system

Abstract

Provide a system. A method for capturing a user's voice; A means for converting the captured audio into text data; means for analyzing a user's intent based on the converted text data; A means for generating appropriate text based on the user's intent; means for converting the generated text into audio data and presenting the audio data to a user; means for modifying and regenerating the text according to user confirmation and modification instructions; means for saving the generated and modified text in a final document format; A system including:

Description

[Technical Field]

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

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

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

[0004] As a result of long hours of PC work and online meetings, desk workers are prone to accumulating eye strain, which, if left untreated, can lead to VDT syndrome, which causes mental stress and headaches, or even serious eye diseases such as cataracts and glaucoma.The purpose of this invention is to provide a system that allows desk workers to perform some of their work without using their eyes, thereby reducing eye strain and promoting healthy working styles. [Means for solving the problem]

[0005] The present invention provides a system including means for capturing a user's voice, means for converting the captured voice into text data, means for analyzing the user's intent based on the converted text data, means for generating appropriate text based on the user's intent, means for converting the generated text into voice data and presenting it to the user, means for correcting and regenerating the text in accordance with the user's confirmation and correction instructions, and means for saving the generated and corrected text in a final document format. This system allows desk workers to work without using their eyes and reduces eye strain.

[0006] "User" refers to the person who operates the system and issues instructions via voice input.

[0007] "Audio capturing means" refers to a device or method for collecting a user's voice using a microphone or the like.

[0008] "Means for converting speech to text data" refers to a device or method for converting captured speech data into text format data using speech recognition technology.

[0009] "Means for analyzing user intent based on text data" refers to a device or method for understanding user instructions and intent from text data using natural language processing technology.

[0010] The "means for generating appropriate text" refers to a device or method for generating appropriate sentences or information according to the user's intentions.

[0011] The term "means for converting text into audio data and presenting it to a user" refers to a device or method for converting generated text into audio data using speech synthesis technology and playing it back to a user.

[0012] "Means for correcting and regenerating text according to user confirmation and correction instructions" refers to a device or method that corrects and regenerates text based on confirmation and correction instructions given by the user via voice.

[0013] "Means for saving in final document format" refers to a device or method for saving the generated and modified text in a particular document format (e.g., Word, PDF). [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] This invention relates to a system that allows desk workers to create, check, and correct documents by voice without using their eyes. This system generates text data based on the user's voice input and allows checking and correction through voice feedback. Each component and its operation are explained below.

[0036] System Configuration

[0037] This system consists of three main components: users, terminals, and servers.

[0038] Audio capture and preprocessing

[0039] The user gives voice instructions using a dedicated microphone or a microphone built into the device, and the device collects the user's voice in real time using a voice capture module, allowing the user to start working without needing an input device.

[0040] Once the voice is captured, the device converts the voice data into text using a speech recognition engine that is highly accurate and takes into account a wide range of pronunciations and accents.

[0041] Intention Analysis and Text Generation

[0042] The converted text data is sent to the server and analyzed by a natural language processing engine. The server analyzes this text data to recognize the user's intent. For example, if a user is given an instruction such as "Please write about sales analysis," the server analyzes the intent and creates an appropriate text generation request.

[0043] Based on the analyzed intent, the server uses chat generation AI to generate text according to the instructions. The generated text reflects industry-standard formatting and structure, and is tailored to the user's needs.

[0044] Audio feedback and confirmation / correction

[0045] The generated text is sent to the device, which converts it into voice data using a speech synthesis engine. The device then plays the converted voice data back to the user, prompting them to confirm the content. The user can then use the voice feedback to confirm the content of the document and, if necessary, provide voice instructions for corrections.

[0046] For example, if a user says, "Please change the expression 'year-on-year comparison' to 'the percentage increase from the same month last year,'" the device captures this speech, converts it back into text data, and sends it to the server. The server again analyzes the intent, corrects it as necessary, and regenerates the text. This process is repeated until the user is satisfied.

[0047] Save the final document

[0048] When the user makes a final confirmation and gives the instruction to "save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., Word, PDF) and saves it in a database or cloud storage. After the saving process is complete, the device will inform the user by voice that the save is complete.

[0049] Specific examples

[0050] Example 1:

[0051] Launch and Start: When the user says "Start creating a report," the device captures the voice, converts it into text using a speech recognition engine, and sends it to the server.

[0052] Intent analysis and text generation: The server analyzes the intent of "Start creating a report," generates the text "Start creating a report. What would you like as your first item?", and sends it to the device.

