the user's text and voice data to understand the user's intent. This analysis allows the server to determine what the user is looking for and generate appropriate prompts based on that. As a result of this analysis, the server may generate a prompt such as "Summary of the planning meeting, including key points."

[0043] Automatically Generated Prompts

[0044] Based on the analyzed user intent, the server automatically generates a prompt containing the necessary information. For example, if a user inputs "Please create a summary of a product review," the server generates a prompt "Summary of a product review, including key points." This prompt is sent to the generative model.

[0045] Prompting the Generative Model

[0046] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques (e.g., a Generative Pre-trained Transformer) to generate output based on the user's prompts. This output can include text, images, video data, etc.

[0047] Receiving a response from the generative model

[0048] The results generated by the generative model based on the specified prompts are sent back to the server. For example, if the generative model generates a "summary of the planning meeting" based on the user's request, the summary is sent to the server.

[0049] Displaying the results

[0050] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. This step allows the user to easily check the generated output, such as text, images, and videos. For example, a summary of a planning meeting generated can be displayed on the screen, allowing the user to confirm the response to their request.

[0051] Example: A series of steps for generating sentences

[0052] Consider the case where a user voice-inputs "Please create a summary of a product review" into a device. This voice input is received by the device and sent to the server. The server converts the voice data into text and analyzes the user's intent through natural language processing. Based on the results, it automatically generates a prompt "Summary of a product review, including key points" and sends it to the generative model. The generative model generates a product review summary based on this prompt and sends it back to the server. Finally, the server sends the received summary to the user's device, which displays it to the user.

[0053] As described above, the present invention provides a system for easily and efficiently using generative models, eliminating the need for users to manually input detailed prompts.

[0054] The processing flow will be explained below.

[0055] Step 1:

[0056] A user accesses a terminal.

[0057] Action: A user launches an application on the system using a device such as a smartphone or PC.

[0058] Step 2:

[0059] The user provides the input information.

[0060] Action: User clicks icon or speaks or writes "I want to create a summary of the planning meeting."

[0061] Step 3:

[0062] The terminal receives the input information.

[0063] How it works: The device receives the user's clicks, voice, and text input in real time and transmits that information to the server.

[0064] Step 4:

[0065] The server analyzes the input information.

[0066] How it works: The server analyzes the data received from the device, converts voice data into text, and uses natural language processing (NLP) techniques to understand the user's intent.

[0067] Step 5:

[0068] The server generates the appropriate prompt.

[0069] How it works: Based on the analysis results, the server automatically generates a prompt such as "Summary of the planning meeting, including key points."

[0070] Step 6:

[0071] The server sends a prompt to the generative model.

[0072] How it works: The server sends the generated prompt as an HTTP request to the API endpoint of the generating AI.

[0073] Step 7:

[0074] The server receives the response from the generative model.

[0075] How it works: The generative AI model generates output based on the prompts, and the results are sent back to the server.

[0076] Step 8:

[0077] The server sends the results to the terminal.

[0078] How it works: The server formats the generated output and creates data packets to send to the user's terminal.

[0079] Step 9:

[0080] The terminal displays the results to the user.

[0081] Operation: The device visualizes the received data and presents it to the user. For example, it displays a summary of the planning meeting on the screen.

[0082] Examples:

[0083] 1. The user speaks, "I want a summary of a product review."

[0084] 2. The device sends the audio data to the server.

[0085] 3. The server transcribes the audio data, performs natural language processing, and generates prompts.

[0086] 4. The server sends a prompt to the generative model.

[0087] 5. The generative model generates a summary of the review and sends it back to the server.

[0088] 6. The server sends the results to the user's terminal, which displays a summary to the user.

[0089] In this way, the system automatically performs analysis and generation without the user having to enter complex prompts, making generative models easier to use.

[0090] Example 1

[0091] 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."

[0092] When users use complex generative models, many detailed operations are required, which places a significant burden on them. In particular, when using voice or text input, users must manually input appropriate prompts into the generative model, which is a time-consuming and labor-intensive task. A solution to this problem is needed that allows users to easily use generative models.

[0093] 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.

[0094] In this invention, the server includes a means for converting voice data into text data, a means for analyzing the converted text data using natural language processing technology, and a means for formatting the generated prompt into a format that is easy for the generative model to understand, thereby enabling the user to use the generative model simply and quickly without performing complex operations.

[0095] "User input information" refers to data that a user provides to the system in the form of voice, text, icon manipulation, or the like.

[0096] "Means for receiving" refers to a function for receiving information input by a user and incorporating it into the system.

[0097] "Means for analyzing" refers to methods and techniques for processing received user input information and understanding its content.

[0098] A "prompt" is a short instruction that specifically specifies the desired output from the generative model.

[0099] A "generative model" refers to an artificial intelligence system that generates text, images, video, etc. based on a given prompt.

[0100] "Means of conversion" refers to the methods and techniques for converting audio data into text data.

[0101] "Natural language processing technology" refers to a technological system that enables computers to understand and analyze human language and perform appropriate information processing.

[0102] "Formatting means" refers to a method for adjusting the format of the generated prompt so that the generative model can efficiently understand it and generate a response.

[0103] A "response" refers to the output that a generative model generates based on a prompt.

[0104] "Means for displaying" refers to a function for visually or audibly presenting the response received from the generative model to the user.

[0105] The present invention relates to a system that utilizes a generative model based on user input information. A specific embodiment of this system is described below. The system consists of three elements: a server, a terminal, and a user.

[0106] System configuration and operation

[0107] User Input

[0108] Users access the system using devices such as smartphones, tablets, and PCs. When sending requests to the system, users can choose from a variety of input methods, such as voice recognition, text input, and icon manipulation. For example, a user may input a request in text format, such as "Please create a product review summary." Alternatively, a user may make a similar request by voice.

[0109] Providing an interface

[0110] The terminal provides an interface for the user to provide input information, including a text input field, a user interface including icons, and microphone access permission for voice input. When the user inputs information through these interfaces, the information is transmitted to the server in real time.

[0111] Analyzing User Intent

[0112] The server analyzes the user's input information sent from the device. It uses natural language processing (NLP) techniques to analyze text and voice data and understand the user's intent. For example, it uses an NLP engine (e.g., a common NLP library or API) to identify that the user's request is to "create a summary of a product review."

[0113] Automatically Generated Prompts

[0114] Based on the analysis results, the server automatically generates a prompt containing the necessary information. For example, it generates a prompt such as "A summary of product reviews, including key points." This prompt is formatted in a way that is easy for the generative model to understand.

[0115] Prompting the Generative Model

[0116] The server sends the generated prompts to a generative model (e.g., an artificial intelligence system such as a Generative Pre-trained Transformer) that uses the prompts to generate output for the user's request, such as text, images, or video. The generative model uses advanced AI algorithms to generate the requested data in real time.

[0117] Receiving a response from the generative model

[0118] The output generated by the generative model is sent back to the server. For example, a "product review summary" is generated and sent to the server. This generated result is sent as structured data, such as text.

[0119] Displaying the results

[0120] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. The device displays the generated text, image, or video output to the user on a screen or using a voice assistant. For example, a generated "product review summary" is displayed on the device screen.

[0121] Specific examples

[0122] A specific example is a case where the user vocally inputs into the terminal, "I want you to create a review summary of the product."

[0123] 1. User input:

[0124] The user speaks to the terminal, "I want a review summary of the product."

[0125] 2. Receiving audio data:

[0126] The device provides microphone access permission for voice input and transmits the received voice data to the server, for example, when the user makes a request to "activate the voice assistant."

[0127] 3. Text Conversion and Intent Analysis:

[0128] The server converts the voice data into text and analyzes the user's intent using natural language processing technology. The server converts the voice data into text using the Google® Speech-to-Text API, and then analyzes the intent using a common NLP engine.

[0129] 4. Generate prompt:

[0130] The server generates a prompt such as "Product review summary, including key points" and sends this prompt to the generative model.

[0131] 5. Generative model response:

[0132] A generative model (e.g., GPT-3®) generates a review summary based on the prompts and sends the summary back to the server.

[0133] 6. View the response:

[0134] The server sends the summary to the device, which then displays it to the user. The device screen displays the review summary: "I found this product easy to use and very durable. I was particularly impressed by its light weight."

[0135] By following the above steps, users can use generative AI models easily and efficiently.

[0136] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0137] Step 1: User Input

[0138] Users access the system using devices such as smartphones, tablets, and PCs. For example, a user inputs a request in text format, such as "Please create a product review summary." The user's input is entered into the device as text data, such as "Please create a product review summary." This input data is sent to the server through the device's interface.

[0139] Step 2: Providing an interface

[0140] The device provides a user interface that includes text input fields, icons, and microphone access permission for voice input. When a user inputs information through these interfaces, the information is transmitted to the server in real time. This transmitted data becomes the input information required for the next analysis step.

[0141] Step 3: Analyze user intent

[0142] The server analyzes the user's input information received from the terminal. Specifically, it analyzes the text data using natural language processing technology (e.g., a standard NLP engine). The server receives the text data "Please create a product review summary" as input and analyzes this data to identify the user's intent. The output of the analysis is the intent to "generate a product review summary."

[0143] Step 4: Automatically generate prompts

[0144] The server automatically generates a prompt containing the necessary information based on the analyzed user intent. For example, it generates a prompt such as "Product review summary, including important points." The prompt generated by the server is text data such as "Product review summary, including important points." This prompt is sent as input to the generative model.

[0145] Step 5: Sending prompts to the generative model

[0146] The server sends the generated prompt to a generative model (e.g., a Generative Pre-trained Transformer), which generates an output based on the prompt. The generative model receives the prompt "Product review summary, including key points" as input and generates the appropriate text output. The generated output (e.g., "Product review summary") is sent back to the server.

[0147] Step 6: Receiving a response from the generative model

[0148] The server receives the output results returned from the generative model. Text data containing the "product review summary" generated by the generative model is returned to the server. This received data becomes the input data for proceeding to the next display step.

[0149] Step 7: View the results

[0150] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, specific text data for displaying a "product review summary" is sent to the device and displayed to the user on the device screen in the form of "I thought this product was easy to use and very durable. I was particularly impressed by its light weight." This displayed text is the final output, and the user can confirm the generated information.

[0151] The above are the specific processing steps of this system. Each step starts with user input, then includes analysis, prompt generation, output of the generative model, and finally display to the user.

[0152] (Application example 1)

[0153] 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."

[0154] In autonomous vehicles, there is a need for a means to efficiently provide passengers with the information they request. In particular, there is a need for a system that can instantly obtain the information passengers need using voice input or screen operation, and provide it on the spot by display or voice. It is also important that this will improve passenger satisfaction and enhance the convenience of autonomous vehicles.

[0155] 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.

[0156] In this invention, the server includes means for receiving user input information, means for analyzing the user's intention from the received input information, means for automatically generating appropriate prompts based on the analyzed intention, means for transmitting the generated prompts to a generative model, means for receiving a response from the generative model, means for displaying the received response to the user, and means for providing information based on passenger questions using a voice recognition and display device installed in the autonomous vehicle. This enables passengers to instantly obtain necessary information through voice input or screen operation, improving the user experience in the autonomous vehicle.

[0157] "User input information" refers to information such as voice, text, and icon operations that the user provides to the system.

[0158] The "receiving means" is a mechanism for receiving user input information at a terminal or server.

[0159] The "means for analyzing intent" is a method for understanding and analyzing the user's intent based on the received input information.

[0160] The "means for automatically generating a prompt" is a function that automatically generates an appropriate instruction sentence based on the analyzed user's intention.

[0161] A "generative model" is an artificial intelligence technique for generating text, images, videos, etc. based on specific prompts.

[0162] A "means for sending a prompt" is a mechanism for passing a generated prompt to a Generative Model.

[0163] The "means for receiving a response" is a mechanism for receiving a response from the Generative Model.

[0164] A "display means" is a device for visually or audibly presenting the received response to the user.

[0165] An "autonomous vehicle" is a vehicle that can perform driving operations automatically.

[0166] A "voice recognition device" is a device for converting passenger voice input into text data.

[0167] A "display device" is a display device for displaying data such as text and images.

[0168] A "passenger" is a person using an autonomous vehicle.

[0169] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described in detail with reference to the accompanying drawings.

[0170] System configuration

[0171] This system consists of three elements: a server, a terminal, and a user, in order to realize information provision services in autonomous vehicles.

[0172] 1. Receiving user input information

[0173] Users input information using terminals (for example, a voice recognition device or a display device) installed in the autonomous vehicle. The voice recognition device is a device that converts passengers' voice input into text data. The display device is a device that displays data such as text and images. Users make requests by voice, such as "Please tell me about nearby tourist spots."

[0174] 2. Providing an interface

[0175] The terminal provides an interface for the user to provide input information, which may include a microphone for voice input, a display screen, etc. The information input by the user is transmitted to the server in real time.

[0176] 3. Analyzing User Intent

[0177] The server analyzes the user's input information received from the device. It uses natural language processing (NLP) technology to analyze the user's voice data and understand the user's intent. Through this analysis, the server identifies the information the user is seeking and generates appropriate prompts based on that information.

[0178] 4. Automatically Generate Prompts

[0179] Based on the analyzed user intent, the server automatically generates a prompt containing the necessary information, for example, "A passenger is asking about nearby tourist attractions. Please provide the most suitable tourist attraction information based on the following."

[0180] 5. Sending prompts to the generative model

[0181] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques such as Generative Pre-trained Transformer (GPT) to generate outputs based on the user's prompts.

[0182] 6. Receiving a response from the generative model

[0183] The results that the generative model generates based on the specified prompts are sent back to the server. For example, if the generative model generates a "list of nearby tourist attractions" based on the user's request, that information is sent to the server.

[0184] 7. Displaying the results

[0185] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, the generated tourist information is displayed on a screen or guided by voice.

[0186] Specific examples

[0187] When a passenger says, "Please tell me about nearby tourist spots in this car," the voice input is received by the device and sent to the server. The server converts the voice data into text and analyzes the user's intention through natural language processing. As a result, it automatically generates a "prompt about nearby tourist spots" and sends it to the generative model. The generative model generates tourist spot information based on this prompt and sends it back to the server. Finally, the server sends the received tourist spot information to the user's device and displays it on the screen.

[0188] Prompt Sentence Examples

[0189] "A passenger is asking about nearby tourist attractions. Please provide the best tourist information based on the following:

[0190] Question: Please tell me about nearby tourist spots

[0191] Required information: name of tourist spot, distance, highlights

[0192] As described above, the present invention is a system that allows users to instantly obtain necessary information through voice input or screen operation, thereby improving the user experience in an autonomous vehicle.

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

[0194] Step 1:

[0195] The user enters information by voice.

[0196] Input: User's voice (e.g., "Please tell me about nearby tourist spots")

[0197] Output: Audio data

[0198] Specific operation: A user provides speech input to a speech recognition device installed in an autonomous vehicle, which receives this input as digital voice data.

[0199] Step 2:

[0200] The terminal converts the voice data into text data.

[0201] Input: Digital audio data

[0202] Output: Text data

[0203] What it does: The device's voice recognition software (e.g., Google Speech-to-Text API) analyzes the voice data and converts it into corresponding text data, such as "Please tell me about nearby tourist spots."

[0204] Step 3:

[0205] The server analyzes the text data and understands the user's intent.

[0206] Input: Text data

[0207] Output: Intent information (e.g., "I'm looking for information about tourist spots")

[0208] Specific operation: The server analyzes text data using a natural language processing (NLP) engine (e.g., SpaCy or GPT-3). From the analyzed data, it understands the user's intent and generates intent information according to the purpose.

[0209] Step 4:

[0210] The server automatically generates prompts based on the user's intent.

[0211] Input: Intention information

[0212] Output: The generated prompt (e.g., "Please tell me about nearby tourist attractions")

[0213] What it does: The server uses the parsed intent information to construct an appropriate prompt, which is automatically generated using a preset template.

[0214] Step 5:

[0215] The server sends the generated prompt to the Generative Model.

[0216] Input: Generated prompt

[0217] Output: A request to the generative model with the prompt sent

[0218] Specific operation: The server sends the generated prompt to the generative model (e.g., OpenAI (registered trademark) API) via an HTTP request.

[0219] Step 6:

[0220] The generative model generates a response based on the prompt and sends it back to the server.

[0221] Input: prompt

[0222] Output: Generated response (e.g., list of tourist attractions)

[0223] Specific operation: The generative model analyzes the prompt and generates an appropriate response, which is sent back to the server in the form of text data, image data, or other data.

[0224] Step 7:

[0225] The server receives the response from the generative model and sends it to the terminal.

[0226] Input: Generated response

[0227] Output: Response data sent to the terminal

[0228] Specific operation: The server transfers the response data received from the generative model via the network to the terminal.

[0229] Step 8:

[0230] The terminal displays or audibly presents the received response to the user.

[0231] Input: Response data (e.g., a list of tourist attractions)

[0232] Output: Display or audio output to the user

[0233] Specific actions: Display a list of tourist attractions on the device's display device, or use a speech synthesizer to provide the information to the user aloud.

[0234] 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.

[0235] This invention relates to a system that receives user input information, analyzes the user's intentions and emotions, automatically generates appropriate prompts, and sends the prompts to a generative model. This system consists of three elements: a server, a terminal, and a user, and is also combined with an emotion engine.

[0236] System configuration and operation

[0237] User Input

[0238] Users access the system using a device (e.g., smartphone, tablet, PC, etc.). Users can input requests using a variety of methods, including voice recognition, text input, and icon manipulation. For example, a user might input by voice, "I would like to create a summary of the planning meeting."

[0239] Providing an interface

[0240] The device provides an interface for the user to input information, which may include icons, a microphone for voice input, text input fields, etc. As the user provides the input, it is transmitted to the server in real time.

[0241] Analyzing User Intent

[0242] The server analyzes the user's input information received from the device. It uses natural language processing (NLP) techniques to understand the user's request and identify the user's intent. This analysis generates a prompt, such as "Summary of the planning meeting, including key points."

[0243] User Emotion Recognition

[0244] In addition, the server uses an emotion engine to analyze the user's emotions, recognizing the user's emotions from voice and text data and identifying emotional states such as "happiness," "sadness," and "surprise."

[0245] Automatically Generated Prompts

[0246] Based on the analyzed user intent and emotional information, the server automatically generates an appropriate prompt. For example, if a user inputs "Please create a summary of a product review," the server will generate a prompt such as "A summary of a product review, including key points," complementing the user's emotional information.