[0053] Confirm and correct: If the user instructs "Write about sales analysis," the same process is followed to generate the text "This month's sales increased by 5% compared to the same month last year." The user then requests corrections, and the final content is "This month's sales increased by 5% compared to the same month last year."

[0054] Example 2:

[0055] Save: When the user says "Save this content," the server saves it as the final document and the device notifies the user by voice that the save is complete.

[0056] In this way, the system of the present invention enables desk workers to create and edit documents without using their eyes, reducing eye strain and supporting healthy working styles.

[0057] The processing flow will be explained below.

[0058] Step 1:

[0059] A user logs into a dedicated application or device and dictates "Start report creation."

[0060] Step 2:

[0061] The device captures the user's voice through the microphone and stores the voice data in a buffer.

[0062] Step 3:

[0063] The voice data captured by the device is sent to a voice recognition library and converted into text data.

[0064] Step 4:

[0065] The voice recognition library converts the voice data into text data and sends it back to the device.

[0066] Step 5:

[0067] The terminal encodes the converted text data and sends it to the server as an HTTP request.

[0068] Step 6:

[0069] The server decodes the encoded text data and analyzes it using a natural language processing engine.

[0070] Step 7:

[0071] The server recognizes the user's intention based on the analysis results and sends a request to the chat generation AI to generate text based on that intention.

[0072] Step 8:

[0073] The chat generation AI generates appropriate text based on the instructions and sends it back to the server.

[0074] Step 9:

[0075] The server receives the generated text data and sends it to the terminal as an HTTP response.

[0076] Step 10:

[0077] The text data received by the terminal is sent to a speech synthesis engine and converted into voice data.

[0078] Step 11:

[0079] The speech synthesis engine converts the text into speech data and sends it back to the device.

[0080] Step 12:

[0081] The terminal plays back the generated voice data to the user, prompting them to confirm the content.

[0082] Step 13:

[0083] The user confirms the result through voice feedback and, if necessary, gives voice instructions for correction.

[0084] Step 14:

[0085] The device captures the correction instructions and converts them into text data using a voice recognition library again.

[0086] Step 15:

[0087] The terminal transmits the corrected text data to the server.

[0088] Step 16:

[0089] The server parses the modified text data and regenerates the text as necessary.

[0090] Step 17:

[0091] The server transmits the regenerated text data to the terminal, where it is converted back into voice data and played back to the user.

[0092] Step 18:

[0093] This process is repeated, with revisions and validations being made until the user is satisfied.

[0094] Step 19:

[0095] The user issues a voice command saying, "Please save this content."

[0096] Step 20:

[0097] The terminal sends a final instruction to the server.

[0098] Step 21:

[0099] The server converts the verified text into a final document format (e.g., Word, PDF) and stores it in a database or cloud storage.

[0100] Step 22:

[0101] The server sends a notification of completion of saving to the terminal.

[0102] Step 23:

[0103] The device will notify the user by voice when the save is complete.

[0104] This allows users to create, edit, and save documents without using their eyes, reducing physical strain.

[0105] Example 1

[0106] 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."

[0107] Modern desk workers are suffering from health problems such as eye strain and vision loss due to long hours of computer work. Creating, reviewing, and correcting documents in particular places a heavy visual burden on the eyes, reducing efficiency and causing health problems. Furthermore, traditional document creation methods are difficult for visually impaired users. A system that can solve these problems and achieve a more efficient and healthy work environment is needed.

[0108] 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.

[0109] In this invention, the server includes a means for capturing a user's voice, a means for converting the captured voice into text data, and a means for analyzing the user's intent based on the converted text data. This allows the user to create, review, and modify documents using only voice instructions. Furthermore, by providing a means for creating documents using a generative AI model and a means for saving the final formatted document in data storage, efficient and accurate document creation and storage can be achieved. This reduces visual strain and supports healthy working styles.

[0110] A "means for capturing audio" is a device or system for capturing audio uttered by a user and processing it as a digital signal.

[0111] The "means for converting into text data" is a technique for analyzing the captured audio data and converting it into a corresponding text format.

[0112] "Means for analyzing user intent" refers to natural language processing technology used to understand the converted text data and identify what the user is looking for.

[0113] "Means for generating documents" refers to technology that uses a generative AI model to create text according to the specified content based on the user's intentions.

[0114] The "means for converting the text into audio data and presenting it to the user" refers to a technique for converting the generated or corrected text into audio and providing it to the user as an anti-visual interface.

[0115] The "means for correcting and regenerating a document according to confirmation and correction instructions" is a processing system for correcting the contents of a document in response to feedback from a user and regenerating the document as necessary.

[0116] "Means for saving in final format" refers to the technology for converting the completed document into a specified file format and saving it in a database, cloud storage, etc.

[0117] A "data analysis module" is a program or function that analyzes data based on the user's voice instructions and sends appropriate instructions to the generative AI model.

[0118] A "generative AI model" is an artificial intelligence technology that generates text based on a given prompt.

[0119] "Data storage" refers to a system or device for storing and managing data over the long term.

[0120] This invention relates to a system that allows desk workers to create, check, and correct documents by voice without using their eyes. This system generates text data based on the user's voice input and allows checking and correction through voice feedback. Each component and its operation are explained below.

[0121] System Configuration

[0122] This system consists of three main components: users, terminals, and servers.

[0123] Audio capture and preprocessing

[0124] The user gives voice instructions using a dedicated microphone or a microphone built into the device, and the device collects the user's voice in real time using a voice capture module, allowing the user to start working without needing an input device.

[0125] Voice Recognition

[0126] Once the audio is captured, the device converts the audio data into text using a speech recognition engine (e.g., Google® Cloud Speech-to-Text) that has high accuracy and takes into account various pronunciations and accents.

[0127] Intention Analysis and Text Generation

[0128] The converted text data is sent to the server and analyzed by a natural language processing engine (e.g., spaCy). The server analyzes this text data to recognize the user's intent. For example, if an instruction such as "Please write about sales analysis" is given, the server analyzes the intent and creates an appropriate text generation request.

[0129] Based on the analyzed intent, the server uses a generative AI model (e.g., OpenAI® GPT-3®) to generate text based on the instructions. The generated text reflects industry-standard formatting and structure and is tailored to the user's intended content.

[0130] Audio feedback and confirmation / correction

[0131] The generated text is sent to the device, which converts it into voice data using a speech synthesis engine (e.g., Amazon Polly). The device then plays the converted voice data back to the user, prompting them to confirm the content. The user can then confirm the content of the document through voice feedback and, if necessary, provide voice instructions for correction.

[0132] For example, if a user says, "Please change the expression 'compared to the same month last year' to 'the percentage increase from the same month last year,'" the device captures this speech, converts it into text data using a speech recognition engine, and sends it to the server. The server again analyzes the intent, corrects the text as necessary, and regenerates it. This process is repeated until the user is satisfied.

[0133] Save the final document

[0134] When the user makes a final confirmation and gives the instruction "Save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., Word, PDF) and saves it in a database or cloud storage (e.g., Google Drive, Dropbox). After the saving process is complete, the device will inform the user by voice that the save is complete.

[0135] Specific examples

[0136] Below is a concrete example of how this system can be used.

[0137] Example 1:

[0138] Boot and start:

[0139] When a user says "start report creation," the device captures the voice, converts it into text using a speech recognition engine, and sends it to the server.

[0140] The server analyzes the intent of "Start creating a report," generates the text "Start creating a report. What would you like as your first item?", and sends it to the terminal.

[0141] The user gives the voice instruction "Write about your sales analysis."

[0142] The generative AI model generates the text, "Sales this month increased by 5% compared to the same month last year."

[0143] The user further instructs, "Please change the year-on-year comparison to the increase rate from the same month last year."

[0144] The device captures the audio again and sends correction instructions to the server.

[0145] The server generates the text "Sales this month increased 5% from the same month last year" reflecting the correction.

[0146] Example 2:

[0147] keep:

[0148] When the user says "Save this content," the device captures the audio, converts it into text, and sends it to the server.

[0149] The server saves it as the final document, and the device notifies you with an audio message that the save is complete.

[0150] As described above, the system of the present invention allows desk workers to create, check, and edit documents using only voice commands, thereby reducing visual strain and supporting healthy working styles.

[0151] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0152] Step 1: Capture audio

[0153] Specific description:

[0154] Input: The user gives a voice command such as "Start report creation."

[0155] Specific operation: The user's voice is captured using a dedicated microphone or the device's built-in microphone.

[0156] Data processing: The terminal collects audio data using an audio capture module.

[0157] Output: The captured voice data is passed to the speech recognition engine.

[0158] Step 2: Voice Recognition

[0159] Specific description:

[0160] Input: Captured audio data.