[0247] Prompting the Generative Model

[0248] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques to generate output based on the user's prompts, which can include text, images, and video data.

[0249] Receiving a response from the generative model

[0250] The results generated by the generative model based on the specified prompts are sent back to the server. For example, if the generative model generates a "summary of the planning meeting" based on the user's request, the summary is sent to the server.

[0251] Displaying the results

[0252] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, the generated summary of a planning meeting can be displayed on the screen, allowing the user to confirm the response to their request.

[0253] Example: A text generation workflow including emotion recognition

[0254] Consider the case where a user voice-inputs "Please create a summary of a product review" into a device. This voice input is received by the device and sent to the server. The server converts the voice data into text and analyzes the user's intent using natural language processing technology. At the same time, it uses an emotion engine to recognize the user's emotions and includes that information in the prompt. For example, it generates a prompt such as "Summary of a product review, including key points, user is satisfied" and sends it to the generative model. The generative model generates a product review summary based on this prompt and sends it back to the server. Finally, the server sends the generated result to the user's device, which displays the summary to the user.

[0255] As described above, the present invention is a system that provides more appropriate responses by eliminating the need for users to manually input detailed prompts, by utilizing generative models simply and efficiently, and by generating prompts that take the user's emotions into consideration.

[0256] The processing flow will be explained below.

[0257] Step 1:

[0258] A user accesses a terminal.

[0259] Action: A user launches an application on the system using a device such as a smartphone or PC.

[0260] Step 2:

[0261] The user provides the input information.

[0262] Action: User clicks icon or speaks or writes "I want to create a summary of the planning meeting."

[0263] Step 3:

[0264] The terminal receives the input information.

[0265] How it works: The device receives the user's clicks, voice, and text input in real time and transmits that information to the server.

[0266] Step 4:

[0267] The server analyzes the input information.

[0268] How it works: The server analyzes the data received from the device, converts voice data into text, and uses natural language processing (NLP) techniques to understand the user's intent.

[0269] Step 5:

[0270] The server recognizes the user's emotions.

[0271] How it works: The server uses an emotion engine to analyze the user's emotional state from the received voice and text data, for example, identifying emotions such as "happiness," "sadness," or "surprise" from the user's tone of voice and the context of the text.

[0272] Step 6:

[0273] The server generates the appropriate prompt.

[0274] How it works: Based on the analyzed user intent and emotion information, the server automatically generates a prompt such as "Summary of the planning meeting, including key points." It may also fine-tune the prompt based on the recognized emotion.

[0275] Step 7:

[0276] The server sends a prompt to the generative model.

[0277] How it works: The server sends the generated prompt to the API endpoint of the AI ​​generator as an HTTP request. The prompt also contains the user's emotional information.

[0278] Step 8:

[0279] The server receives the response from the generative model.

[0280] How it works: The generative model generates output based on the prompts, and the results are sent back to the server, for example, a summary of the analyzed planning meeting.

[0281] Step 9:

[0282] The server sends the results to the terminal.

[0283] How it works: The server formats the generated output and creates data packets to send to the user's terminal.

[0284] Step 10:

[0285] The terminal displays the results to the user.

[0286] Operation: The device visualizes the received data and presents it to the user. For example, it displays a summary of the planning meeting on the screen.

[0287] Examples:

[0288] 1. The user speaks, "I want a summary of a product review."

[0289] 2. The device sends the audio data to the server.

[0290] 3. The server transcribes the audio data, performs natural language processing, and analyzes the user's intent.

[0291] 4. The server uses the emotion engine to recognize the user's emotion as "satisfied."

[0292] 5. The server generates a prompt: "Summary of product review, including key points, user satisfied."

[0293] 6. The server sends a prompt to the generative model.

[0294] 7. The generative model generates a summary of the review and sends it back to the server.

[0295] 8. The server sends the results to the user's device.

[0296] 9. The terminal displays the summary to the user.

[0297] In this way, the burden on the user to input detailed prompts is reduced, and the system automatically analyzes and generates, making it possible to effectively utilize generative models and provide more appropriate responses that take the user's emotions into account.

[0298] Example 2

[0299] 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."

[0300] In conventional systems, users have to manually input detailed prompts, which is time-consuming, and it is difficult to utilize generative models that take into account the user's intentions and emotions. As a result, the responses users receive are often inappropriate.

[0301] 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.

[0302] In this invention, the server includes means for receiving user input information, means for analyzing the user's intention from the received input information, and means for automatically generating appropriate prompts based on the analyzed intention and the user's emotional information. This allows for the automatic generation of prompts that take the user's intention and emotion into consideration, making it possible to produce more appropriate and accurate output from the generative model.

[0303] "User input information" refers to any information provided by the user to the system, including voice, text, icon operations, and the like.

[0304] The "receiving means" refers to a function or device that allows the system to receive input information provided by the user.

[0305] "User intent" refers to the user's goals and desires as interpreted by the system based on input information.

[0306] "Means for analyzing" refers to technology or devices for analyzing received input information and understanding the user's intentions.

[0307] "Emotion information" refers to the user's emotional state analyzed from voice and text data, and includes "joy," "sadness," "surprise," and the like.

[0308] A "prompt" is text information that provides instructions or guidelines for a generative model when generating output.

[0309] "Automatic generation means" refers to technology or devices for automatically creating appropriate prompts based on the user's intentions and emotional information.

[0310] A "generative model" is a model that includes artificial intelligence techniques that generate output, such as text, images, or video, based on input prompts.

[0311] The "sending means" is a function or device for sending the generated prompt to the Generative Model.

[0312] A "response" is the output result generated by a generative model based on a prompt.

[0313] The "means for displaying" is a function or device for visually or audibly presenting the response received from the generative model to the user.

[0314] This invention relates to a system that receives user input information, analyzes the user's intentions and emotions, automatically generates appropriate prompts, and sends the prompts to a generative model. This system consists of three elements: a server, a terminal, and a user, and is also combined with an emotion engine.

[0315] System configuration and operation

[0316] User Input

[0317] Users access the system using a device (e.g., smartphone, tablet, PC, etc.). Users can input requests using a variety of methods, including voice recognition, text input, and icon manipulation. For example, a user might input by voice, "I would like to create a summary of the planning meeting."

[0318] Providing an interface

[0319] The device provides an interface for the user to input information, which may include icons, a microphone for voice input, text input fields, etc. As the user provides the input, it is transmitted to the server in real time.

[0320] Analyzing User Intent

[0321] The server analyzes the user's input information received from the device. It uses natural language processing (NLP) techniques to understand the user's request and identify the user's intent. This analysis generates a prompt, such as "Summary of the planning meeting, including key points."

[0322] User Emotion Recognition

[0323] In addition, the server uses an emotion engine to analyze the user's emotions, recognizing the user's emotions from voice and text data and identifying emotional states such as "happiness," "sadness," and "surprise."

[0324] Automatically Generated Prompts

[0325] Based on the analyzed user intent and emotional information, the server automatically generates an appropriate prompt. For example, if a user inputs "Please create a summary of a product review," the server will generate a prompt such as "A summary of a product review, including key points," complementing the user's emotional information.

[0326] Prompting the Generative Model

[0327] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques to generate output based on the user's prompts, which can include text, images, and video data.

[0328] Receiving a response from the generative model

[0329] The results generated by the generative model based on the specified prompts are sent back to the server. For example, if the generative model generates a "summary of the planning meeting" based on the user's request, the summary is sent to the server.

[0330] Displaying the results

[0331] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, the generated summary of a planning meeting can be displayed on the screen, allowing the user to confirm the response to their request.

[0332] Example: A text generation workflow including emotion recognition

[0333] Consider the case where a user voice-inputs "Please create a summary of a product review" into a device. This voice input is received by the device and sent to the server. The server converts the voice data into text and analyzes the user's intent using natural language processing technology. At the same time, it uses an emotion engine to recognize the user's emotions and includes that information in the prompt. For example, it generates a prompt such as "Summary of a product review, including key points, user is satisfied" and sends it to the generative model. The generative model generates a product review summary based on this prompt and sends it back to the server. Finally, the server sends the generated result to the user's device, which displays the summary to the user.

[0334] As described above, the present invention is a system that provides more appropriate responses by eliminating the need for users to manually input detailed prompts, by utilizing generative models simply and efficiently, and by generating prompts that take the user's emotions into consideration.

[0335] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0336] Step 1:

[0337] Receiving user input information

[0338] Users use devices (smartphones, tablets, PCs, etc.) to input information by voice, text, or icon operation. The input information is transferred from the device to the server in real time.

[0339] Input: User's voice command "Please create a summary of the product review."

[0340] Output: User request in text format: "Please create a summary of product reviews."

[0341] Step 2:

[0342] Providing an interface

[0343] The device provides an interface for the user to input information, which may include a microphone for voice input, text entry fields, icons, etc.

[0344] Input: User operations (voice input, text input, icon operation)

[0345] Output: Input data (voice, text data) from the device

[0346] Step 3:

[0347] Transfer of input information to the server

[0348] The terminal transfers the information entered by the user to the server in real time.

[0349] Input: Voice or text data received by the device

[0350] Output: Audio or text data sent to the server

[0351] Step 4:

[0352] Analyzing User Intent

[0353] The server analyzes the received user input (voice or text) and uses natural language processing (NLP) techniques to understand the user's intent. This analysis identifies the intent and generates a specific prompt.

[0354] Input: Voice or text data sent to the server

[0355] Output: Parsed user intent: "Summary of product review, including key points"

[0356] Step 5:

[0357] User sentiment analysis

[0358] The server uses an emotion engine to analyze the user's emotions from the received voice and text data, identifying emotions such as "happiness," "sadness," and "surprise."

[0359] Input: Audio or text data

[0360] Output: Identified user emotional state "Satisfied"

[0361] Step 6:

[0362] Automatically Generated Prompts

[0363] The server automatically generates an appropriate prompt based on the analyzed user intent and emotion information. For example, a prompt such as "Summary of product review, including key points, user satisfied" may be generated.

[0364] Input: Parsed user intent and emotion information

[0365] Output: Auto-generated prompt "Summary of product reviews, including key points, satisfied users"

[0366] Step 7:

[0367] Prompting the Generative Model

[0368] The server sends the generated prompts to a generative AI model, which generates an output based on the prompts.

[0369] Input: Auto-generated prompt

[0370] Output: The prompt sent to the generative AI model

[0371] Step 8:

[0372] Receiving a response from the generative model

[0373] The generative AI model generates output based on the prompts sent and sends the results back to the server.

[0374] Input: Prompt sent

[0375] Output: Generated text "Product Review Summary"

[0376] Step 9:

[0377] Transfer of generated results from the server to the device

[0378] The server transmits the results received from the generative model to the user's terminal.

[0379] Input: The generated results sent from the generative model

[0380] Output: Generated results sent to the terminal

[0381] Step 10:

[0382] Viewing the generated results

[0383] The terminal presents the generated summary of the product review to the user visually or audibly, for example, by displaying it on a screen so that the user can confirm the response to the request.

[0384] Input: Generated result

[0385] Output: The generated result displayed on the user's terminal.

[0386] (Application example 2)

[0387] 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."

[0388] Conventional systems that accept user input can analyze the user's intentions and provide corresponding generated prompts, but they do not take the user's emotions into account when generating prompts, making it impossible to provide responses appropriate to the user's psychological state.In addition, there was a lack of concrete application examples for improving the user's shopping experience, making it difficult to improve customer satisfaction in physical stores.

[0389] The identification process 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 receiving user input information, means for analyzing the user's intention from the received input information, means for analyzing the user's emotion from the received input information, means for automatically generating an appropriate prompt based on the analyzed intention and emotion, means for sending the generated prompt to the generative model, means for receiving a response from the generative model, and means for displaying the received response to the user. This makes it possible to provide a more accurate and satisfying response based on the user's intention and emotion.

[0390] "User input information" refers to information input by the user through various means such as voice, text, or icon operation.

[0391] "Intent analysis" is the process of determining the content and purpose of received user input.

[0392] "Emotion analysis" is the process of identifying a user's emotional state from speech or text data, for example, detecting emotions such as joy, sadness, or surprise.

[0393] "Automatic prompt generation" refers to the process by which the system automatically generates appropriate instructions or commands based on the analyzed intent and sentiment.

[0394] A "generative model" is a system that uses artificial intelligence technology to generate text, images, and videos based on user requests.

[0395] "Receiving a response" refers to the process by which the server receives the results generated from the Generative Model.

[0396] "Displaying to the user" refers to the process of presenting the generated results to the user through the screen or audio of the device they use.

[0397] This invention relates to a system that receives user input, analyzes the user's intentions and emotions, automatically generates appropriate prompts, and sends the prompts to a generative model. This system mainly consists of three elements: a server, a terminal, and a user, and is also combined with an emotion engine.

[0398] System configuration and operation

[0399] User Input

[0400] Users access the system using a device (e.g., a smartphone or smart glasses). Users can input requests using a variety of methods, including voice recognition, text input, and icon manipulation. For example, a user might say, "Please give me a review of this bag."

[0401] Providing an interface

[0402] The device provides an interface for the user to input information, including a microphone for voice input, text input fields, and a UI for icon manipulation. As the user provides input, it is transmitted to the server in real time.

[0403] Analyzing user intent and emotions

[0404] The server analyzes the user input information received from the device. It uses natural language processing (NLP) technology to understand the user's request and identify the user's intent. This analysis is performed using, for example, the TextBlob library. It also analyzes the user's emotions using an emotion engine. It recognizes the user's emotions from voice and text data and identifies emotional states such as "happy," "unhappy," or "neutral." This process uses Google's speech recognition API and TextBlob library.

[0405] Automatically Generated Prompts

[0406] Based on the analyzed user intent and emotion information, the server automatically generates an appropriate prompt. For example, if a user inputs "Please give me a review about this bag," the server generates a prompt such as "The user said the following in a happy state: Please give me a review about this bag. Please generate the optimal product review based on this," and sends it to the generative model.

[0407] Prompting the Generative Model

[0408] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques to generate output based on the user's prompts. This output can include text, images, and video data. Specifically, OpenAI's API is used.

[0409] Receiving a response from the generative model

[0410] The results generated by the generative model based on the specified prompts are sent back to the server. For example, if the generative model generates a "review of this bag" based on the user's request, the review is sent to the server.

[0411] Displaying the results

[0412] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, the generated bag review can be displayed on the screen, allowing the user to confirm the response to their request.

[0413] Specific examples

[0414] Consider a scenario where a user uses smart glasses to voice-input "Please give me a review about this bag." This voice input is received by the device and sent to a server, which converts the voice data into text and analyzes the user's intent and emotions using natural language processing technology and an emotion engine. Based on the analysis, the server generates the following prompt:

[0415] Example prompt sentence:

[0416] In the delighted state, the user said: Give me a review about this bag

[0417] Generate optimal product reviews based on this.

[0418] The generative model generates product reviews based on the prompts and sends them back to the server, which then sends the generated results to the user's device, which then displays the reviews to the user.

[0419] As described above, the present invention is a system that generates prompts based on the user's intentions and emotions, thereby providing more accurate responses.

[0420] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0421] Step 1:

[0422] The user inputs a request to the terminal by voice or text. For example, the user inputs "Please give me a review about this bag." This voice data is received by the terminal.

[0423] Step 2:

[0424] The device converts the voice data into text data using Google's speech recognition API, resulting in the text data "Write a review about this bag."

[0425] Step 3:

[0426] The device sends text data to the server, which then receives the text and analyzes the user's intent. This is done using natural language processing (NLP) techniques, such as the TextBlob library. The analysis identifies the user's intent as a request for a bag review.

[0427] Step 4:

[0428] At the same time, the server analyzes the user's emotions from the text data using the TextBlob library, for example, to identify the emotion "joy."

[0429] Step 5:

[0430] The server automatically generates prompts based on the analyzed user intent and emotion, such as "The user said the following in the joy state: Give us your review of this bag. Based on this, generate the best product review."

[0431] Step 6:

[0432] The server sends the automatically generated prompts to a generative model, which uses OpenAI's API, and the generative model receives the prompts and generates a product review based on them.

[0433] Step 7:

[0434] The product reviews generated by the generative model are sent back to the server. For example, a review might be generated that reads, "This bag is very user-friendly and has an excellent design. It is particularly notable for the high-quality materials used."

[0435] Step 8:

[0436] The server sends the reviews received from the generative model to the device, which receives the reviews and displays them visually or audibly to the user, allowing the user to confirm the response to their request.

[0437] Through the above processing steps, the user can obtain accurate product reviews based on his or her own intentions and feelings.

[0438] 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.

[0439] 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.

[0440] 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.

[0441] [Second embodiment]

[0442] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0443] 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.

[0444] 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).

[0445] 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.

[0446] 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.

[0447] 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).

[0448] 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.

[0449] 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.

[0450] 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.

[0451] 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.

[0452] 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.

[0453] 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."

[0454] The present invention relates to a system that utilizes a generative model using user input information, and a specific embodiment thereof will be described below. The system consists of three elements: a server, a terminal, and a user.

[0455] System configuration and operation

[0456] User Input

[0457] A user accesses the system using a device (e.g., a smartphone, tablet, or PC). At this stage, the user can select from a variety of input methods, such as voice recognition, text input, and icon manipulation, to input requests to the generative model. For example, a user may input in text format, "I would like to create a summary of the planning meeting," or make a similar request by voice.

[0458] Providing an interface

[0459] The terminal provides an interface for the user to provide input information, which may include a user interface including icons, microphone access permission for voice input, text input fields, etc. When the user inputs information through these interfaces, the information is transmitted to the server in real time.

[0460] Analyzing user intent

[0461] The server analyzes the user's input information received from the device. Using natural language processing (NLP) technology, it analyzes the user's text and voice data to understand the user's intent. This analysis allows the server to determine what the user is looking for and generate appropriate prompts based on that. As a result of this analysis, the server may generate a prompt such as "Summary of the planning meeting, including key points."

[0462] Automatically Generated Prompts

[0463] Based on the analyzed user intent, the server automatically generates a prompt containing the necessary information. For example, if a user inputs "Please create a summary of a product review," the server generates a prompt "Summary of a product review, including key points." This prompt is sent to the generative model.

[0464] Prompting the Generative Model

[0465] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques (e.g., a Generative Pre-trained Transformer) to generate output based on the user's prompts. This output can include text, images, video data, etc.

[0466] Receiving a response from the generative model

[0467] The results generated by the generative model based on the specified prompts are sent back to the server. For example, if the generative model generates a "summary of the planning meeting" based on the user's request, the summary is sent to the server.