[0161] Specific operation: The terminal converts voice data into text data using a voice recognition engine (e.g., Google Cloud Speech-to-Text).

[0162] Data processing: The speech recognition engine analyzes the speech waveform and converts it into a corresponding string of characters.

[0163] Output: The converted text data is generated.

[0164] Step 3: Intent Analysis

[0165] Specific description:

[0166] Input: Text data output from the speech recognition engine.

[0167] Specific operation: The server receives the text data sent from the terminal and analyzes it using a natural language processing engine (e.g., spaCy).

[0168] Data processing: Natural language processing engines analyze text data to identify user intent.

[0169] Output: Analysis results that reflect the user's intentions are generated.

[0170] Step 4: Text Generation

[0171] Specific description:

[0172] Input: Analysis results obtained from intent analysis.

[0173] Specific operation: The server uses a generative AI model (e.g., OpenAI GPT-3) to generate text based on the user's intent.

[0174] Data processing: The generative AI model generates relevant text based on the input prompt.

[0175] Output: Generated text data is generated.

[0176] Step 5: Audio feedback

[0177] Specific description:

[0178] Input: Generated text data.

[0179] Specific operation: The device converts the generated text data into audio data using a speech synthesis engine (e.g., Amazon Polly) and plays it back.

[0180] Data processing: The speech synthesis engine converts text data into speech waveforms.

[0181] Output: Audio data is generated that is played back to the user.

[0182] Step 6: Accepting correction instructions

[0183] Specific description:

[0184] Input: User's spoken correction instructions (e.g., "Please change the expression 'year-over-year comparison' to 'percent increase from the same month last year'").

[0185] Specific operation: The device again captures the user's voice correction instructions and converts them into text data using a voice recognition engine.

[0186] Data processing: The new voice command is converted into text and sent to the server.

[0187] Output: Text data containing correction instructions is passed to the server.

[0188] Step 7: Fix and Regenerate

[0189] Specific description:

[0190] Input: Text data containing correction instructions.

[0191] Specific behavior: The server parses the correction instructions, corrects the text as necessary, and regenerates it.

[0192] Data processing: The data analysis module analyzes the correction instructions and sends new prompts to the generative AI model to regenerate the text.

[0193] Output: The corrected text data is generated.

[0194] Step 8: Save the final document

[0195] Specific description:

[0196] Input: Final text data after correction (e.g., "Sales this month increased 5% from the same month last year.").

[0197] Specific operation: The server converts the final text data into a document format (e.g., Word, PDF) and saves it in data storage (e.g., Google Drive, Dropbox).

[0198] Data processing: The text data is converted into the specified file format and uploaded to cloud storage.

[0199] Output: A sound is generated to notify the user that the save is complete.

[0200] (Application example 1)

[0201] 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."

[0202] The objective of this invention is to provide a system that reduces the burden of deskwork in a factory environment and allows users to create, check, modify, and finally save reports using only voice commands. This system allows users to operate it by voice without using their eyes, improving worker health and increasing productivity. Furthermore, in a voice-only work environment, rapid and accurate data processing is required.

[0203] 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.

[0204] In this invention, the server includes means for capturing a user's voice, means for converting the captured voice into text data, means for analyzing the user's intent based on the converted text data, means for generating appropriate text based on the user's intent, means for converting the generated text into voice data and presenting it to the user, means for correcting and regenerating the text according to the user's confirmation and correction instructions, means for saving the generated and corrected text in a final document format, and robot means for realizing use in a voice-specific work environment, a factory environment where reports are created by voice instructions. This allows workers to create documents using only voice instructions, reducing eye fatigue and improving work efficiency.

[0205] "User" refers to the person who operates the system and gives voice instructions.

[0206] "Audio capturing means" refers to a device or method that captures the audio produced by a user as electronic data.

[0207] "Means for converting voice into text data" refers to technology that analyzes captured voice data and converts it into corresponding text data.

[0208] "Means of intent analysis" refers to technology that understands what a user wants from text data and identifies appropriate actions.

[0209] "Means for generating appropriate text" refers to technology that automatically creates the necessary information and sentences based on the user's intentions.

[0210] The "means for converting the generated text into audio data and presenting it to the user" refers to a device or method for converting the generated text back into audio data and allowing the user to listen to it.

[0211] "Means for correcting and regenerating text" refers to a technique for revising existing text and generating new text according to user instructions.

[0212] "Means for saving in final document format" refers to a technique for saving the final verified text in a specified format.

[0213] A "voice-focused work environment" refers to a workplace where operations and instructions are primarily performed by voice.

[0214] "Use in a factory environment where reports are created using voice instructions" refers to performing a series of tasks such as creating, modifying, and saving reports using only voice within the factory.

[0215] "Robotic means" refers to automated equipment or systems that implement voice commands in a factory environment.

[0216] This invention is a system for efficiently creating reports using voice instructions in a factory environment. The system generates text data based on the user's voice input and allows confirmation and correction through voice feedback. The specific configuration and processing procedure for realizing this system will be described below.

[0217] System Configuration

[0218] The system consists of four main components: users, terminals, servers, and robots.

[0219] 1. Audio capture and preprocessing

[0220] The user gives voice instructions using a dedicated microphone or the device's built-in microphone. The device collects the user's voice in real time using a voice capture module. The collected voice data is converted into text data using a voice recognition engine (e.g., Google Speech-to-Text API).

[0221] 2. Intention Analysis and Text Generation

[0222] The converted text data is sent to a server and analyzed by a natural language processing engine (e.g., OpenAI GPT-4 (registered trademark)). The server analyzes this text data to recognize the user's intent. For example, if an instruction such as "Please write about sales analysis" is given, the server analyzes the intent and creates an appropriate text generation request. Based on the analyzed intent, the server uses a generative AI model to generate text according to the instruction.

[0223] 3. Audio feedback and confirmation / correction

[0224] The generated text is sent to the device, which converts it into audio data using a speech synthesis engine (e.g., Google Text-to-Speech API). The device then plays the converted audio data back to the user, prompting them to confirm the content. The user can then use the audio feedback to confirm the content of the document and, if necessary, provide voice instructions for corrections.

[0225] 4. Preservation of the final document

[0226] When the user makes a final confirmation and gives the instruction to "save this content," the device sends this to the server. The server converts the confirmed text into a final document format (e.g., PDF) and saves it in a database or cloud storage (e.g., AWS (registered trademark) S3). After this saving process is complete, the device will notify the user by voice that the save is complete.

[0227] Specific examples

[0228] Example 1: Starting a report

[0229] User: Says "Start creating a report."

[0230] The device collects the voice using a voice capture module and converts it into text data such as "Start creating report" using a voice recognition engine.

[0231] The server receives this text, generates the text "Let's start creating a report. What would you like as the first item?" and sends it to the terminal.

[0232] The terminal uses a speech synthesis engine to convert this text into audio data and play it back to the user.

[0233] Example 2: Sales analysis instructions and corrections

[0234] User: "Write about your sales analysis."

[0235] The terminal collects voice using a voice capture module and converts it into text data using a voice recognition engine.

[0236] The server analyzes this text data, recognizes the intent of "Please write about sales analysis," and generates the text "This month's sales increased by 5% compared to the same month last year."

[0237] The terminal converts the generated text into voice data using a voice synthesis engine and plays it back to the user.

[0238] When the user instructs, "Please change it to the increase rate compared to the same month last year," the device captures this voice, converts it back into text data, and sends it to the server.

[0239] The server again parses the intent and generates the corrected text, "Sales this month increased 5% from the same month last year."

[0240] The terminal converts the corrected text into voice data using a voice synthesis engine and plays it back to the user.

[0241] Prompt Sentence Examples

[0242] "Please write a report on your sales analysis. Please include specific figures and year-on-year increases or decreases."

[0243] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0244] Step 1: User initiates voice command

[0245] The user gives voice instructions (such as "Start creating a report") using a dedicated microphone or the device's built-in microphone. The device collects voice data in real time using a voice capture module. At this time, the user's voice is input, and the captured voice data is output.

[0246] Step 2: Convert the audio data to text

[0247] The device converts the collected voice data into text data using a voice recognition engine (e.g., Google Speech-to-Text API). Specifically, the captured voice data is input and passed through an advanced voice analysis algorithm to generate corresponding text data. This text data is then output.

[0248] Step 3: Analyze the text data

[0249] The server receives the converted text data and analyzes the user's intent through a natural language processing engine (e.g., OpenAI GPT-4). For example, it interprets the purpose of an instruction such as "start creating a report." In this case, the input is text data, and the output is the analysis result (data indicating the user's intent).

[0250] Step 4: Generate the appropriate text

[0251] The server uses a generative AI model based on the results of intent analysis to generate the required text. Based on the converted text data, a response text such as "I'm starting to write a report. What would you like as the first item?" is generated. The input is the analysis result, and the output is the generated text data.