[0468] Displaying the results

[0469] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. This step allows the user to easily check the generated output, such as text, images, and videos. For example, a summary of a planning meeting generated can be displayed on the screen, allowing the user to confirm the response to their request.

[0470] Example: A series of steps for generating sentences

[0471] Consider the case where a user voice-inputs "Please create a summary of a product review" into a device. This voice input is received by the device and sent to the server. The server converts the voice data into text and analyzes the user's intent through natural language processing. Based on the results, it automatically generates a prompt "Summary of a product review, including key points" and sends it to the generative model. The generative model generates a product review summary based on this prompt and sends it back to the server. Finally, the server sends the received summary to the user's device, which displays it to the user.

[0472] As described above, the present invention provides a system for easily and efficiently using generative models, eliminating the need for users to manually input detailed prompts.

[0473] The processing flow will be explained below.

[0474] Step 1:

[0475] A user accesses a terminal.

[0476] Action: A user launches an application on the system using a device such as a smartphone or PC.

[0477] Step 2:

[0478] The user provides the input information.

[0479] Action: User clicks icon or speaks or writes "I want to create a summary of the planning meeting."

[0480] Step 3:

[0481] The terminal receives the input information.

[0482] How it works: The device receives the user's clicks, voice, and text input in real time and transmits that information to the server.

[0483] Step 4:

[0484] The server analyzes the input information.

[0485] How it works: The server analyzes the data received from the device, converts voice data into text, and uses natural language processing (NLP) techniques to understand the user's intent.

[0486] Step 5:

[0487] The server generates the appropriate prompt.

[0488] How it works: Based on the analysis results, the server automatically generates a prompt such as "Summary of the planning meeting, including key points."

[0489] Step 6:

[0490] The server sends a prompt to the generative model.

[0491] How it works: The server sends the generated prompt as an HTTP request to the API endpoint of the generating AI.

[0492] Step 7:

[0493] The server receives the response from the generative model.

[0494] How it works: The generative AI model generates output based on the prompts, and the results are sent back to the server.

[0495] Step 8:

[0496] The server sends the results to the terminal.

[0497] How it works: The server formats the generated output and creates data packets to send to the user's terminal.

[0498] Step 9:

[0499] The terminal displays the results to the user.

[0500] Operation: The device visualizes the received data and presents it to the user. For example, it displays a summary of the planning meeting on the screen.

[0501] Examples:

[0502] 1. The user speaks, "I want a summary of a product review."

[0503] 2. The device sends the audio data to the server.

[0504] 3. The server transcribes the audio data, performs natural language processing, and generates prompts.

[0505] 4. The server sends a prompt to the generative model.

[0506] 5. The generative model generates a summary of the review and sends it back to the server.

[0507] 6. The server sends the results to the user's terminal, which displays a summary to the user.

[0508] In this way, the system automatically performs analysis and generation without the user having to enter complex prompts, making generative models easier to use.

[0509] Example 1

[0510] 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."

[0511] When users use complex generative models, many detailed operations are required, which places a heavy burden on them. In particular, when using voice or text input, users must manually input appropriate prompts into the generative model, which is a time-consuming and labor-intensive task. There is a need for a method that solves this problem and allows users to easily use generative models.

[0512] 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.

[0513] In this invention, the server includes a means for converting voice data into text data, a means for analyzing the converted text data using natural language processing technology, and a means for formatting the generated prompt into a format that is easy for the generative model to understand, thereby enabling the user to use the generative model simply and quickly without performing complex operations.

[0514] "User input information" refers to data that a user provides to the system in the form of voice, text, icon manipulation, or the like.

[0515] "Means for receiving" refers to a function for receiving information input by a user and incorporating it into the system.

[0516] "Means for analyzing" refers to methods and techniques for processing received user input information and understanding its content.

[0517] A "prompt" is a short instruction that specifically specifies the desired output from the generative model.

[0518] A "generative model" refers to an artificial intelligence system that generates text, images, video, etc. based on a given prompt.

[0519] "Means of conversion" refers to the methods and techniques for converting audio data into text data.

[0520] "Natural language processing technology" refers to a technological system that enables computers to understand and analyze human language and perform appropriate information processing.

[0521] "Formatting means" refers to a method for adjusting the format of the generated prompt so that the generative model can efficiently understand it and generate a response.

[0522] A "response" refers to the output that a generative model generates based on a prompt.

[0523] "Means for displaying" refers to a function for visually or audibly presenting the response received from the generative model to the user.

[0524] The present invention relates to a system that utilizes a generative model based on user input information. A specific embodiment of this system is described below. The system consists of three elements: a server, a terminal, and a user.

[0525] System configuration and operation

[0526] User Input

[0527] Users access the system using devices such as smartphones, tablets, and PCs. When sending requests to the system, users can choose from a variety of input methods, such as voice recognition, text input, and icon manipulation. For example, a user may input a request in text format, such as "Please create a product review summary." Alternatively, a user may make a similar request by voice.

[0528] Providing an interface

[0529] The terminal provides an interface for the user to provide input information, including a text input field, a user interface including icons, and microphone access permission for voice input. When the user inputs information through these interfaces, the information is transmitted to the server in real time.

[0530] Analyzing user intent

[0531] The server analyzes the user's input information sent from the device. It uses natural language processing (NLP) techniques to analyze text and voice data and understand the user's intent. For example, it uses an NLP engine (e.g., a common NLP library or API) to identify that the user's request is to "create a summary of a product review."

[0532] Automatically Generated Prompts

[0533] Based on the analysis results, the server automatically generates a prompt containing the necessary information. For example, it generates a prompt such as "A summary of product reviews, including key points." This prompt is formatted in a way that is easy for the generative model to understand.

[0534] Prompting the Generative Model

[0535] The server sends the generated prompts to a generative model (e.g., an artificial intelligence system such as a Generative Pre-trained Transformer) that uses the prompts to generate output for the user's request, such as text, images, or video. The generative model uses advanced AI algorithms to generate the requested data in real time.

[0536] Receiving a response from the generative model

[0537] The output generated by the generative model is sent back to the server. For example, a "product review summary" is generated and sent to the server. This generated result is sent as structured data, such as text.

[0538] Displaying the results

[0539] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. The device displays the generated text, image, or video output to the user on a screen or using a voice assistant. For example, a generated "product review summary" is displayed on the device screen.

[0540] Specific examples

[0541] A specific example is a case where the user vocally inputs into the terminal, "I want you to create a review summary of the product."

[0542] 1. User input:

[0543] The user speaks to the terminal, "I want a review summary of the product."

[0544] 2. Receiving audio data:

[0545] The device provides microphone access permission for voice input and transmits the received voice data to the server, for example, when the user makes a request to "activate the voice assistant."

[0546] 3. Text Conversion and Intent Analysis:

[0547] The server converts the voice data into text and analyzes the user's intent using natural language processing technology. The server converts the voice data into text using the Google Speech-to-Text API, and then analyzes the intent using a common NLP engine.

[0548] 4. Generate prompt:

[0549] The server generates a prompt such as "Product review summary, including key points" and sends this prompt to the generative model.

[0550] 5. Generative model response:

[0551] A generative model (e.g., GPT-3) generates a review summary based on the prompts and sends the summary back to the server.

[0552] 6. View the response:

[0553] The server sends the summary to the device, which then displays it to the user. The device screen displays the review summary: "I found this product easy to use and very durable. I was particularly impressed by its light weight."

[0554] By following the above steps, users can use generative AI models easily and efficiently.

[0555] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0556] Step 1: User Input

[0557] Users access the system using devices such as smartphones, tablets, and PCs. For example, a user inputs a request in text format, such as "Please create a product review summary." The user's input is entered into the device as text data, such as "Please create a product review summary." This input data is sent to the server through the device's interface.

[0558] Step 2: Provide an interface

[0559] The device provides a user interface that includes text input fields, icons, and microphone access permission for voice input. When a user inputs information through these interfaces, the information is transmitted to the server in real time. This transmitted data becomes the input information required for the next analysis step.

[0560] Step 3: Analyze user intent

[0561] The server analyzes the user's input information received from the terminal. Specifically, it analyzes the text data using natural language processing technology (e.g., a standard NLP engine). The server receives the text data "Please create a product review summary" as input and analyzes this data to identify the user's intent. The output of the analysis is the intent to "generate a product review summary."

[0562] Step 4: Automatically generate prompts

[0563] The server automatically generates a prompt containing the necessary information based on the analyzed user intent. For example, it generates a prompt such as "Product review summary, including important points." The prompt generated by the server is text data such as "Product review summary, including important points." This prompt is sent as input to the generative model.

[0564] Step 5: Sending prompts to the generative model

[0565] The server sends the generated prompt to a generative model (e.g., a Generative Pre-trained Transformer), which generates an output based on the prompt. The generative model receives the prompt "Product review summary, including key points" as input and generates the appropriate text output. The generated output (e.g., "Product review summary") is sent back to the server.

[0566] Step 6: Receiving a response from the generative model

[0567] The server receives the output results returned from the generative model. Text data containing the "product review summary" generated by the generative model is returned to the server. This received data becomes the input data for proceeding to the next display step.

[0568] Step 7: View the results

[0569] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, specific text data for displaying a "product review summary" is sent to the device and displayed to the user on the device screen in the form of "I thought this product was easy to use and very durable. I was particularly impressed by its light weight." This displayed text is the final output, and the user can confirm the generated information.

[0570] The above are the specific processing steps of this system. Each step starts with user input, then includes analysis, prompt generation, output of the generative model, and finally display to the user.

[0571] (Application example 1)

[0572] 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."

[0573] In autonomous vehicles, there is a need for a means to efficiently provide passengers with the information they request. In particular, there is a need for a system that can instantly obtain the information passengers need using voice input or screen operation, and provide it on the spot by display or voice. It is also important that this will improve passenger satisfaction and enhance the convenience of autonomous vehicles.

[0574] 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.

[0575] In this invention, the server includes means for receiving user input information, means for analyzing the user's intention from the received input information, means for automatically generating appropriate prompts based on the analyzed intention, means for transmitting the generated prompts to a generative model, means for receiving a response from the generative model, means for displaying the received response to the user, and means for providing information based on passenger questions using a voice recognition and display device installed in the autonomous vehicle. This enables passengers to instantly obtain necessary information through voice input or screen operation, improving the user experience in the autonomous vehicle.

[0576] "User input information" refers to information such as voice, text, and icon operations that the user provides to the system.

[0577] The "receiving means" is a mechanism for receiving user input information at a terminal or server.

[0578] The "means for analyzing intent" is a method for understanding and analyzing the user's intent based on the received input information.

[0579] The "means for automatically generating a prompt" is a function that automatically generates an appropriate instruction sentence based on the analyzed user's intention.

[0580] A "generative model" is an artificial intelligence technique for generating text, images, videos, etc. based on specific prompts.

[0581] A "means for sending a prompt" is a mechanism for passing a generated prompt to a Generative Model.

[0582] The "means for receiving a response" is a mechanism for receiving a response from the Generative Model.

[0583] A "display means" is a device for visually or audibly presenting the received response to the user.

[0584] An "autonomous vehicle" is a vehicle that can perform driving operations automatically.

[0585] A "voice recognition device" is a device for converting passenger voice input into text data.

[0586] A "display device" is a display device for displaying data such as text and images.

[0587] A "passenger" is a person using an autonomous vehicle.

[0588] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described in detail with reference to the accompanying drawings.

[0589] System configuration

[0590] This system consists of three elements: a server, a terminal, and a user, in order to realize information provision services in autonomous vehicles.

[0591] 1. Receiving user input information

[0592] Users input information using terminals (for example, a voice recognition device or a display device) installed in the autonomous vehicle. The voice recognition device is a device that converts passengers' voice input into text data. The display device is a device that displays data such as text and images. Users make requests by voice, such as "Please tell me about nearby tourist spots."

[0593] 2. Providing an interface

[0594] The terminal provides an interface for the user to provide input information, which may include a microphone for voice input, a display screen, etc. The information input by the user is transmitted to the server in real time.

[0595] 3. Analyzing User Intent

[0596] The server analyzes the user's input information received from the device. It uses natural language processing (NLP) technology to analyze the user's voice data and understand the user's intent. Through this analysis, the server identifies the information the user is seeking and generates appropriate prompts based on that information.

[0597] 4. Automatically Generate Prompts

[0598] Based on the analyzed user intent, the server automatically generates a prompt containing the necessary information, for example, "A passenger is asking about nearby tourist attractions. Please provide the most suitable tourist attraction information based on the following."

[0599] 5. Sending prompts to the generative model

[0600] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques such as Generative Pre-trained Transformer (GPT) to generate outputs based on the user's prompts.

[0601] 6. Receiving a response from the generative model

[0602] The results that the generative model generates based on the specified prompts are sent back to the server. For example, if the generative model generates a "list of nearby tourist attractions" based on the user's request, that information is sent to the server.

[0603] 7. Displaying the results

[0604] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, the generated tourist information is displayed on a screen or guided by voice.

[0605] Specific examples

[0606] When a passenger says, "Please tell me about nearby tourist spots in this car," the voice input is received by the device and sent to the server. The server converts the voice data into text and analyzes the user's intention through natural language processing. As a result, it automatically generates a "prompt about nearby tourist spots" and sends it to the generative model. The generative model generates tourist spot information based on this prompt and sends it back to the server. Finally, the server sends the received tourist spot information to the user's device and displays it on the screen.

[0607] Prompt Sentence Examples

[0608] "A passenger is asking about nearby tourist attractions. Please provide the best tourist information based on the following:

[0609] Question: Please tell me about nearby tourist spots

[0610] Required information: name of tourist spot, distance, highlights

[0611] As described above, the present invention is a system that allows users to instantly obtain necessary information through voice input or screen operation, thereby improving the user experience in an autonomous vehicle.

[0612] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0613] Step 1:

[0614] The user enters information by voice.

[0615] Input: User's voice (e.g., "Please tell me about nearby tourist spots")

[0616] Output: Audio data

[0617] Specific operation: A user provides speech input to a speech recognition device installed in an autonomous vehicle, which receives this input as digital voice data.

[0618] Step 2:

[0619] The terminal converts the voice data into text data.

[0620] Input: Digital audio data

[0621] Output: Text data

[0622] What it does: The device's voice recognition software (e.g., Google Speech-to-Text API) analyzes the voice data and converts it into corresponding text data, such as "Please tell me about nearby tourist spots."

[0623] Step 3:

[0624] The server analyzes the text data and understands the user's intent.

[0625] Input: Text data

[0626] Output: Intent information (e.g., "I'm looking for information about tourist spots")

[0627] Specific operation: The server analyzes text data using a natural language processing (NLP) engine (e.g., SpaCy or GPT-3). From the analyzed data, it understands the user's intent and generates intent information according to the purpose.

[0628] Step 4:

[0629] The server automatically generates prompts based on the user's intent.

[0630] Input: Intention information

[0631] Output: The generated prompt (e.g., "Please tell me about nearby tourist attractions")

[0632] What it does: The server uses the parsed intent information to construct an appropriate prompt, which is automatically generated using a preset template.

[0633] Step 5:

[0634] The server sends the generated prompt to the Generative Model.

[0635] Input: Generated prompt

[0636] Output: A request to the generative model with the prompt sent

[0637] Specific behavior: The server sends the generated prompt via an HTTP request to the generative model (e.g., OpenAI API).

[0638] Step 6:

[0639] The generative model generates a response based on the prompt and sends it back to the server.

[0640] Input: prompt

[0641] Output: Generated response (e.g., list of tourist attractions)

[0642] Specific operation: The generative model analyzes the prompt and generates an appropriate response, which is sent back to the server in the form of text data, image data, or other data.

[0643] Step 7:

[0644] The server receives the response from the generative model and sends it to the terminal.

[0645] Input: Generated response

[0646] Output: Response data sent to the terminal

[0647] Specific operation: The server transfers the response data received from the generative model via the network to the terminal.

[0648] Step 8:

[0649] The terminal displays or audibly presents the received response to the user.

[0650] Input: Response data (e.g., a list of tourist attractions)

[0651] Output: Display or audio output to the user

[0652] Specific actions: Display a list of tourist attractions on the device's display device, or use a speech synthesizer to provide the information to the user aloud.

[0653] 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.

[0654] This invention relates to a system that receives user input information, analyzes the user's intentions and emotions, automatically generates appropriate prompts, and sends the prompts to a generative model. This system consists of three elements: a server, a terminal, and a user, and is also combined with an emotion engine.

[0655] System configuration and operation

[0656] User Input

[0657] Users access the system using a device (e.g., smartphone, tablet, PC, etc.). Users can input requests using a variety of methods, including voice recognition, text input, and icon manipulation. For example, a user might input by voice, "I would like to create a summary of the planning meeting."

[0658] Providing an interface

[0659] The device provides an interface for the user to input information, which may include icons, a microphone for voice input, text input fields, etc. As the user provides the input, it is transmitted to the server in real time.

[0660] Analyzing User Intent

[0661] The server analyzes the user's input information received from the device. It uses natural language processing (NLP) techniques to understand the user's request and identify the user's intent. This analysis generates a prompt, such as "Summary of the planning meeting, including key points."

[0662] User Emotion Recognition

[0663] In addition, the server uses an emotion engine to analyze the user's emotions, recognizing the user's emotions from voice and text data and identifying emotional states such as "happiness," "sadness," and "surprise."

[0664] Automatically Generated Prompts

[0665] Based on the analyzed user intent and emotional information, the server automatically generates an appropriate prompt. For example, if a user inputs "Please create a summary of a product review," the server will generate a prompt such as "A summary of a product review, including key points," complementing the user's emotional information.

[0666] Prompting the Generative Model

[0667] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques to generate output based on the user's prompts, which can include text, images, and video data.

[0668] Receiving a response from the generative model

[0669] The results generated by the generative model based on the specified prompts are sent back to the server. For example, if the generative model generates a "summary of the planning meeting" based on the user's request, the summary is sent to the server.

[0670] Displaying the results

[0671] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, the generated summary of a planning meeting can be displayed on the screen, allowing the user to confirm the response to their request.

[0672] Example: A text generation workflow including emotion recognition

[0673] Consider the case where a user voice-inputs "Please create a summary of a product review" into a device. This voice input is received by the device and sent to the server. The server converts the voice data into text and analyzes the user's intent using natural language processing technology. At the same time, it uses an emotion engine to recognize the user's emotions and includes that information in the prompt. For example, it generates a prompt such as "Summary of a product review, including key points, user is satisfied" and sends it to the generative model. The generative model generates a product review summary based on this prompt and sends it back to the server. Finally, the server sends the generated result to the user's device, which displays the summary to the user.

[0674] As described above, the present invention is a system that provides more appropriate responses by eliminating the need for users to manually input detailed prompts, by utilizing generative models simply and efficiently, and by generating prompts that take the user's emotions into consideration.

[0675] The processing flow will be explained below.

[0676] Step 1:

[0677] A user accesses a terminal.

[0678] Action: A user launches an application on the system using a device such as a smartphone or PC.

[0679] Step 2:

[0680] The user provides the input information.

[0681] Action: User clicks icon or speaks or writes "I want to create a summary of the planning meeting."

[0682] Step 3:

[0683] The terminal receives the input information.

[0684] How it works: The device receives the user's clicks, voice, and text input in real time and transmits that information to the server.

[0685] Step 4:

[0686] The server analyzes the input information.

[0687] How it works: The server analyzes the data received from the device, converts voice data into text, and uses natural language processing (NLP) techniques to understand the user's intent.

[0688] Step 5:

[0689] The server recognizes the user's emotions.

[0690] How it works: The server uses an emotion engine to analyze the user's emotional state from the received voice and text data, for example, identifying emotions such as "happiness," "sadness," or "surprise" from the user's tone of voice and the context of the text.

[0691] Step 6:

[0692] The server generates the appropriate prompt.

[0693] How it works: Based on the analyzed user intent and emotion information, the server automatically generates a prompt such as "Summary of the planning meeting, including key points." It may also fine-tune the prompt based on the recognized emotion.

[0694] Step 7:

[0695] The server sends a prompt to the generative model.

[0696] How it works: The server sends the generated prompt to the API endpoint of the AI ​​generator as an HTTP request. The prompt also contains the user's emotional information.

[0697] Step 8:

[0698] The server receives the response from the generative model.

[0699] How it works: The generative model generates output based on the prompts, and the results are sent back to the server, for example, a summary of the analyzed planning meeting.

[0700] Step 9:

[0701] The server sends the results to the terminal.

[0702] How it works: The server formats the generated output and creates data packets to send to the user's terminal.

[0703] Step 10:

[0704] The terminal displays the results to the user.

[0705] Operation: The device visualizes the received data and presents it to the user. For example, it displays a summary of the planning meeting on the screen.

[0706] Examples:

[0707] 1. The user speaks, "I want a summary of a product review."

[0708] 2. The device sends the audio data to the server.

[0709] 3. The server transcribes the audio data, performs natural language processing, and analyzes the user's intent.

[0710] 4. The server uses the emotion engine to recognize the user's emotion as "satisfied."

[0711] 5. The server generates a prompt: "Summary of product review, including key points, user satisfied."

[0712] 6. The server sends a prompt to the generative model.

[0713] 7. The generative model generates a summary of the review and sends it back to the server.

[0714] 8. The server sends the results to the user's device.

[0715] 9. The terminal displays the summary to the user.

[0716] In this way, the burden on the user to input detailed prompts is reduced, and the system automatically analyzes and generates, making it possible to effectively utilize generative models and provide more appropriate responses that take the user's emotions into account.

[0717] Example 2

[0718] 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."

[0719] In conventional systems, users have to manually input detailed prompts, which is time-consuming, and it is difficult to utilize generative models that take into account the user's intentions and emotions. As a result, the responses users receive are often inappropriate.

[0720] 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.

[0721] In this invention, the server includes means for receiving user input information, means for analyzing the user's intention from the received input information, and means for automatically generating appropriate prompts based on the analyzed intention and the user's emotional information. This allows for the automatic generation of prompts that take the user's intention and emotion into consideration, making it possible to produce more appropriate and accurate output from the generative model.

[0722] "User input information" refers to any information provided by the user to the system, including voice, text, icon operations, and the like.

[0723] The "receiving means" refers to a function or device that allows the system to receive input information provided by the user.

[0724] "User intent" refers to the user's goals and desires as interpreted by the system based on input information.

[0725] "Means for analyzing" refers to technology or devices for analyzing received input information and understanding the user's intentions.

[0726] "Emotion information" refers to the user's emotional state analyzed from voice and text data, and includes "joy," "sadness," "surprise," and the like.

[0727] A "prompt" is text information that provides instructions or guidelines for a generative model when generating output.

[0728] "Automatic generation means" refers to technology or devices for automatically creating appropriate prompts based on the user's intentions and emotional information.

[0729] A "generative model" is a model that includes artificial intelligence techniques that generate output, such as text, images, or video, based on input prompts.

[0730] The "sending means" is a function or device for sending the generated prompt to the Generative Model.

[0731] A "response" is the output result generated by a generative model based on a prompt.

[0732] The "means for displaying" is a function or device for visually or audibly presenting the response received from the generative model to the user.

[0733] This invention relates to a system that receives user input information, analyzes the user's intentions and emotions, automatically generates appropriate prompts, and sends the prompts to a generative model. This system consists of three elements: a server, a terminal, and a user, and is also combined with an emotion engine.

[0734] System configuration and operation

[0735] User Input

[0736] Users access the system using a device (e.g., smartphone, tablet, PC, etc.). Users can input requests using a variety of methods, including voice recognition, text input, and icon manipulation. For example, a user might input by voice, "I would like to create a summary of the planning meeting."

[0737] Providing an interface

[0738] The device provides an interface for the user to input information, which may include icons, a microphone for voice input, text input fields, etc. As the user provides the input, it is transmitted to the server in real time.

[0739] Analyzing User Intent

[0740] The server analyzes the user's input information received from the device. It uses natural language processing (NLP) techniques to understand the user's request and identify the user's intent. This analysis generates a prompt, such as "Summary of the planning meeting, including key points."

[0741] User Emotion Recognition

[0742] In addition, the server uses an emotion engine to analyze the user's emotions, recognizing the user's emotions from voice and text data and identifying emotional states such as "happiness," "sadness," and "surprise."

[0743] Automatically Generated Prompts

[0744] Based on the analyzed user intent and emotional information, the server automatically generates an appropriate prompt. For example, if a user inputs "Please create a summary of a product review," the server will generate a prompt such as "A summary of a product review, including key points," complementing the user's emotional information.

[0745] Prompting the Generative Model

[0746] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques to generate output based on the user's prompts, which can include text, images, and video data.

[0747] Receiving a response from the generative model

[0748] The results generated by the generative model based on the specified prompts are sent back to the server. For example, if the generative model generates a "summary of the planning meeting" based on the user's request, the summary is sent to the server.

[0749] Displaying the results

[0750] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, the generated summary of a planning meeting can be displayed on the screen, allowing the user to confirm the response to their request.

[0751] Example: A text generation workflow including emotion recognition

[0752] Consider the case where a user voice-inputs "Please create a summary of a product review" into a device. This voice input is received by the device and sent to the server. The server converts the voice data into text and analyzes the user's intent using natural language processing technology. At the same time, it uses an emotion engine to recognize the user's emotions and includes that information in the prompt. For example, it generates a prompt such as "Summary of a product review, including key points, user is satisfied" and sends it to the generative model. The generative model generates a product review summary based on this prompt and sends it back to the server. Finally, the server sends the generated result to the user's device, which displays the summary to the user.

[0753] As described above, the present invention is a system that provides more appropriate responses by eliminating the need for users to manually input detailed prompts, by utilizing generative models simply and efficiently, and by generating prompts that take the user's emotions into consideration.

[0754] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0755] Step 1:

[0756] Receiving user input information

[0757] Users use devices (smartphones, tablets, PCs, etc.) to input information by voice, text, or icon operation. The input information is transferred from the device to the server in real time.

[0758] Input: User's voice command "Please create a summary of the product review."

[0759] Output: User request in text format: "Please create a summary of product reviews."

[0760] Step 2:

[0761] Providing an interface

[0762] The device provides an interface for the user to input information, which may include a microphone for voice input, text entry fields, icons, etc.

[0763] Input: User operations (voice input, text input, icon operation)

[0764] Output: Input data (voice, text data) from the device

[0765] Step 3:

[0766] Transfer of input information to the server

[0767] The terminal transfers the information entered by the user to the server in real time.

[0768] Input: Voice or text data received by the device

[0769] Output: Audio or text data sent to the server

[0770] Step 4:

[0771] Analyzing User Intent

[0772] The server analyzes the received user input (voice or text) and uses natural language processing (NLP) techniques to understand the user's intent. This analysis identifies the intent and generates a specific prompt.

[0773] Input: Voice or text data sent to the server

[0774] Output: Parsed user intent: "Summary of product review, including key points"

[0775] Step 5:

[0776] User sentiment analysis

[0777] The server uses an emotion engine to analyze the user's emotions from the received voice and text data, identifying emotions such as "happiness," "sadness," and "surprise."

[0778] Input: Audio or text data

[0779] Output: Identified user emotional state "Satisfied"

[0780] Step 6:

[0781] Automatically Generated Prompts

[0782] The server automatically generates an appropriate prompt based on the analyzed user intent and emotion information. For example, a prompt such as "Summary of product review, including key points, user satisfied" may be generated.

[0783] Input: Parsed user intent and emotion information

[0784] Output: Auto-generated prompt "Summary of product reviews, including key points, satisfied users"

[0785] Step 7:

[0786] Prompting the Generative Model

[0787] The server sends the generated prompts to a generative AI model, which generates an output based on the prompts.

[0788] Input: Auto-generated prompt

[0789] Output: The prompt sent to the generative AI model

[0790] Step 8:

[0791] Receiving a response from the generative model

[0792] The generative AI model generates output based on the prompts sent and sends the results back to the server.

[0793] Input: Prompt sent

[0794] Output: Generated text "Product Review Summary"

[0795] Step 9:

[0796] Transfer of generated results from the server to the device

[0797] The server transmits the results received from the generative model to the user's terminal.

[0798] Input: The generated results sent from the generative model

[0799] Output: Generated results sent to the terminal

[0800] Step 10:

[0801] Viewing the generated results

[0802] The terminal presents the generated summary of the product review to the user visually or audibly, for example, by displaying it on a screen so that the user can confirm the response to the request.

[0803] Input: Generated result

[0804] Output: The generated result displayed on the user's terminal.

[0805] (Application example 2)

[0806] 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."

[0807] Conventional systems that accept user input can analyze the user's intentions and provide corresponding generated prompts, but they do not take the user's emotions into account when generating prompts, making it impossible to provide responses appropriate to the user's psychological state.In addition, there was a lack of concrete application examples for improving the user's shopping experience, making it difficult to improve customer satisfaction in physical stores.

[0808] The identification process 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 receiving user input information, means for analyzing the user's intention from the received input information, means for analyzing the user's emotion from the received input information, means for automatically generating an appropriate prompt based on the analyzed intention and emotion, means for sending the generated prompt to the generative model, means for receiving a response from the generative model, and means for displaying the received response to the user. This makes it possible to provide a more accurate and satisfying response based on the user's intention and emotion.

[0809] "User input information" refers to information input by the user through various means such as voice, text, or icon operation.

[0810] "Intent analysis" is the process of determining the content and purpose of received user input.

[0811] "Emotion analysis" is the process of identifying a user's emotional state from speech or text data, for example, detecting emotions such as joy, sadness, or surprise.

[0812] "Automatic prompt generation" refers to the process by which the system automatically generates appropriate instructions or commands based on the analyzed intent and sentiment.

[0813] A "generative model" is a system that uses artificial intelligence technology to generate text, images, and videos based on user requests.

[0814] "Receiving a response" refers to the process by which the server receives the results generated from the Generative Model.

[0815] "Displaying to the user" refers to the process of presenting the generated results to the user through the screen or audio of the device they use.

[0816] This invention relates to a system that receives user input, analyzes the user's intentions and emotions, automatically generates appropriate prompts, and sends the prompts to a generative model. This system mainly consists of three elements: a server, a terminal, and a user, and is also combined with an emotion engine.

[0817] System configuration and operation

[0818] User Input

[0819] Users access the system using a device (e.g., a smartphone or smart glasses). Users can input requests using a variety of methods, including voice recognition, text input, and icon manipulation. For example, a user might say, "Please give me a review of this bag."

[0820] Providing an interface

[0821] The device provides an interface for the user to input information, including a microphone for voice input, text input fields, and a UI for icon manipulation. As the user provides input, it is transmitted to the server in real time.

[0822] Analyzing user intent and emotions

[0823] The server analyzes the user input information received from the device. It uses natural language processing (NLP) technology to understand the user's request and identify the user's intent. This analysis is performed using, for example, the TextBlob library. It also analyzes the user's emotions using an emotion engine. It recognizes the user's emotions from voice and text data and identifies emotional states such as "happy," "unhappy," or "neutral." This process uses Google's speech recognition API and TextBlob library.

[0824] Automatically Generated Prompts

[0825] Based on the analyzed user intent and emotion information, the server automatically generates an appropriate prompt. For example, if a user inputs "Please give me a review about this bag," the server generates a prompt such as "The user said the following in a happy state: Please give me a review about this bag. Please generate the optimal product review based on this," and sends it to the generative model.

[0826] Prompting the Generative Model

[0827] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques to generate output based on the user's prompts. This output can include text, images, and video data. Specifically, OpenAI's API is used.

[0828] Receiving a response from the generative model

[0829] The results generated by the generative model based on the specified prompts are sent back to the server. For example, if the generative model generates a "review of this bag" based on the user's request, the review is sent to the server.

[0830] Displaying the results

[0831] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, the generated bag review can be displayed on the screen, allowing the user to confirm the response to their request.

[0832] Specific examples

[0833] Consider a scenario where a user uses smart glasses to voice-input "Please give me a review about this bag." This voice input is received by the device and sent to a server, which converts the voice data into text and analyzes the user's intent and emotions using natural language processing technology and an emotion engine. Based on the analysis, the server generates the following prompt:

[0834] Example prompt sentence:

[0835] In the delighted state, the user said: Give me a review about this bag

[0836] Generate optimal product reviews based on this.

[0837] The generative model generates product reviews based on the prompts and sends them back to the server, which then sends the generated results to the user's device, which then displays the reviews to the user.

[0838] As described above, the present invention is a system that generates prompts based on the user's intentions and emotions, thereby providing more accurate responses.

[0839] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0840] Step 1:

[0841] The user inputs a request to the terminal by voice or text. For example, the user inputs "Please give me a review about this bag." This voice data is received by the terminal.

[0842] Step 2:

[0843] The device converts the voice data into text data using Google's speech recognition API, resulting in the text data "Write a review about this bag."

[0844] Step 3:

[0845] The device sends text data to the server, which then receives the text and analyzes the user's intent. This is done using natural language processing (NLP) techniques, such as the TextBlob library. The analysis identifies the user's intent as a request for a bag review.

[0846] Step 4:

[0847] At the same time, the server analyzes the user's emotions from the text data using the TextBlob library, for example, to identify the emotion "joy."

[0848] Step 5:

[0849] The server automatically generates prompts based on the analyzed user intent and emotion, such as "The user said the following in the joy state: Give us your review of this bag. Based on this, generate the best product review."

[0850] Step 6:

[0851] The server sends the automatically generated prompts to a generative model, which uses OpenAI's API, and the generative model receives the prompts and generates a product review based on them.

[0852] Step 7:

[0853] The product reviews generated by the generative model are sent back to the server. For example, a review might be generated that reads, "This bag is very user-friendly and has an excellent design. It is particularly notable for the high-quality materials used."

[0854] Step 8:

[0855] The server sends the reviews received from the generative model to the device, which receives the reviews and displays them visually or audibly to the user, allowing the user to confirm the response to their request.

[0856] Through the above processing steps, the user can obtain accurate product reviews based on his or her own intentions and feelings.

[0857] 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.

[0858] 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.

[0859] 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.

[0860] [Third embodiment]

[0861] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0862] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0863] 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).

[0864] 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.

[0865] 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.

[0866] 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).

[0867] 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.

[0868] 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.

[0869] 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.

[0870] 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.

[0871] 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.

[0872] 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."

[0873] The present invention relates to a system that utilizes a generative model using user input information, and a specific embodiment thereof will be described below. The system consists of three elements: a server, a terminal, and a user.

[0874] System configuration and operation

[0875] User Input

[0876] A user accesses the system using a device (e.g., a smartphone, tablet, or PC). At this stage, the user can select from a variety of input methods, such as voice recognition, text input, and icon manipulation, to input requests to the generative model. For example, a user may input in text format, "I would like to create a summary of the planning meeting," or make a similar request by voice.

[0877] Providing an interface

[0878] The terminal provides an interface for the user to provide input information, which may include a user interface including icons, microphone access permission for voice input, text input fields, etc. When the user inputs information through these interfaces, the information is transmitted to the server in real time.

[0879] Analyzing User Intent

[0880] The server analyzes the user's input information received from the device. Using natural language processing (NLP) technology, it analyzes the user's text and voice data to understand the user's intent. This analysis allows the server to determine what the user is looking for and generate appropriate prompts based on that. As a result of this analysis, the server may generate a prompt such as "Summary of the planning meeting, including key points."

[0881] Automatically Generated Prompts

[0882] Based on the analyzed user intent, the server automatically generates a prompt containing the necessary information. For example, if a user inputs "Please create a summary of a product review," the server generates a prompt "Summary of a product review, including key points." This prompt is sent to the generative model.

[0883] Prompting the Generative Model

[0884] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques (e.g., a Generative Pre-trained Transformer) to generate output based on the user's prompts. This output can include text, images, video data, etc.

[0885] Receiving a response from the generative model

[0886] The results generated by the generative model based on the specified prompts are sent back to the server. For example, if the generative model generates a "summary of the planning meeting" based on the user's request, the summary is sent to the server.

[0887] Displaying the results

[0888] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. This step allows the user to easily check the generated output, such as text, images, and videos. For example, a summary of a planning meeting generated can be displayed on the screen, allowing the user to confirm the response to their request.

[0889] Example: A series of steps for generating sentences

[0890] Consider the case where a user voice-inputs "Please create a summary of a product review" into a device. This voice input is received by the device and sent to the server. The server converts the voice data into text and analyzes the user's intent through natural language processing. Based on the results, it automatically generates a prompt "Summary of a product review, including key points" and sends it to the generative model. The generative model generates a product review summary based on this prompt and sends it back to the server. Finally, the server sends the received summary to the user's device, which displays it to the user.

[0891] As described above, the present invention provides a system for easily and efficiently using generative models, eliminating the need for users to manually input detailed prompts.

[0892] The processing flow will be explained below.

[0893] Step 1:

[0894] A user accesses a terminal.