[0252] Step 5: Convert the generated text to audio data

[0253] The device receives the generated text data and converts it into voice data using a speech synthesis engine (e.g., Google Text-to-Speech API). Specifically, it takes the generated text data as input and outputs it as voice waveform data. This voice data is then played back to the user.

[0254] Step 6: Receive user confirmation and correction instructions

[0255] The user can check the generated text through voice feedback and make corrections as necessary. For example, they can say, "Please change it to the increase rate compared to the same month last year." This voice instruction is input, captured by the device, and converted back into text data through the voice recognition engine (output).

[0256] Step 7: Regenerate the text based on the correction instructions

[0257] The server receives the text data again, analyzes the intent, and makes the necessary corrections. For example, based on the instruction "Please change this to the increase rate compared to the same month last year," the corrected text "This month's sales increased by 5% compared to the same month last year" is generated. The input is the text data of the correction instruction, and the output is the corrected text data.

[0258] Step 8: Convert the corrected text back into audio data and present it

[0259] The corrected text data is again converted into voice data by the terminal using a speech synthesis engine. It is then played back to the user for reconfirmation. This process is repeated until the user is satisfied. The input is the corrected text data, and the output is the corrected voice data.

[0260] Step 9: Save in the final document format

[0261] Finally, when the user instructs "Save this content," the server converts the final confirmed text data into a specified document format (e.g., PDF) and saves it in cloud storage (e.g., AWS S3). The input is the final confirmed text data, and the output is the saved document. The device notifies the user by voice when saving is complete.

[0262] 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.

[0263] This invention relates to a system that allows users to generate text through voice input and confirm and correct it by voice. A feature of this invention is that by incorporating an emotion engine that recognizes the user's emotions, the generated text and voice feedback are controlled to match the user's emotions. This improves the user experience and supports healthier working styles.

[0264] System Configuration

[0265] This system consists of four main components: the user, the terminal, the server, and the emotion engine.

[0266] Audio capture and preprocessing

[0267] The user gives voice instructions using a dedicated microphone or a microphone built into the device. The device collects the user's voice in real time using a voice capture module. This voice data is sent to a voice recognition engine and converted into text data.

[0268] emotion recognition

[0269] The converted text data and voice data are sent by the terminal to the emotion engine, which analyzes the user's emotions from the voice and text and generates emotion data. This emotion data is used for subsequent text generation and voice synthesis.

[0270] Intention Analysis and Text Generation

[0271] The server receives the text data and emotion data and analyzes the text data using a natural language processing engine. The server recognizes the user's intention based on the analysis results, and sends a request to the chat generation AI to generate text based on that intention, taking into account the emotion data. The generated text is expected to have a style and tone that matches the user's emotion.

[0272] Audio feedback and confirmation / correction

[0273] The generated text is sent to the device and converted into voice data by a speech synthesis engine. At this time, the tone and speed of the voice are adjusted based on the emotional data obtained from the emotion engine. The device then plays the adjusted voice data back to the user, prompting them to confirm the content. The user can check the content of the document through voice feedback and, if necessary, give voice instructions for corrections.

[0274] For example, if a user instructs the server to "Please prepare a sales report for the previous month," the server analyzes the instruction and generates text such as "Last month's sales increased by 10% compared to the plan." If the emotion engine recognizes that the user is nervous, the speech synthesis engine generates speech in a slightly slower, calmer tone.

[0275] Save the final document

[0276] When the user makes a final confirmation and gives the instruction to "save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., Word, PDF) and saves it in a database or cloud storage. After the saving process is complete, the device will inform the user by voice that the save is complete.

[0277] Specific examples

[0278] Example 1:

[0279] Launch and Start: The user says "Start report creation." The device captures the voice and converts it into text data using the speech recognition engine. At the same time, the emotion engine analyzes the user's current emotion and generates emotion data.

[0280] Intention analysis and text generation: The server analyzes the intention of "Start creating a report" and generates the text "Start creating a report. What would you like to do first?" taking into account the user's relaxation state.

[0281] Confirmation and correction: If the user instructs "Please write about sales analysis," the text generated reflects that state of mind: "This month's sales increased by 5% compared to the same month last year." If the user requests a correction, such as "Please change the expression "compared to the same month last year" to "the percentage increase from the same month last year," the text is corrected based on that request.

[0282] Example 2:

[0283] Save: When the user says "Save this content," the server saves it as the final document and the device notifies the user by voice that the save is complete.