[0895] Action: A user launches an application on the system using a device such as a smartphone or PC.

[0896] Step 2:

[0897] The user provides the input information.

[0898] Action: User clicks icon or speaks or writes "I want to create a summary of the planning meeting."

[0899] Step 3:

[0900] The terminal receives the input information.

[0901] How it works: The device receives the user's clicks, voice, and text input in real time and transmits that information to the server.

[0902] Step 4:

[0903] The server analyzes the input information.

[0904] How it works: The server analyzes the data received from the device, converts voice data into text, and uses natural language processing (NLP) techniques to understand the user's intent.

[0905] Step 5:

[0906] The server generates the appropriate prompt.

[0907] How it works: Based on the analysis results, the server automatically generates a prompt such as "Summary of the planning meeting, including key points."

[0908] Step 6:

[0909] The server sends a prompt to the generative model.

[0910] How it works: The server sends the generated prompt as an HTTP request to the API endpoint of the generating AI.

[0911] Step 7:

[0912] The server receives the response from the generative model.

[0913] How it works: The generative AI model generates output based on the prompts, and the results are sent back to the server.

[0914] Step 8:

[0915] The server sends the results to the terminal.

[0916] How it works: The server formats the generated output and creates data packets to send to the user's terminal.

[0917] Step 9:

[0918] The terminal displays the results to the user.

[0919] Operation: The device visualizes the received data and presents it to the user. For example, it displays a summary of the planning meeting on the screen.

[0920] Examples:

[0921] 1. The user speaks, "I want a summary of a product review."

[0922] 2. The device sends the audio data to the server.

[0923] 3. The server transcribes the audio data, performs natural language processing, and generates prompts.

[0924] 4. The server sends a prompt to the generative model.

[0925] 5. The generative model generates a summary of the review and sends it back to the server.

[0926] 6. The server sends the results to the user's terminal, which displays a summary to the user.

[0927] In this way, the system automatically performs analysis and generation without the user having to enter complex prompts, making generative models easier to use.

[0928] Example 1

[0929] 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."

[0930] When users use complex generative models, many detailed operations are required, which places a significant burden on them. In particular, when using voice or text input, users must manually input appropriate prompts into the generative model, which is a time-consuming and labor-intensive task. A solution to this problem is needed that allows users to easily use generative models.

[0931] 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.

[0932] In this invention, the server includes a means for converting voice data into text data, a means for analyzing the converted text data using natural language processing technology, and a means for formatting the generated prompt into a format that is easy for the generative model to understand, thereby enabling the user to use the generative model simply and quickly without performing complex operations.

[0933] "User input information" refers to data that a user provides to the system in the form of voice, text, icon manipulation, or the like.

[0934] "Means for receiving" refers to a function for receiving information input by a user and incorporating it into the system.

[0935] "Means for analyzing" refers to methods and techniques for processing received user input information and understanding its content.

[0936] A "prompt" is a short instruction that specifically specifies the desired output from the generative model.

[0937] A "generative model" refers to an artificial intelligence system that generates text, images, video, etc. based on a given prompt.

[0938] "Means of conversion" refers to the methods and techniques for converting audio data into text data.

[0939] "Natural language processing technology" refers to a technological system that enables computers to understand and analyze human language and perform appropriate information processing.

[0940] "Formatting means" refers to a method for adjusting the format of the generated prompt so that the generative model can efficiently understand it and generate a response.

[0941] A "response" refers to the output that a generative model generates based on a prompt.

[0942] "Means for displaying" refers to a function for visually or audibly presenting the response received from the generative model to the user.

[0943] The present invention relates to a system that utilizes a generative model based on user input information. A specific embodiment of this system is described below. The system consists of three elements: a server, a terminal, and a user.

[0944] System configuration and operation

[0945] User Input

[0946] Users access the system using devices such as smartphones, tablets, and PCs. When sending requests to the system, users can choose from a variety of input methods, such as voice recognition, text input, and icon manipulation. For example, a user may input a request in text format, such as "Please create a product review summary." Alternatively, a user may make a similar request by voice.

[0947] Providing an interface

[0948] The terminal provides an interface for the user to provide input information, including a text input field, a user interface including icons, and microphone access permission for voice input. When the user inputs information through these interfaces, the information is transmitted to the server in real time.

[0949] Analyzing User Intent

[0950] The server analyzes the user's input information sent from the device. It uses natural language processing (NLP) techniques to analyze text and voice data and understand the user's intent. For example, it uses an NLP engine (e.g., a common NLP library or API) to identify that the user's request is to "create a summary of a product review."

[0951] Automatically Generated Prompts

[0952] Based on the analysis results, the server automatically generates a prompt containing the necessary information. For example, it generates a prompt such as "A summary of product reviews, including key points." This prompt is formatted in a way that is easy for the generative model to understand.

[0953] Prompting the Generative Model

[0954] The server sends the generated prompts to a generative model (e.g., an artificial intelligence system such as a Generative Pre-trained Transformer) that uses the prompts to generate output for the user's request, such as text, images, or video. The generative model uses advanced AI algorithms to generate the requested data in real time.

[0955] Receiving a response from the generative model

[0956] The output generated by the generative model is sent back to the server. For example, a "product review summary" is generated and sent to the server. This generated result is sent as structured data, such as text.

[0957] Displaying the results

[0958] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. The device displays the generated text, image, or video output to the user on a screen or using a voice assistant. For example, a generated "product review summary" is displayed on the device screen.

[0959] Specific examples

[0960] A specific example is a case where the user vocally inputs into the terminal, "I want you to create a review summary of the product."

[0961] 1. User input:

[0962] The user speaks to the terminal, "I want a review summary of the product."

[0963] 2. Receiving audio data:

[0964] The device provides microphone access permission for voice input and transmits the received voice data to the server, for example, when the user makes a request to "activate the voice assistant."

[0965] 3. Text Conversion and Intent Analysis:

[0966] The server converts the voice data into text and analyzes the user's intent using natural language processing technology. The server converts the voice data into text using the Google Speech-to-Text API, and then analyzes the intent using a common NLP engine.

[0967] 4. Generate prompt:

[0968] The server generates a prompt such as "Product review summary, including key points" and sends this prompt to the generative model.

[0969] 5. Generative model response:

[0970] A generative model (e.g., GPT-3) generates a review summary based on the prompts and sends the summary back to the server.

[0971] 6. View the response:

[0972] The server sends the summary to the device, which then displays it to the user. The device screen displays the review summary: "I found this product easy to use and very durable. I was particularly impressed by its light weight."

[0973] By following the above steps, users can use generative AI models easily and efficiently.

[0974] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0975] Step 1: User Input

[0976] Users access the system using devices such as smartphones, tablets, and PCs. For example, a user inputs a request in text format, such as "Please create a product review summary." The user's input is entered into the device as text data, such as "Please create a product review summary." This input data is sent to the server through the device's interface.

[0977] Step 2: Providing an interface

[0978] The device provides a user interface that includes text input fields, icons, and microphone access permission for voice input. When a user inputs information through these interfaces, the information is transmitted to the server in real time. This transmitted data becomes the input information required for the next analysis step.

[0979] Step 3: Analyze user intent

[0980] The server analyzes the user's input information received from the terminal. Specifically, it analyzes the text data using natural language processing technology (e.g., a standard NLP engine). The server receives the text data "Please create a product review summary" as input and analyzes this data to identify the user's intent. The output of the analysis is the intent to "generate a product review summary."

[0981] Step 4: Automatically generate prompts

[0982] The server automatically generates a prompt containing the necessary information based on the analyzed user intent. For example, it generates a prompt such as "Product review summary, including important points." The prompt generated by the server is text data such as "Product review summary, including important points." This prompt is sent as input to the generative model.

[0983] Step 5: Sending prompts to the generative model

[0984] The server sends the generated prompt to a generative model (e.g., a Generative Pre-trained Transformer), which generates an output based on the prompt. The generative model receives the prompt "Product review summary, including key points" as input and generates the appropriate text output. The generated output (e.g., "Product review summary") is sent back to the server.

[0985] Step 6: Receiving a response from the generative model

[0986] The server receives the output results returned from the generative model. Text data containing the "product review summary" generated by the generative model is returned to the server. This received data becomes the input data for proceeding to the next display step.

[0987] Step 7: View the results

[0988] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, specific text data for displaying a "product review summary" is sent to the device and displayed to the user on the device screen in the form of "I thought this product was easy to use and very durable. I was particularly impressed by its light weight." This displayed text is the final output, and the user can confirm the generated information.

[0989] These are the specific processing steps of this system. Each step starts with user input, then includes analysis, prompt generation, output of the generative model, and finally display to the user.

[0990] (Application example 1)

[0991] 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."

[0992] In autonomous vehicles, there is a need for a means to efficiently provide passengers with the information they request. In particular, there is a need for a system that can instantly obtain the information passengers need using voice input or screen operation, and provide it on the spot by display or voice. It is also important that this will improve passenger satisfaction and enhance the convenience of autonomous vehicles.

[0993] 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.

[0994] In this invention, the server includes means for receiving user input information, means for analyzing the user's intention from the received input information, means for automatically generating appropriate prompts based on the analyzed intention, means for transmitting the generated prompts to a generative model, means for receiving a response from the generative model, means for displaying the received response to the user, and means for providing information based on passenger questions using a voice recognition and display device installed in the autonomous vehicle. This enables passengers to instantly obtain necessary information through voice input or screen operation, improving the user experience in the autonomous vehicle.

[0995] "User input information" refers to information such as voice, text, and icon operations that the user provides to the system.

[0996] The "receiving means" is a mechanism for receiving user input information at a terminal or server.

[0997] The "means for analyzing intent" is a method for understanding and analyzing the user's intent based on the received input information.

[0998] The "means for automatically generating a prompt" is a function that automatically generates an appropriate instruction sentence based on the analyzed user's intention.

[0999] A "generative model" is an artificial intelligence technique for generating text, images, videos, etc. based on specific prompts.

[1000] A "means for sending a prompt" is a mechanism for passing a generated prompt to a Generative Model.

[1001] The "means for receiving a response" is a mechanism for receiving a response from the Generative Model.

[1002] A "display means" is a device for visually or audibly presenting the received response to the user.

[1003] An "autonomous vehicle" is a vehicle that can perform driving operations automatically.

[1004] A "voice recognition device" is a device for converting passenger voice input into text data.

[1005] A "display device" is a display device for displaying data such as text and images.

[1006] A "passenger" is a person using an autonomous vehicle.

[1007] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described in detail with reference to the accompanying drawings.

[1008] System configuration

[1009] This system consists of three elements: a server, a terminal, and a user, in order to realize information provision services in autonomous vehicles.

[1010] 1. Receiving user input information

[1011] Users input information using terminals (for example, a voice recognition device or a display device) installed in the autonomous vehicle. The voice recognition device is a device that converts passengers' voice input into text data. The display device is a device that displays data such as text and images. Users make requests by voice, such as "Please tell me about nearby tourist spots."

[1012] 2. Providing an interface

[1013] The terminal provides an interface for the user to provide input information, which may include a microphone for voice input, a display screen, etc. The information input by the user is transmitted to the server in real time.

[1014] 3. Analyzing User Intent

[1015] The server analyzes the user's input information received from the device. It uses natural language processing (NLP) technology to analyze the user's voice data and understand the user's intent. Through this analysis, the server identifies the information the user is seeking and generates appropriate prompts based on that information.

[1016] 4. Automatically Generate Prompts

[1017] Based on the analyzed user intent, the server automatically generates a prompt containing the necessary information, for example, "A passenger is asking about nearby tourist attractions. Please provide the most suitable tourist attraction information based on the following."

[1018] 5. Sending prompts to the generative model

[1019] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques such as Generative Pre-trained Transformer (GPT) to generate outputs based on the user's prompts.

[1020] 6. Receiving a response from the generative model

[1021] The results that the generative model generates based on the specified prompts are sent back to the server. For example, if the generative model generates a "list of nearby tourist attractions" based on the user's request, that information is sent to the server.

[1022] 7. Displaying the results

[1023] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, the generated tourist information is displayed on a screen or guided by voice.

[1024] Specific examples

[1025] When a passenger says, "Please tell me about nearby tourist spots in this car," the voice input is received by the device and sent to the server. The server converts the voice data into text and analyzes the user's intention through natural language processing. As a result, it automatically generates a "prompt about nearby tourist spots" and sends it to the generative model. The generative model generates tourist spot information based on this prompt and sends it back to the server. Finally, the server sends the received tourist spot information to the user's device and displays it on the screen.

[1026] Prompt Sentence Examples

[1027] "A passenger is asking about nearby tourist attractions. Please provide the best tourist information based on the following:

[1028] Question: Please tell me about nearby tourist spots

[1029] Required information: name of tourist spot, distance, highlights

[1030] As described above, the present invention is a system that allows users to instantly obtain necessary information through voice input or screen operation, thereby improving the user experience in an autonomous vehicle.

[1031] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1032] Step 1:

[1033] The user enters information by voice.

[1034] Input: User's voice (e.g., "Please tell me about nearby tourist spots")

[1035] Output: Audio data

[1036] Specific operation: A user provides speech input to a speech recognition device installed in an autonomous vehicle, which receives this input as digital voice data.

[1037] Step 2:

[1038] The terminal converts the voice data into text data.

[1039] Input: Digital audio data

[1040] Output: Text data

[1041] What it does: The device's voice recognition software (e.g., Google Speech-to-Text API) analyzes the voice data and converts it into corresponding text data, such as "Please tell me about nearby tourist spots."

[1042] Step 3:

[1043] The server analyzes the text data and understands the user's intent.

[1044] Input: Text data

[1045] Output: Intent information (e.g., "I'm looking for information about tourist spots")

[1046] Specific operation: The server analyzes text data using a natural language processing (NLP) engine (e.g., SpaCy or GPT-3). From the analyzed data, it understands the user's intent and generates intent information according to the purpose.

[1047] Step 4:

[1048] The server automatically generates prompts based on the user's intent.

[1049] Input: Intention information

[1050] Output: The generated prompt (e.g., "Please tell me about nearby tourist attractions")

[1051] What it does: The server uses the parsed intent information to construct an appropriate prompt, which is automatically generated using a preset template.

[1052] Step 5:

[1053] The server sends the generated prompt to the Generative Model.

[1054] Input: Generated prompt

[1055] Output: A request to the generative model with the prompt sent

[1056] Specific behavior: The server sends the generated prompt via an HTTP request to the generative model (e.g., OpenAI API).

[1057] Step 6:

[1058] The generative model generates a response based on the prompt and sends it back to the server.

[1059] Input: prompt

[1060] Output: Generated response (e.g., list of tourist attractions)

[1061] Specific operation: The generative model analyzes the prompt and generates an appropriate response, which is sent back to the server in the form of text data, image data, or other data.

[1062] Step 7:

[1063] The server receives the response from the generative model and sends it to the terminal.

[1064] Input: Generated response

[1065] Output: Response data sent to the terminal

[1066] Specific operation: The server transfers the response data received from the generative model via the network to the terminal.

[1067] Step 8:

[1068] The terminal displays or audibly presents the received response to the user.

[1069] Input: Response data (e.g., a list of tourist attractions)

[1070] Output: Display or audio output to the user

[1071] Specific actions: Display a list of tourist attractions on the device's display device, or use a speech synthesizer to provide the information to the user aloud.

[1072] 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.

[1073] This invention relates to a system that receives user input information, analyzes the user's intentions and emotions, automatically generates appropriate prompts, and sends the prompts to a generative model. This system consists of three elements: a server, a terminal, and a user, and is also combined with an emotion engine.

[1074] System configuration and operation

[1075] User Input

[1076] Users access the system using a device (e.g., smartphone, tablet, PC, etc.). Users can input requests using a variety of methods, including voice recognition, text input, and icon manipulation. For example, a user might input by voice, "I would like to create a summary of the planning meeting."

[1077] Providing an interface

[1078] The device provides an interface for the user to input information, which may include icons, a microphone for voice input, text input fields, etc. As the user provides the input, it is transmitted to the server in real time.

[1079] Analyzing user intent

[1080] The server analyzes the user's input information received from the device. It uses natural language processing (NLP) techniques to understand the user's request and identify the user's intent. This analysis generates a prompt, such as "Summary of the planning meeting, including key points."

[1081] Recognizing user emotions

[1082] In addition, the server uses an emotion engine to analyze the user's emotions, recognizing the user's emotions from voice and text data and identifying emotional states such as "happiness," "sadness," and "surprise."

[1083] Automatically Generated Prompts

[1084] Based on the analyzed user intent and emotional information, the server automatically generates an appropriate prompt. For example, if a user inputs "Please create a summary of a product review," the server will generate a prompt such as "A summary of a product review, including key points," complementing the user's emotional information.

[1085] Prompting the Generative Model

[1086] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques to generate output based on the user's prompts, which can include text, images, and video data.

[1087] Receiving a response from the generative model

[1088] The results generated by the generative model based on the specified prompts are sent back to the server. For example, if the generative model generates a "summary of the planning meeting" based on the user's request, the summary is sent to the server.

[1089] Displaying the results

[1090] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, the generated summary of a planning meeting can be displayed on the screen, allowing the user to confirm the response to their request.

[1091] Example: A text generation workflow including emotion recognition

[1092] Consider the case where a user voice-inputs "Please create a summary of a product review" into a device. This voice input is received by the device and sent to the server. The server converts the voice data into text and analyzes the user's intent using natural language processing technology. At the same time, it uses an emotion engine to recognize the user's emotions and includes that information in the prompt. For example, it generates a prompt such as "Summary of a product review, including key points, user is satisfied" and sends it to the generative model. The generative model generates a product review summary based on this prompt and sends it back to the server. Finally, the server sends the generated result to the user's device, which displays the summary to the user.

[1093] As described above, the present invention is a system that provides more appropriate responses by eliminating the need for users to manually input detailed prompts, by utilizing generative models simply and efficiently, and by generating prompts that take the user's emotions into consideration.

[1094] The processing flow will be explained below.

[1095] Step 1:

[1096] A user accesses a terminal.

[1097] Action: A user launches an application on the system using a device such as a smartphone or PC.

[1098] Step 2:

[1099] The user provides the input information.

[1100] Action: User clicks icon or speaks or writes "I want to create a summary of the planning meeting."

[1101] Step 3:

[1102] The terminal receives the input information.

[1103] How it works: The device receives the user's clicks, voice, and text input in real time and transmits that information to the server.

[1104] Step 4:

[1105] The server analyzes the input information.

[1106] How it works: The server analyzes the data received from the device, converts voice data into text, and uses natural language processing (NLP) techniques to understand the user's intent.