[0284] In this way, the system of the present invention allows document creation and editing to be performed while taking the user's emotions into consideration, improving the user experience while reducing eye strain.

[0285] The processing flow will be explained below.

[0286] Step 1:

[0287] A user logs into a dedicated application or device and dictates "Start report creation."

[0288] Step 2:

[0289] The device captures the user's voice through the microphone and stores the voice data in a buffer.

[0290] Step 3:

[0291] The voice data captured by the device is sent to a voice recognition library and converted into text data.

[0292] Step 4:

[0293] The voice recognition library converts the voice data into text data and sends it back to the device.

[0294] Step 5:

[0295] The terminal transmits the converted text data to an emotion engine, which analyzes the user's emotions.

[0296] Step 6:

[0297] The emotion engine analyzes the voice and text data, generates the user's emotion data, and sends it back to the terminal.

[0298] Step 7:

[0299] The device encodes the text data and emotion data and sends them to the server as an HTTP request.

[0300] Step 8:

[0301] The server decodes the encoded text data and emotion data and analyzes the text data using a natural language processing engine.

[0302] Step 9:

[0303] The server recognizes the user's intention based on the analysis results, takes emotional data into consideration, and sends a request for text generation to the chat generation AI.

[0304] Step 10:

[0305] The chat generation AI generates appropriate text based on the intent and emotion data and sends it back to the server.

[0306] Step 11:

[0307] The server receives the generated text data and sends it to the terminal as an HTTP response.

[0308] Step 12:

[0309] The text data received by the terminal is sent to a speech synthesis engine and converted into voice data based on emotional data.

[0310] Step 13:

[0311] The speech synthesis engine takes into account the emotional data, converts the text into speech data, and sends it back to the device.

[0312] Step 14:

[0313] The terminal plays back the generated voice data to the user, prompting them to confirm the content.

[0314] Step 15:

[0315] The user confirms the result through voice feedback and, if necessary, gives voice instructions for correction.

[0316] Step 16:

[0317] The device captures the correction instructions and converts them into text data using a voice recognition library again.

[0318] Step 17:

[0319] The terminal transmits the corrected text data and emotion data to the server.

[0320] Step 18:

[0321] The server parses the modified text data and regenerates the text as necessary.

[0322] Step 19:

[0323] The server transmits the regenerated text data to the terminal, where it is converted back into voice data and played back to the user.

[0324] Step 20:

[0325] This process is repeated, with revisions and validations being made until the user is satisfied.

[0326] Step 20:

[0327] The user issues a voice command saying, "Please save this content."

[0328] Step 22:

[0329] The terminal sends a final instruction to the server.

[0330] Step 23:

[0331] The server converts the verified text into a final document format (e.g., Word, PDF) and stores it in a database or cloud storage.

[0332] Step 24:

[0333] The server sends a notification of completion of saving to the terminal.

[0334] Step 25:

[0335] The device will notify the user by voice when the save is complete.

[0336] This allows users to create, edit and save documents without using their eyes, and when combined with an emotion engine, provides more relevant and emotionally tailored feedback.

[0337] Example 2

[0338] 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."

[0339] Conventional document creation systems fail to consider the user's emotions when generating text using voice instructions and then reviewing and correcting it. This prevents the system from adjusting text and voice feedback based on the user's emotional state, resulting in a lack of user experience. This also makes it difficult to adjust the tempo and mood while creating documents, hindering efficient and healthy work practices. To address these issues, a system is needed that can recognize the user's emotions in real time and generate text and provide voice feedback that adapts to those emotions.

[0340] 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.

[0341] In this invention, the server includes means for capturing a user's voice, means for converting the captured voice into text data, means for analyzing the user's emotions based on the converted text data and voice data and generating emotion data, means for analyzing the user's intention based on the emotion data and text data, means for generating appropriate text based on the user's intention and emotion data, means for converting the generated text into voice data, adjusting the tone and speed of the voice based on the emotion data and presenting it to the user, means for correcting and regenerating the text in accordance with the user's confirmation and correction instructions, and means for saving the generated and corrected text in a final document format. This enables text generation and voice feedback adapted to the user's emotions, improving the user experience and realizing an efficient and healthy way of working.

[0342] An "audio capture means" is a device or software for collecting a user's voice in real time.

[0343] "Speech recognition means" refers to a device or software for converting collected voice data into text data.