[1107] Step 5:

[1108] The server recognizes the user's emotions.

[1109] How it works: The server uses an emotion engine to analyze the user's emotional state from the received voice and text data, for example, identifying emotions such as "happiness," "sadness," or "surprise" from the user's tone of voice and the context of the text.

[1110] Step 6:

[1111] The server generates the appropriate prompt.

[1112] How it works: Based on the analyzed user intent and emotion information, the server automatically generates a prompt such as "Summary of the planning meeting, including key points." It may also fine-tune the prompt based on the recognized emotion.

[1113] Step 7:

[1114] The server sends a prompt to the generative model.

[1115] How it works: The server sends the generated prompt to the API endpoint of the AI ​​generator as an HTTP request. The prompt also contains the user's emotional information.

[1116] Step 8:

[1117] The server receives the response from the generative model.

[1118] How it works: The generative model generates output based on the prompts, and the results are sent back to the server, for example, a summary of the analyzed planning meeting.

[1119] Step 9:

[1120] The server sends the results to the terminal.

[1121] How it works: The server formats the generated output and creates data packets to send to the user's terminal.

[1122] Step 10:

[1123] The terminal displays the results to the user.

[1124] Operation: The device visualizes the received data and presents it to the user. For example, it displays a summary of the planning meeting on the screen.

[1125] Examples:

[1126] 1. The user speaks, "I want a summary of a product review."

[1127] 2. The device sends the audio data to the server.

[1128] 3. The server transcribes the audio data, performs natural language processing, and analyzes the user's intent.

[1129] 4. The server uses the emotion engine to recognize the user's emotion as "satisfied."

[1130] 5. The server generates a prompt: "Summary of product review, including key points, user satisfied."

[1131] 6. The server sends a prompt to the generative model.

[1132] 7. The generative model generates a summary of the review and sends it back to the server.

[1133] 8. The server sends the results to the user's device.

[1134] 9. The terminal displays the summary to the user.

[1135] In this way, the burden on the user to input detailed prompts is reduced, and the system automatically analyzes and generates, making it possible to effectively utilize generative models and provide more appropriate responses that take the user's emotions into account.

[1136] Example 2

[1137] 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."

[1138] In conventional systems, users have to manually input detailed prompts, which is time-consuming, and it is difficult to utilize generative models that take into account the user's intentions and emotions. As a result, the responses users receive are often inappropriate.

[1139] 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.

[1140] In this invention, the server includes means for receiving user input information, means for analyzing the user's intention from the received input information, and means for automatically generating appropriate prompts based on the analyzed intention and the user's emotional information. This allows for the automatic generation of prompts that take the user's intention and emotion into consideration, making it possible to produce more appropriate and accurate output from the generative model.

[1141] "User input information" refers to any information provided by the user to the system, including voice, text, icon operations, and the like.

[1142] The "receiving means" refers to a function or device that allows the system to receive input information provided by the user.

[1143] "User intent" refers to the user's goals and desires as interpreted by the system based on input information.

[1144] "Means for analyzing" refers to technology or devices for analyzing received input information and understanding the user's intentions.

[1145] "Emotion information" refers to the user's emotional state analyzed from voice and text data, and includes "joy," "sadness," "surprise," and the like.

[1146] A "prompt" is text information that provides instructions or guidelines for a generative model when generating output.

[1147] "Automatic generation means" refers to technology or devices for automatically creating appropriate prompts based on the user's intentions and emotional information.

[1148] A "generative model" is a model that includes artificial intelligence techniques that generate output, such as text, images, or video, based on input prompts.

[1149] The "sending means" is a function or device for sending the generated prompt to the Generative Model.

[1150] A "response" is the output result generated by a generative model based on a prompt.

[1151] The "means for displaying" is a function or device for visually or audibly presenting the response received from the generative model to the user.

[1152] This invention relates to a system that receives user input information, analyzes the user's intentions and emotions, automatically generates appropriate prompts, and sends the prompts to a generative model. This system consists of three elements: a server, a terminal, and a user, and is also combined with an emotion engine.

[1153] System configuration and operation

[1154] User Input

[1155] Users access the system using a device (e.g., smartphone, tablet, PC, etc.). Users can input requests using a variety of methods, including voice recognition, text input, and icon manipulation. For example, a user might input by voice, "I would like to create a summary of the planning meeting."

[1156] Providing an interface

[1157] The device provides an interface for the user to input information, which may include icons, a microphone for voice input, text input fields, etc. As the user provides the input, it is transmitted to the server in real time.

[1158] Analyzing User Intent

[1159] The server analyzes the user's input information received from the device. It uses natural language processing (NLP) techniques to understand the user's request and identify the user's intent. This analysis generates a prompt, such as "Summary of the planning meeting, including key points."

[1160] User Emotion Recognition

[1161] In addition, the server uses an emotion engine to analyze the user's emotions, recognizing the user's emotions from voice and text data and identifying emotional states such as "happiness," "sadness," and "surprise."

[1162] Automatically Generated Prompts

[1163] Based on the analyzed user intent and emotional information, the server automatically generates an appropriate prompt. For example, if a user inputs "Please create a summary of a product review," the server will generate a prompt such as "A summary of a product review, including key points," complementing the user's emotional information.

[1164] Prompting the Generative Model

[1165] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques to generate output based on the user's prompts, which can include text, images, and video data.

[1166] Receiving a response from the generative model

[1167] The results generated by the generative model based on the specified prompts are sent back to the server. For example, if the generative model generates a "summary of the planning meeting" based on the user's request, the summary is sent to the server.

[1168] Displaying the results

[1169] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, the generated summary of a planning meeting can be displayed on the screen, allowing the user to confirm the response to their request.

[1170] Example: A text generation workflow including emotion recognition

[1171] Consider the case where a user voice-inputs "Please create a summary of a product review" into a device. This voice input is received by the device and sent to the server. The server converts the voice data into text and analyzes the user's intent using natural language processing technology. At the same time, it uses an emotion engine to recognize the user's emotions and includes that information in the prompt. For example, it generates a prompt such as "Summary of a product review, including key points, user is satisfied" and sends it to the generative model. The generative model generates a product review summary based on this prompt and sends it back to the server. Finally, the server sends the generated result to the user's device, which displays the summary to the user.

[1172] As described above, the present invention is a system that provides more appropriate responses by eliminating the need for users to manually input detailed prompts, by utilizing generative models simply and efficiently, and by generating prompts that take the user's emotions into consideration.

[1173] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1174] Step 1:

[1175] Receiving user input information

[1176] Users use devices (smartphones, tablets, PCs, etc.) to input information by voice, text, or icon operation. The input information is transferred from the device to the server in real time.

[1177] Input: User's voice command "Please create a summary of the product review."

[1178] Output: User request in text format: "Please create a summary of product reviews."

[1179] Step 2:

[1180] Providing an interface

[1181] The device provides an interface for the user to input information, which may include a microphone for voice input, text entry fields, icons, etc.

[1182] Input: User operations (voice input, text input, icon operation)

[1183] Output: Input data (voice, text data) from the device

[1184] Step 3:

[1185] Transfer of input information to the server

[1186] The terminal transfers the information entered by the user to the server in real time.

[1187] Input: Voice or text data received by the device

[1188] Output: Audio or text data sent to the server

[1189] Step 4:

[1190] Analyzing User Intent

[1191] The server analyzes the received user input (voice or text) and uses natural language processing (NLP) techniques to understand the user's intent. This analysis identifies the intent and generates a specific prompt.

[1192] Input: Voice or text data sent to the server

[1193] Output: Parsed user intent: "Summary of product review, including key points"

[1194] Step 5:

[1195] User sentiment analysis

[1196] The server uses an emotion engine to analyze the user's emotions from the received voice and text data, identifying emotions such as "happiness," "sadness," and "surprise."

[1197] Input: Audio or text data

[1198] Output: Identified user emotional state "Satisfied"

[1199] Step 6:

[1200] Automatically Generated Prompts

[1201] The server automatically generates an appropriate prompt based on the analyzed user intent and emotion information. For example, a prompt such as "Summary of product review, including key points, user satisfied" may be generated.

[1202] Input: Parsed user intent and emotion information

[1203] Output: Auto-generated prompt "Summary of product reviews, including key points, satisfied users"

[1204] Step 7:

[1205] Prompting the Generative Model

[1206] The server sends the generated prompts to a generative AI model, which generates an output based on the prompts.

[1207] Input: Auto-generated prompt

[1208] Output: The prompt sent to the generative AI model

[1209] Step 8:

[1210] Receiving a response from the generative model

[1211] The generative AI model generates output based on the prompts sent and sends the results back to the server.

[1212] Input: Prompt sent

[1213] Output: Generated text "Product Review Summary"

[1214] Step 9:

[1215] Transfer of generated results from the server to the device

[1216] The server transmits the results received from the generative model to the user's terminal.

[1217] Input: The generated results sent from the generative model

[1218] Output: Generated results sent to the terminal

[1219] Step 10:

[1220] Viewing the generated results

[1221] The terminal presents the generated summary of the product review to the user visually or audibly, for example, by displaying it on a screen so that the user can confirm the response to the request.

[1222] Input: Generated result

[1223] Output: The generated result displayed on the user's terminal.

[1224] (Application example 2)

[1225] 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."

[1226] Conventional systems that accept user input can analyze the user's intentions and provide corresponding generated prompts, but they do not take the user's emotions into account when generating prompts, making it impossible to provide responses appropriate to the user's psychological state.In addition, there was a lack of concrete application examples for improving the user's shopping experience, making it difficult to improve customer satisfaction in physical stores.

[1227] The identification process 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 receiving user input information, means for analyzing the user's intention from the received input information, means for analyzing the user's emotion from the received input information, means for automatically generating an appropriate prompt based on the analyzed intention and emotion, means for sending the generated prompt to the generative model, means for receiving a response from the generative model, and means for displaying the received response to the user. This makes it possible to provide a more accurate and satisfying response based on the user's intention and emotion.

[1228] "User input information" refers to information input by the user through various means such as voice, text, or icon operation.

[1229] "Intent analysis" is the process of determining the content and purpose of received user input.

[1230] "Emotion analysis" is the process of identifying a user's emotional state from speech or text data, for example, detecting emotions such as joy, sadness, or surprise.

[1231] "Automatic prompt generation" refers to the process by which the system automatically generates appropriate instructions or commands based on the analyzed intent and sentiment.

[1232] A "generative model" is a system that uses artificial intelligence technology to generate text, images, and videos based on user requests.

[1233] "Receiving a response" refers to the process by which the server receives the results generated from the Generative Model.

[1234] "Displaying to the user" refers to the process of presenting the generated results to the user through the screen or audio of the device they use.

[1235] This invention relates to a system that receives user input, analyzes the user's intentions and emotions, automatically generates appropriate prompts, and sends the prompts to a generative model. This system mainly consists of three elements: a server, a terminal, and a user, and is also combined with an emotion engine.

[1236] System configuration and operation

[1237] User Input

[1238] Users access the system using a device (e.g., a smartphone or smart glasses). Users can input requests using a variety of methods, including voice recognition, text input, and icon manipulation. For example, a user might say, "Please give me a review of this bag."

[1239] Providing an interface

[1240] The device provides an interface for the user to input information, including a microphone for voice input, text input fields, and a UI for icon manipulation. As the user provides input, it is transmitted to the server in real time.

[1241] Analyzing user intent and emotions

[1242] The server analyzes the user input information received from the device. It uses natural language processing (NLP) technology to understand the user's request and identify the user's intent. This analysis is performed using, for example, the TextBlob library. It also analyzes the user's emotions using an emotion engine. It recognizes the user's emotions from voice and text data and identifies emotional states such as "happy," "unhappy," or "neutral." This process uses Google's speech recognition API and TextBlob library.

[1243] Automatically Generated Prompts

[1244] Based on the analyzed user intent and emotion information, the server automatically generates an appropriate prompt. For example, if a user inputs "Please give me a review about this bag," the server generates a prompt such as "The user said the following in a happy state: Please give me a review about this bag. Please generate the optimal product review based on this," and sends it to the generative model.

[1245] Prompting the Generative Model

[1246] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques to generate output based on the user's prompts. This output can include text, images, and video data. Specifically, OpenAI's API is used.

[1247] Receiving a response from the generative model

[1248] The results generated by the generative model based on the specified prompts are sent back to the server. For example, if the generative model generates a "review of this bag" based on the user's request, the review is sent to the server.

[1249] Displaying the results

[1250] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, the generated bag review can be displayed on the screen, allowing the user to confirm the response to their request.

[1251] Specific examples

[1252] Consider a scenario where a user uses smart glasses to voice-input "Please give me a review about this bag." This voice input is received by the device and sent to a server, which converts the voice data into text and analyzes the user's intent and emotions using natural language processing technology and an emotion engine. Based on the analysis, the server generates the following prompt:

[1253] Example prompt sentence:

[1254] In the delighted state, the user said: Give me a review about this bag

[1255] Generate optimal product reviews based on this.

[1256] The generative model generates product reviews based on the prompts and sends them back to the server, which then sends the generated results to the user's device, which then displays the reviews to the user.

[1257] As described above, the present invention is a system that generates prompts based on the user's intentions and emotions, thereby providing more accurate responses.

[1258] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1259] Step 1:

[1260] The user inputs a request to the terminal by voice or text. For example, the user inputs "Please give me a review about this bag." This voice data is received by the terminal.

[1261] Step 2:

[1262] The device converts the voice data into text data using Google's speech recognition API, resulting in the text data "Write a review about this bag."

[1263] Step 3:

[1264] The device sends text data to the server, which then receives the text and analyzes the user's intent. This is done using natural language processing (NLP) techniques, such as the TextBlob library. The analysis identifies the user's intent as a request for a bag review.

[1265] Step 4:

[1266] At the same time, the server analyzes the user's emotions from the text data using the TextBlob library, for example, to identify the emotion "joy."

[1267] Step 5:

[1268] The server automatically generates prompts based on the analyzed user intent and emotion, such as "The user said the following in the joy state: Give us your review of this bag. Based on this, generate the best product review."

[1269] Step 6:

[1270] The server sends the automatically generated prompts to a generative model, which uses OpenAI's API, and the generative model receives the prompts and generates a product review based on them.

[1271] Step 7:

[1272] The product reviews generated by the generative model are sent back to the server. For example, a review might be generated that reads, "This bag is very user-friendly and has an excellent design. It is particularly notable for the high-quality materials used."

[1273] Step 8:

[1274] The server sends the reviews received from the generative model to the device, which receives the reviews and displays them visually or audibly to the user, allowing the user to confirm the response to their request.

[1275] Through the above processing steps, the user can obtain accurate product reviews based on his or her own intentions and feelings.

[1276] 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.

[1277] 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.

[1278] 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.

[1279] [Fourth embodiment]

[1280] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1281] 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.

[1282] 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).

[1283] 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.

[1284] 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.

[1285] 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).

[1286] 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.

[1287] 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.

[1288] 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.

[1289] 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.

[1290] 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.

[1291] 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.

[1292] 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."

[1293] The present invention relates to a system that utilizes a generative model using user input information, and a specific embodiment thereof will be described below. The system consists of three elements: a server, a terminal, and a user.

[1294] System configuration and operation

[1295] User Input

[1296] A user accesses the system using a device (e.g., a smartphone, tablet, or PC). At this stage, the user can select from a variety of input methods, such as voice recognition, text input, and icon manipulation, to input requests to the generative model. For example, a user may input in text format, "I would like to create a summary of the planning meeting," or make a similar request by voice.

[1297] Providing an interface

[1298] The terminal provides an interface for the user to provide input information, which may include a user interface including icons, microphone access permission for voice input, text input fields, etc. When the user inputs information through these interfaces, the information is transmitted to the server in real time.

[1299] Analyzing User Intent

[1300] The server analyzes the user's input information received from the device. Using natural language processing (NLP) technology, it analyzes the user's text and voice data to understand the user's intent. This analysis allows the server to determine what the user is looking for and generate appropriate prompts based on that. As a result of this analysis, the server may generate a prompt such as "Summary of the planning meeting, including key points."

[1301] Automatically Generated Prompts

[1302] Based on the analyzed user intent, the server automatically generates a prompt containing the necessary information. For example, if a user inputs "Please create a summary of a product review," the server generates a prompt "Summary of a product review, including key points." This prompt is sent to the generative model.

[1303] Prompting the Generative Model

[1304] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques (e.g., a Generative Pre-trained Transformer) to generate output based on the user's prompts. This output can include text, images, video data, etc.

[1305] Receiving a response from the generative model

[1306] The results generated by the generative model based on the specified prompts are sent back to the server. For example, if the generative model generates a "summary of the planning meeting" based on the user's request, the summary is sent to the server.

[1307] Displaying the results

[1308] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. This step allows the user to easily check the generated output, such as text, images, and videos. For example, a summary of a planning meeting generated can be displayed on the screen, allowing the user to confirm the response to their request.

[1309] Example: A series of steps for generating sentences

[1310] Consider the case where a user voice-inputs "Please create a summary of a product review" into a device. This voice input is received by the device and sent to the server. The server converts the voice data into text and analyzes the user's intent through natural language processing. Based on the results, it automatically generates a prompt "Summary of a product review, including key points" and sends it to the generative model. The generative model generates a product review summary based on this prompt and sends it back to the server. Finally, the server sends the received summary to the user's device, which displays it to the user.

[1311] As described above, the present invention provides a system for easily and efficiently using generative models, eliminating the need for users to manually input detailed prompts.

[1312] The processing flow will be explained below.

[1313] Step 1:

[1314] A user accesses a terminal.

[1315] Action: A user launches an application on the system using a device such as a smartphone or PC.

[1316] Step 2:

[1317] The user provides the input information.

[1318] Action: User clicks icon or speaks or writes "I want to create a summary of the planning meeting."

[1319] Step 3:

[1320] The terminal receives the input information.

[1321] How it works: The device receives the user's clicks, voice, and text input in real time and transmits that information to the server.

[1322] Step 4:

[1323] The server analyzes the input information.

[1324] How it works: The server analyzes the data received from the device, converts voice data into text, and uses natural language processing (NLP) techniques to understand the user's intent.

[1325] Step 5:

[1326] The server generates the appropriate prompt.

[1327] How it works: Based on the analysis results, the server automatically generates a prompt such as "Summary of the planning meeting, including key points."

[1328] Step 6:

[1329] The server sends a prompt to the generative model.

[1330] How it works: The server sends the generated prompt as an HTTP request to the API endpoint of the generating AI.

[1331] Step 7:

[1332] The server receives the response from the generative model.

[1333] How it works: The generative AI model generates output based on the prompts, and the results are sent back to the server.

[1334] Step 8:

[1335] The server sends the results to the terminal.

[1336] How it works: The server formats the generated output and creates data packets to send to the user's terminal.

[1337] Step 9:

[1338] The terminal displays the results to the user.

[1339] Operation: The device visualizes the received data and presents it to the user. For example, it displays a summary of the planning meeting on the screen.

[1340] Examples:

[1341] 1. The user speaks, "I want a summary of a product review."

[1342] 2. The device sends the audio data to the server.

[1343] 3. The server transcribes the audio data, performs natural language processing, and generates prompts.

[1344] 4. The server sends a prompt to the generative model.

[1345] 5. The generative model generates a summary of the review and sends it back to the server.

[1346] 6. The server sends the results to the user's terminal, which displays a summary to the user.

[1347] In this way, the system automatically performs analysis and generation without the user having to enter complex prompts, making generative models easier to use.

[1348] Example 1

[1349] 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."

[1350] When users use complex generative models, many detailed operations are required, which places a significant burden on them. In particular, when using voice or text input, users must manually input appropriate prompts into the generative model, which is a time-consuming and labor-intensive task. A solution to this problem is needed that allows users to easily use generative models.

[1351] 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.

[1352] In this invention, the server includes a means for converting voice data into text data, a means for analyzing the converted text data using natural language processing technology, and a means for formatting the generated prompt into a format that is easy for the generative model to understand, thereby enabling the user to use the generative model simply and quickly without performing complex operations.

[1353] "User input information" refers to data that a user provides to the system in the form of voice, text, icon manipulation, or the like.

[1354] "Means for receiving" refers to a function for receiving information input by a user and incorporating it into the system.

[1355] "Means for analyzing" refers to methods and techniques for processing received user input information and understanding its content.

[1356] A "prompt" is a short instruction that specifically specifies the desired output from the generative model.

[1357] A "generative model" refers to an artificial intelligence system that generates text, images, video, etc. based on a given prompt.

[1358] "Means of conversion" refers to the methods and techniques for converting audio data into text data.

[1359] "Natural language processing technology" refers to a technological system that enables computers to understand and analyze human language and perform appropriate information processing.

[1360] "Formatting means" refers to a method for adjusting the format of the generated prompt so that the generative model can efficiently understand it and generate a response.

[1361] A "response" refers to the output that a generative model generates based on a prompt.

[1362] "Means for displaying" refers to a function for visually or audibly presenting the response received from the generative model to the user.

[1363] The present invention relates to a system that utilizes a generative model based on user input information. A specific embodiment of this system is described below. The system consists of three elements: a server, a terminal, and a user.

[1364] System configuration and operation

[1365] User Input

[1366] Users access the system using devices such as smartphones, tablets, and PCs. When sending requests to the system, users can choose from a variety of input methods, such as voice recognition, text input, and icon manipulation. For example, a user may input a request in text format, such as "Please create a product review summary." Alternatively, a user may make a similar request by voice.

[1367] Providing an interface

[1368] The terminal provides an interface for the user to provide input information, including a text input field, a user interface including icons, and microphone access permission for voice input. When the user inputs information through these interfaces, the information is transmitted to the server in real time.

[1369] Analyzing User Intent

[1370] The server analyzes the user's input information sent from the device. It uses natural language processing (NLP) techniques to analyze text and voice data and understand the user's intent. For example, it uses an NLP engine (e.g., a common NLP library or API) to identify that the user's request is to "create a summary of a product review."

[1371] Automatically Generated Prompts

[1372] Based on the analysis results, the server automatically generates a prompt containing the necessary information. For example, it generates a prompt such as "A summary of product reviews, including key points." This prompt is formatted in a way that is easy for the generative model to understand.

[1373] Prompting the Generative Model

[1374] The server sends the generated prompts to a generative model (e.g., an artificial intelligence system such as a Generative Pre-trained Transformer) that uses the prompts to generate output for the user's request, such as text, images, or video. The generative model uses advanced AI algorithms to generate the requested data in real time.

[1375] Receiving a response from the generative model

[1376] The output generated by the generative model is sent back to the server. For example, a "product review summary" is generated and sent to the server. This generated result is sent as structured data, such as text.

[1377] Displaying the results

[1378] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. The device displays the generated text, image, or video output to the user on a screen or using a voice assistant. For example, a generated "product review summary" is displayed on the device screen.

[1379] Specific examples

[1380] A specific example is a case where the user vocally inputs into the terminal, "I want you to create a review summary of the product."

[1381] 1. User input:

[1382] The user speaks to the terminal, "I want a review summary of the product."

[1383] 2. Receiving audio data:

[1384] The device provides microphone access permission for voice input and transmits the received voice data to the server, for example, when the user makes a request to "activate the voice assistant."

[1385] 3. Text Conversion and Intent Analysis:

[1386] The server converts the voice data into text and analyzes the user's intent using natural language processing technology. The server converts the voice data into text using the Google Speech-to-Text API, and then analyzes the intent using a common NLP engine.

[1387] 4. Generate prompt:

[1388] The server generates a prompt such as "Product review summary, including key points" and sends this prompt to the generative model.

[1389] 5. Generative model response:

[1390] A generative model (e.g., GPT-3) generates a review summary based on the prompts and sends the summary back to the server.

[1391] 6. View the response:

[1392] The server sends the summary to the device, which then displays it to the user. The device screen displays the review summary: "I found this product easy to use and very durable. I was particularly impressed by its light weight."

[1393] By following the above steps, users can use generative AI models easily and efficiently.

[1394] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1395] Step 1: User Input

[1396] Users access the system using devices such as smartphones, tablets, and PCs. For example, a user inputs a request in text format, such as "Please create a product review summary." The user's input is entered into the device as text data, such as "Please create a product review summary." This input data is sent to the server through the device's interface.

[1397] Step 2: Provide an interface

[1398] The device provides a user interface that includes text input fields, icons, and microphone access permission for voice input. When a user inputs information through these interfaces, the information is transmitted to the server in real time. This transmitted data becomes the input information required for the next analysis step.

[1399] Step 3: Analyze user intent

[1400] The server analyzes the user's input information received from the terminal. Specifically, it analyzes the text data using natural language processing technology (e.g., a standard NLP engine). The server receives the text data "Please create a product review summary" as input and analyzes this data to identify the user's intent. The output of the analysis is the intent to "generate a product review summary."

[1401] Step 4: Automatically generate prompts

[1402] The server automatically generates a prompt containing the necessary information based on the analyzed user intent. For example, it generates a prompt such as "Product review summary, including important points." The prompt generated by the server is text data such as "Product review summary, including important points." This prompt is sent as input to the generative model.

[1403] Step 5: Sending prompts to the generative model

[1404] The server sends the generated prompt to a generative model (e.g., a Generative Pre-trained Transformer), which generates an output based on the prompt. The generative model receives the prompt "Product review summary, including key points" as input and generates the appropriate text output. The generated output (e.g., "Product review summary") is sent back to the server.

[1405] Step 6: Receiving a response from the generative model

[1406] The server receives the output results returned from the generative model. Text data containing the "product review summary" generated by the generative model is returned to the server. This received data becomes the input data for proceeding to the next display step.

[1407] Step 7: View the results

[1408] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, specific text data for displaying a "product review summary" is sent to the device and displayed to the user on the device screen in the form of "I thought this product was easy to use and very durable. I was particularly impressed by its light weight." This displayed text is the final output, and the user can confirm the generated information.

[1409] The above are the specific processing steps of this system. Each step starts with user input, then includes analysis, prompt generation, output of the generative model, and finally display to the user.

[1410] (Application example 1)

[1411] 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."

[1412] In autonomous vehicles, there is a need for a means to efficiently provide passengers with the information they request. In particular, there is a need for a system that can instantly obtain the information passengers need using voice input or screen operation, and provide it on the spot by display or voice. It is also important that this will improve passenger satisfaction and enhance the convenience of autonomous vehicles.

[1413] 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.

[1414] In this invention, the server includes means for receiving user input information, means for analyzing the user's intention from the received input information, means for automatically generating appropriate prompts based on the analyzed intention, means for transmitting the generated prompts to a generative model, means for receiving a response from the generative model, means for displaying the received response to the user, and means for providing information based on passenger questions using a voice recognition and display device installed in the autonomous vehicle. This enables passengers to instantly obtain necessary information through voice input or screen operation, improving the user experience in the autonomous vehicle.

[1415] "User input information" refers to information such as voice, text, and icon operations that the user provides to the system.

[1416] The "receiving means" is a mechanism for receiving user input information at a terminal or server.

[1417] The "means for analyzing intent" is a method for understanding and analyzing the user's intent based on the received input information.

[1418] The "means for automatically generating a prompt" is a function that automatically generates an appropriate instruction sentence based on the analyzed user's intention.

[1419] A "generative model" is an artificial intelligence technique for generating text, images, videos, etc. based on specific prompts.

[1420] A "means for sending a prompt" is a mechanism for passing a generated prompt to a Generative Model.

[1421] The "means for receiving a response" is a mechanism for receiving a response from the Generative Model.

[1422] A "display means" is a device for visually or audibly presenting the received response to the user.

[1423] An "autonomous vehicle" is a vehicle that can perform driving operations automatically.

[1424] A "voice recognition device" is a device for converting passenger voice input into text data.

[1425] A "display device" is a display device for displaying data such as text and images.

[1426] A "passenger" is a person using an autonomous vehicle.

[1427] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described in detail with reference to the accompanying drawings.

[1428] System configuration

[1429] This system consists of three elements: a server, a terminal, and a user, in order to realize information provision services in autonomous vehicles.

[1430] 1. Receiving user input information

[1431] Users input information using terminals (for example, a voice recognition device or a display device) installed in the autonomous vehicle. The voice recognition device is a device that converts passengers' voice input into text data. The display device is a device that displays data such as text and images. Users make requests by voice, such as "Please tell me about nearby tourist spots."

[1432] 2. Providing an interface

[1433] The terminal provides an interface for the user to provide input information, which may include a microphone for voice input, a display screen, etc. The information input by the user is transmitted to the server in real time.

[1434] 3. Analyzing User Intent

[1435] The server analyzes the user's input information received from the device. It uses natural language processing (NLP) technology to analyze the user's voice data and understand the user's intent. Through this analysis, the server identifies the information the user is seeking and generates appropriate prompts based on that information.

[1436] 4. Automatically Generate Prompts

[1437] Based on the analyzed user intent, the server automatically generates a prompt containing the necessary information, for example, "A passenger is asking about nearby tourist attractions. Please provide the most suitable tourist attraction information based on the following."

[1438] 5. Sending prompts to the generative model

[1439] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques such as Generative Pre-trained Transformer (GPT) to generate outputs based on the user's prompts.

[1440] 6. Receiving a response from the generative model

[1441] The results that the generative model generates based on the specified prompts are sent back to the server. For example, if the generative model generates a "list of nearby tourist attractions" based on the user's request, that information is sent to the server.

[1442] 7. Displaying the results

[1443] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, the generated tourist information is displayed on a screen or guided by voice.

[1444] Specific examples

[1445] When a passenger says, "Please tell me about nearby tourist spots in this car," the voice input is received by the device and sent to the server. The server converts the voice data into text and analyzes the user's intention through natural language processing. As a result, it automatically generates a "prompt about nearby tourist spots" and sends it to the generative model. The generative model generates tourist spot information based on this prompt and sends it back to the server. Finally, the server sends the received tourist spot information to the user's device and displays it on the screen.

[1446] Prompt Sentence Examples

[1447] "A passenger is asking about nearby tourist attractions. Please provide the best tourist information based on the following:

[1448] Question: Please tell me about nearby tourist spots

[1449] Required information: name of tourist spot, distance, highlights

[1450] As described above, the present invention is a system that allows users to instantly obtain necessary information through voice input or screen operation, thereby improving the user experience in an autonomous vehicle.

[1451] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1452] Step 1:

[1453] The user enters information by voice.

[1454] Input: User's voice (e.g., "Please tell me about nearby tourist spots")

[1455] Output: Audio data

[1456] Specific operation: A user provides speech input to a speech recognition device installed in an autonomous vehicle, which receives this input as digital voice data.

[1457] Step 2:

[1458] The terminal converts the voice data into text data.

[1459] Input: Digital audio data

[1460] Output: Text data

[1461] What it does: The device's voice recognition software (e.g., Google Speech-to-Text API) analyzes the voice data and converts it into corresponding text data, such as "Please tell me about nearby tourist spots."

[1462] Step 3:

[1463] The server analyzes the text data and understands the user's intent.

[1464] Input: Text data

[1465] Output: Intent information (e.g., "I'm looking for information about tourist spots")

[1466] Specific operation: The server analyzes text data using a natural language processing (NLP) engine (e.g., SpaCy or GPT-3). From the analyzed data, it understands the user's intent and generates intent information according to the purpose.

[1467] Step 4:

[1468] The server automatically generates prompts based on the user's intent.

[1469] Input: Intention information

[1470] Output: The generated prompt (e.g., "Please tell me about nearby tourist attractions")

[1471] What it does: The server uses the parsed intent information to construct an appropriate prompt, which is automatically generated using a preset template.

[1472] Step 5:

[1473] The server sends the generated prompt to the Generative Model.

[1474] Input: Generated prompt

[1475] Output: A request to the generative model with the prompt sent

[1476] Specific behavior: The server sends the generated prompt via an HTTP request to the generative model (e.g., OpenAI API).

[1477] Step 6:

[1478] The generative model generates a response based on the prompt and sends it back to the server.

[1479] Input: prompt

[1480] Output: Generated response (e.g., list of tourist attractions)

[1481] Specific operation: The generative model analyzes the prompt and generates an appropriate response, which is sent back to the server in the form of text data, image data, or other data.

[1482] Step 7:

[1483] The server receives the response from the generative model and sends it to the terminal.

[1484] Input: Generated response

[1485] Output: Response data sent to the terminal

[1486] Specific operation: The server transfers the response data received from the generative model via the network to the terminal.

[1487] Step 8:

[1488] The terminal displays or audibly presents the received response to the user.

[1489] Input: Response data (e.g., a list of tourist attractions)

[1490] Output: Display or audio output to the user

[1491] Specific actions: Display a list of tourist attractions on the device's display device, or use a speech synthesizer to provide the information to the user aloud.

[1492] 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.

[1493] This invention relates to a system that receives user input information, analyzes the user's intentions and emotions, automatically generates appropriate prompts, and sends the prompts to a generative model. This system consists of three elements: a server, a terminal, and a user, and is also combined with an emotion engine.

[1494] System configuration and operation

[1495] User Input

[1496] Users access the system using a device (e.g., smartphone, tablet, PC, etc.). Users can input requests using a variety of methods, including voice recognition, text input, and icon manipulation. For example, a user might input by voice, "I would like to create a summary of the planning meeting."

[1497] Providing an interface

[1498] The device provides an interface for the user to input information, which may include icons, a microphone for voice input, text input fields, etc. As the user provides the input, it is transmitted to the server in real time.

[1499] Analyzing User Intent

[1500] The server analyzes the user's input information received from the device. It uses natural language processing (NLP) techniques to understand the user's request and identify the user's intent. This analysis generates a prompt, such as "Summary of the planning meeting, including key points."

[1501] User Emotion Recognition

[1502] In addition, the server uses an emotion engine to analyze the user's emotions, recognizing the user's emotions from voice and text data and identifying emotional states such as "happiness," "sadness," and "surprise."

[1503] Automatically Generated Prompts

[1504] Based on the analyzed user intent and emotional information, the server automatically generates an appropriate prompt. For example, if a user inputs "Please create a summary of a product review," the server will generate a prompt such as "A summary of a product review, including key points," complementing the user's emotional information.

[1505] Prompting the Generative Model

[1506] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques to generate output based on the user's prompts, which can include text, images, and video data.

[1507] Receiving a response from the generative model

[1508] The results generated by the generative model based on the specified prompts are sent back to the server. For example, if the generative model generates a "summary of the planning meeting" based on the user's request, the summary is sent to the server.

[1509] Displaying the results

[1510] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, the generated summary of a planning meeting can be displayed on the screen, allowing the user to confirm the response to their request.

[1511] Example: A text generation workflow including emotion recognition

[1512] Consider the case where a user voice-inputs "Please create a summary of a product review" into a device. This voice input is received by the device and sent to the server. The server converts the voice data into text and analyzes the user's intent using natural language processing technology. At the same time, it uses an emotion engine to recognize the user's emotions and includes that information in the prompt. For example, it generates a prompt such as "Summary of a product review, including key points, user is satisfied" and sends it to the generative model. The generative model generates a product review summary based on this prompt and sends it back to the server. Finally, the server sends the generated result to the user's device, which displays the summary to the user.

[1513] As described above, the present invention is a system that provides more appropriate responses by eliminating the need for users to manually input detailed prompts, by utilizing generative models simply and efficiently, and by generating prompts that take the user's emotions into consideration.

[1514] The processing flow will be explained below.

[1515] Step 1:

[1516] A user accesses a terminal.

[1517] Action: A user launches an application on the system using a device such as a smartphone or PC.

[1518] Step 2:

[1519] The user provides the input information.

[1520] Action: User clicks icon or speaks or writes "I want to create a summary of the planning meeting."

[1521] Step 3:

[1522] The terminal receives the input information.

[1523] How it works: The device receives the user's clicks, voice, and text input in real time and transmits that information to the server.

[1524] Step 4:

[1525] The server analyzes the input information.

[1526] How it works: The server analyzes the data received from the device, converts voice data into text, and uses natural language processing (NLP) techniques to understand the user's intent.

[1527] Step 5:

[1528] The server recognizes the user's emotions.

[1529] How it works: The server uses an emotion engine to analyze the user's emotional state from the received voice and text data, for example, identifying emotions such as "happiness," "sadness," or "surprise" from the user's tone of voice and the context of the text.

[1530] Step 6:

[1531] The server generates the appropriate prompt.

[1532] How it works: Based on the analyzed user intent and emotion information, the server automatically generates a prompt such as "Summary of the planning meeting, including key points." It may also fine-tune the prompt based on the recognized emotion.

[1533] Step 7:

[1534] The server sends a prompt to the generative model.

[1535] How it works: The server sends the generated prompt to the API endpoint of the AI ​​generator as an HTTP request. The prompt also contains the user's emotional information.

[1536] Step 8:

[1537] The server receives the response from the generative model.

[1538] How it works: The generative model generates output based on the prompts, and the results are sent back to the server, for example, a summary of the analyzed planning meeting.

[1539] Step 9:

[1540] The server sends the results to the terminal.

[1541] How it works: The server formats the generated output and creates data packets to send to the user's terminal.

[1542] Step 10:

[1543] The terminal displays the results to the user.

[1544] Operation: The device visualizes the received data and presents it to the user. For example, it displays a summary of the planning meeting on the screen.

[1545] Examples:

[1546] 1. The user speaks, "I want a summary of a product review."

[1547] 2. The device sends the audio data to the server.

[1548] 3. The server transcribes the audio data, performs natural language processing, and analyzes the user's intent.

[1549] 4. The server uses the emotion engine to recognize the user's emotion as "satisfied."

[1550] 5. The server generates a prompt: "Summary of product review, including key points, user satisfied."

[1551] 6. The server sends a prompt to the generative model.

[1552] 7. The generative model generates a summary of the review and sends it back to the server.

[1553] 8. The server sends the results to the user's device.

[1554] 9. The terminal displays the summary to the user.

[1555] In this way, the burden on the user to input detailed prompts is reduced, and the system automatically analyzes and generates, making it possible to effectively utilize generative models and provide more appropriate responses that take the user's emotions into account.

[1556] Example 2

[1557] 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."

[1558] In conventional systems, users have to manually input detailed prompts, which is time-consuming, and it is difficult to utilize generative models that take into account the user's intentions and emotions. As a result, the responses users receive are often inappropriate.

[1559] 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.

[1560] In this invention, the server includes means for receiving user input information, means for analyzing the user's intention from the received input information, and means for automatically generating appropriate prompts based on the analyzed intention and the user's emotional information. This allows for the automatic generation of prompts that take the user's intention and emotion into consideration, making it possible to produce more appropriate and accurate output from the generative model.

[1561] "User input information" refers to any information provided by the user to the system, including voice, text, icon operations, and the like.

[1562] The "receiving means" refers to a function or device that allows the system to receive input information provided by the user.

[1563] "User intent" refers to the user's goals and desires as interpreted by the system based on input information.

[1564] "Means for analyzing" refers to technology or devices for analyzing received input information and understanding the user's intentions.

[1565] "Emotion information" refers to the user's emotional state analyzed from voice and text data, and includes "joy," "sadness," "surprise," and the like.

[1566] A "prompt" is text information that provides instructions or guidelines for a generative model when generating output.

[1567] "Automatic generation means" refers to technology or devices for automatically creating appropriate prompts based on the user's intentions and emotional information.

[1568] A "generative model" is a model that includes artificial intelligence techniques that generate output, such as text, images, or video, based on input prompts.

[1569] The "sending means" is a function or device for sending the generated prompt to the Generative Model.

[1570] A "response" is the output result generated by a generative model based on a prompt.

[1571] The "means for displaying" is a function or device for visually or audibly presenting the response received from the generative model to the user.

[1572] This invention relates to a system that receives user input information, analyzes the user's intentions and emotions, automatically generates appropriate prompts, and sends the prompts to a generative model. This system consists of three elements: a server, a terminal, and a user, and is also combined with an emotion engine.

[1573] System configuration and operation

[1574] User Input

[1575] Users access the system using a device (e.g., smartphone, tablet, PC, etc.). Users can input requests using a variety of methods, including voice recognition, text input, and icon manipulation. For example, a user might input by voice, "I would like to create a summary of the planning meeting."

[1576] Providing an interface

[1577] The device provides an interface for the user to input information, which may include icons, a microphone for voice input, text input fields, etc. As the user provides the input, it is transmitted to the server in real time.

[1578] Analyzing User Intent

[1579] The server analyzes the user's input information received from the device. It uses natural language processing (NLP) techniques to understand the user's request and identify the user's intent. This analysis generates a prompt, such as "Summary of the planning meeting, including key points."

[1580] User Emotion Recognition

[1581] In addition, the server uses an emotion engine to analyze the user's emotions, recognizing the user's emotions from voice and text data and identifying emotional states such as "happiness," "sadness," and "surprise."

[1582] Automatically Generated Prompts

[1583] Based on the analyzed user intent and emotional information, the server automatically generates an appropriate prompt. For example, if a user inputs "Please create a summary of a product review," the server will generate a prompt such as "A summary of a product review, including key points," complementing the user's emotional information.

[1584] Prompting the Generative Model

[1585] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques to generate output based on the user's prompts, which can include text, images, and video data.

[1586] Receiving a response from the generative model

[1587] The results generated by the generative model based on the specified prompts are sent back to the server. For example, if the generative model generates a "summary of the planning meeting" based on the user's request, the summary is sent to the server.

[1588] Displaying the results

[1589] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, the generated summary of a planning meeting can be displayed on the screen, allowing the user to confirm the response to their request.

[1590] Example: A text generation workflow including emotion recognition

[1591] Consider the case where a user voice-inputs "Please create a summary of a product review" into a device. This voice input is received by the device and sent to the server. The server converts the voice data into text and analyzes the user's intent using natural language processing technology. At the same time, it uses an emotion engine to recognize the user's emotions and includes that information in the prompt. For example, it generates a prompt such as "Summary of a product review, including key points, user is satisfied" and sends it to the generative model. The generative model generates a product review summary based on this prompt and sends it back to the server. Finally, the server sends the generated result to the user's device, which displays the summary to the user.

[1592] As described above, the present invention is a system that provides more appropriate responses by eliminating the need for users to manually input detailed prompts, by utilizing generative models simply and efficiently, and by generating prompts that take the user's emotions into consideration.

[1593] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1594] Step 1:

[1595] Receiving user input information

[1596] Users use devices (smartphones, tablets, PCs, etc.) to input information by voice, text, or icon operation. The input information is transferred from the device to the server in real time.

[1597] Input: User's voice command "Please create a summary of the product review."

[1598] Output: User request in text format: "Please create a summary of product reviews."

[1599] Step 2:

[1600] Providing an interface

[1601] The device provides an interface for the user to input information, which may include a microphone for voice input, text entry fields, icons, etc.

[1602] Input: User operations (voice input, text input, icon operation)

[1603] Output: Input data (voice, text data) from the device

[1604] Step 3:

[1605] Transfer of input information to the server

[1606] The terminal transfers the information entered by the user to the server in real time.

[1607] Input: Voice or text data received by the device

[1608] Output: Audio or text data sent to the server

[1609] Step 4:

[1610] Analyzing User Intent

[1611] The server analyzes the received user input (voice or text) and uses natural language processing (NLP) techniques to understand the user's intent. This analysis identifies the intent and generates a specific prompt.

[1612] Input: Voice or text data sent to the server

[1613] Output: Parsed user intent: "Summary of product review, including key points"

[1614] Step 5:

[1615] User sentiment analysis

[1616] The server uses an emotion engine to analyze the user's emotions from the received voice and text data, identifying emotions such as "happiness," "sadness," and "surprise."

[1617] Input: Audio or text data

[1618] Output: Identified user emotional state "Satisfied"

[1619] Step 6:

[1620] Automatically Generated Prompts

[1621] The server automatically generates an appropriate prompt based on the analyzed user intent and emotion information. For example, a prompt such as "Summary of product review, including key points, user satisfied" may be generated.

[1622] Input: Parsed user intent and emotion information

[1623] Output: Auto-generated prompt "Summary of product reviews, including key points, satisfied users"

[1624] Step 7:

[1625] Prompting the Generative Model

[1626] The server sends the generated prompts to a generative AI model, which generates an output based on the prompts.

[1627] Input: Auto-generated prompt

[1628] Output: The prompt sent to the generative AI model

[1629] Step 8:

[1630] Receiving a response from the generative model

[1631] The generative AI model generates output based on the prompts sent and sends the results back to the server.

[1632] Input: Prompt sent

[1633] Output: Generated text "Product Review Summary"

[1634] Step 9:

[1635] Transfer of generated results from the server to the device

[1636] The server transmits the results received from the generative model to the user's terminal.

[1637] Input: The generated results sent from the generative model

[1638] Output: Generated results sent to the terminal

[1639] Step 10:

[1640] Viewing the generated results

[1641] The terminal presents the generated summary of the product review to the user visually or audibly, for example, by displaying it on a screen so that the user can confirm the response to the request.

[1642] Input: Generated result

[1643] Output: The generated result displayed on the user's terminal.

[1644] (Application example 2)

[1645] 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."

[1646] Conventional systems that accept user input can analyze the user's intentions and provide corresponding generated prompts, but they do not take the user's emotions into account when generating prompts, making it impossible to provide responses appropriate to the user's psychological state.In addition, there was a lack of concrete application examples for improving the user's shopping experience, making it difficult to improve customer satisfaction in physical stores.

[1647] The identification process 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 receiving user input information, means for analyzing the user's intention from the received input information, means for analyzing the user's emotion from the received input information, means for automatically generating an appropriate prompt based on the analyzed intention and emotion, means for sending the generated prompt to the generative model, means for receiving a response from the generative model, and means for displaying the received response to the user. This makes it possible to provide a more accurate and satisfying response based on the user's intention and emotion.

[1648] "User input information" refers to information input by the user through various means such as voice, text, or icon operation.

[1649] "Intent analysis" is the process of determining the content and purpose of received user input.

[1650] "Emotion analysis" is the process of identifying a user's emotional state from speech or text data, for example, detecting emotions such as joy, sadness, or surprise.

[1651] "Automatic prompt generation" refers to the process by which the system automatically generates appropriate instructions or commands based on the analyzed intent and sentiment.

[1652] A "generative model" is a system that uses artificial intelligence technology to generate text, images, and videos based on user requests.

[1653] "Receiving a response" refers to the process by which the server receives the results generated from the Generative Model.

[1654] "Displaying to the user" refers to the process of presenting the generated results to the user through the screen or audio of the device they use.

[1655] This invention relates to a system that receives user input, analyzes the user's intentions and emotions, automatically generates appropriate prompts, and sends the prompts to a generative model. This system mainly consists of three elements: a server, a terminal, and a user, and is also combined with an emotion engine.

[1656] System configuration and operation

[1657] User Input

[1658] Users access the system using a device (e.g., a smartphone or smart glasses). Users can input requests using a variety of methods, including voice recognition, text input, and icon manipulation. For example, a user might say, "Please give me a review of this bag."

[1659] Providing an interface

[1660] The device provides an interface for the user to input information, including a microphone for voice input, text input fields, and a UI for icon manipulation. As the user provides input, it is transmitted to the server in real time.

[1661] Analyzing user intent and emotions

[1662] The server analyzes the user input information received from the device. It uses natural language processing (NLP) technology to understand the user's request and identify the user's intent. This analysis is performed using, for example, the TextBlob library. It also analyzes the user's emotions using an emotion engine. It recognizes the user's emotions from voice and text data and identifies emotional states such as "happy," "unhappy," or "neutral." This process uses Google's speech recognition API and TextBlob library.

[1663] Automatically Generated Prompts

[1664] Based on the analyzed user intent and emotion information, the server automatically generates an appropriate prompt. For example, if a user inputs "Please give me a review about this bag," the server generates a prompt such as "The user said the following in a happy state: Please give me a review about this bag. Please generate the optimal product review based on this," and sends it to the generative model.

[1665] Prompting the Generative Model

[1666] The server sends the generated prompts to a generative model, which uses artificial intelligence techniques to generate output based on the user's prompts. This output can include text, images, and video data. Specifically, OpenAI's API is used.

[1667] Receiving a response from the generative model

[1668] The results generated by the generative model based on the specified prompts are sent back to the server. For example, if the generative model generates a "review of this bag" based on the user's request, the review is sent to the server.

[1669] Displaying the results

[1670] The server sends the results received from the generative model to the device, which then presents them to the user visually or audibly. For example, the generated bag review can be displayed on the screen, allowing the user to confirm the response to their request.

[1671] Specific examples

[1672] Consider a scenario where a user uses smart glasses to voice-input "Please give me a review about this bag." This voice input is received by the device and sent to a server, which converts the voice data into text and analyzes the user's intent and emotions using natural language processing technology and an emotion engine. Based on the analysis, the server generates the following prompt:

[1673] Example prompt sentence:

[1674] In the delighted state, the user said: Give me a review about this bag

[1675] Generate optimal product reviews based on this.

[1676] The generative model generates product reviews based on the prompts and sends them back to the server, which then sends the generated results to the user's device, which then displays the reviews to the user.

[1677] As described above, the present invention is a system that generates prompts based on the user's intentions and emotions, thereby providing more accurate responses.

[1678] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1679] Step 1:

[1680] The user inputs a request to the terminal by voice or text. For example, the user inputs "Please give me a review about this bag." This voice data is received by the terminal.

[1681] Step 2:

[1682] The device converts the voice data into text data using Google's speech recognition API, resulting in the text data "Write a review about this bag."

[1683] Step 3:

[1684] The device sends text data to the server, which then receives the text and analyzes the user's intent. This is done using natural language processing (NLP) techniques, such as the TextBlob library. The analysis identifies the user's intent as a request for a bag review.

[1685] Step 4:

[1686] At the same time, the server analyzes the user's emotions from the text data using the TextBlob library, for example, to identify the emotion "joy."

[1687] Step 5:

[1688] The server automatically generates prompts based on the analyzed user intent and emotion, such as "The user said the following in the joy state: Give us your review of this bag. Based on this, generate the best product review."

[1689] Step 6:

[1690] The server sends the automatically generated prompts to a generative model, which uses OpenAI's API, and the generative model receives the prompts and generates a product review based on them.

[1691] Step 7:

[1692] The product reviews generated by the generative model are sent back to the server. For example, a review might be generated that reads, "This bag is very user-friendly and has an excellent design. It is particularly notable for the high-quality materials used."

[1693] Step 8:

[1694] The server sends the reviews received from the generative model to the device, which receives the reviews and displays them visually or audibly to the user, allowing the user to confirm the response to their request.

[1695] Through the above processing steps, the user can obtain accurate product reviews based on his or her own intentions and feelings.

[1696] 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.

[1697] 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.

[1698] 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.

[1699] 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.

[1700] FIG. 9 illustrates 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 behaviors 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.

[1701] 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.

[1702] 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).

[1703] 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.

[1704] 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."

[1705] 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.

[1706] 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).

[1707] 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.

[1708] 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.

[1709] 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.

[1710] 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.

[1711] 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.

[1712] 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.

[1713] 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.

[1714] 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.

[1715] 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.

[1716] 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.

[1717] The following is further disclosed regarding the above embodiment.

[1718] (Claim 1)

[1719] means for receiving user input information;

[1720] means for analyzing a user's intent from received input information;

[1721] a means for automatically generating appropriate prompts based on the parsed intent;

[1722] a means for sending the generated prompts to the generative model;

[1723] means for receiving a response from the generative model;

[1724] means for displaying the received response to a user;

[1725] A system including:

[1726] (Claim 2)

[1727] 10. The system of claim 1, wherein the input information is speech, text, and icon manipulation.

[1728] (Claim 3)

[1729] 10. The system of claim 1, wherein the responses generated by the generative model include text, image, and video data.

[1730] "Example 1"

[1731] (Claim 1)

[1732] means for receiving user input information;

[1733] means for analyzing a user's intent from received input information;

[1734] a means for automatically generating appropriate prompts based on the parsed intent;

[1735] a means for sending the generated prompts to the generative model;

[1736] means for receiving a response from the generative model;

[1737] means for displaying the received response to a user;

[1738] means for converting voice data into text data;

[1739] A means for analyzing the converted text data using natural language processing technology;

[1740] A means of formatting the generated prompts into a form that is understandable to the generative model; and

[1741] A system including:

[1742] (Claim 2)

[1743] 10. The system of claim 1, wherein the input information is speech, text, and icon manipulation.

[1744] (Claim 3)

[1745] 10. The system of claim 1, wherein the responses generated by the generative model include text, image, and video data.

[1746] "Application Example 1"

[1747] (Claim 1)

[1748] means for receiving user input information;

[1749] means for analyzing a user's intent from received input information;

[1750] a means for automatically generating appropriate prompts based on the parsed intent;

[1751] a means for sending the generated prompts to the generative model;

[1752] means for receiving a response from the generative model;

[1753] means for displaying the received response to a user;

[1754] a means for utilizing a voice recognition and display device on board the automated driving vehicle to provide information based on a passenger's question;

[1755] A system including:

[1756] (Claim 2)

[1757] 10. The system of claim 1, wherein the input information is speech, text, and icon manipulation.

[1758] (Claim 3)

[1759] 10. The system of claim 1, wherein the responses generated by the generative model include text, still images, and video data.

[1760] "Example 2: Combining Emotion Engines"

[1761] (Claim 1)

[1762] means for receiving user input information;

[1763] means for analyzing a user's intent from received input information;

[1764] a means for automatically generating appropriate prompts based on the analyzed intent and the user's emotional information;

[1765] a means for sending the generated prompts to the generative model;

[1766] means for receiving a response from the generative model;

[1767] means for displaying the received response to a user;

[1768] A system including:

[1769] (Claim 2)

[1770] 10. The system of claim 1, wherein the input information is speech, text, and icon manipulation.

[1771] (Claim 3)

[1772] 10. The system of claim 1, further comprising: an emotion engine for analyzing the user's emotions.

[1773] (Claim 4)

[1774] 10. The system of claim 1, wherein the responses generated by the generative model include text, image, and video data.

[1775] "Application example 2 when combining emotion engines"

[1776] (Claim 1)

[1777] means for receiving user input information;

[1778] means for analyzing a user's intent from received input information;

[1779] means for analyzing user emotions from received input information;

[1780] means for automatically generating appropriate prompts based on the analyzed intent and sentiment;

[1781] a means for sending the generated prompts to the generative model;

[1782] means for receiving a response from the generative model;

[1783] means for displaying the received response to a user;

[1784] A system including:

[1785] (Claim 2)

[1786] 10. The system of claim 1, wherein the input information is speech, text, and icon manipulation.

[1787] (Claim 3)

[1788] 10. The system of claim 1, wherein the responses generated by the generative model include text, image, and video data. [Explanation of symbols]

[1789] 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 for receiving user input information; means for analyzing a user's intent from received input information; a means for automatically generating appropriate prompts based on the parsed intent; a means for sending the generated prompts to the generative model; means for receiving a response from the generative model; means for displaying the received response to a user; A system including:

2. 10. The system of claim 1, wherein the input information is speech, text, and icon manipulation.

3. The system of claim 1 , wherein the responses generated by the generative model include text, image, and video data.

Images