[0344] The "emotion recognition means" is a device or software that analyzes the user's emotions based on the converted text data and voice data, and generates emotion data.

[0345] The "intention analysis means" is a device or software that analyzes the user's intention based on text data and emotion data.

[0346] The "text generation means" is a device or software that generates appropriate text based on the user's intention and emotion data.

[0347] A "speech synthesis means" is a device or software that converts the generated text into speech data and adjusts the tone and speed of the speech based on emotional data.

[0348] A "verification and correction means" is a device or software that corrects and regenerates text according to a user's verification and correction instructions.

[0349] A "data storage means" is a device or software that stores the generated and modified text in its final document format.

[0350] This system allows users to generate text through voice input and confirm and correct it by voice. One of the features of this system is that it incorporates an emotion engine that recognizes the user's emotions, and controls the generated text and voice feedback to match the user's emotions. This improves the user experience and supports more efficient and healthy working styles.

[0351] This system consists of four main components: the user, the terminal, the server, and the emotion engine.

[0352] Audio capture and preprocessing

[0353] The user gives voice instructions using a dedicated microphone or the device's built-in microphone. The device collects the user's voice in real time using a voice capture module. Specifically, Python's speech_recognition library can be used. This voice data is sent to a speech recognition engine (for example, Google Cloud Speech-to-Text API) and converted into text data.

[0354] emotion recognition

[0355] The terminal sends the converted text data and voice data to an emotion engine (e.g., IBM Watson (registered trademark) Tone Analyzer). The emotion engine analyzes the user's emotion from the voice and text and generates emotion data. This emotion data is used for subsequent text generation and voice synthesis.

[0356] Intention Analysis and Text Generation

[0357] The server receives the emotion data and text data and analyzes the text data using a natural language processing engine (e.g., OpenAI's GPT-3). The server recognizes the user's intention based on the analysis results and sends a request to a chat generation AI (e.g., Microsoft® Azure® Chatbot) to generate text based on that intention, taking the emotion data into consideration. The generated text is expected to have a style and tone that matches the user's emotion.

[0358] Audio feedback and confirmation / correction

[0359] The generated text is sent to the device and converted into voice data by a speech synthesis engine (e.g., Amazon Polly). At this time, the tone and speed of the voice are adjusted based on the emotional data obtained from the emotion engine. The device then plays the adjusted voice data back to the user, prompting them to confirm the content. The user can check the content of the document through voice feedback and, if necessary, give voice instructions for corrections.

[0360] For example, if a user instructs the server to "Please prepare a sales report for the previous month," the server analyzes the instruction and generates text such as "Last month's sales increased by 10% compared to the plan." If the emotion engine recognizes that the user is nervous, the speech synthesis engine generates a voice with a slightly slower and calmer tone.

[0361] Save the final document

[0362] When the user makes a final confirmation and gives the instruction to "save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., Word, PDF) and saves it in a database or cloud storage. After the saving process is complete, the device notifies the user by voice that the save is complete.

[0363] Specific examples

[0364] Example 1:

[0365] Launch and start: The user says "Start report creation." The device captures the voice and converts it into text data using the speech recognition engine. At the same time, the emotion engine analyzes the user's current emotion and generates emotion data.

[0366] Intention analysis and text generation: The server analyzes the intention of "Start creating a report" and generates the text "Start creating a report. What would you like to do first?" taking into account the user's relaxation state.

[0367] Confirmation and correction: If the user instructs "Write about sales analysis," the text generated reflects that state of mind: "This month's sales increased by 5% compared to the same month last year." If the user requests a correction, such as "Please change the expression "compared to the same month last year" to "the percentage increase from the same month last year," the text is corrected based on that request.

[0368] Example 2:

[0369] Save: When the user says "Save this content," the server saves it as the final document and the device gives a voice notification that the save is complete.

[0370] In this way, the system of the present invention enables document creation and editing to be performed while taking the user's emotions into consideration, thereby improving the user experience and supporting efficient work methods.

[0371] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0372] Step 1:

[0373] The device uses a dedicated or built-in microphone to capture the user's voice. The user's voice instructions are provided as input. The device uses the Python speech_recognition library to collect voice data in real time. The collected voice data is sent to a speech recognition engine (e.g., Google Cloud Speech-to-Text API) which generates text data as output. This text data becomes the input for the next processing step.

[0374] Step 2:

[0375] The device sends the text data and voice data converted in step 1 to an emotion engine (for example, IBM Watson Tone Analyzer). The text data and voice data are provided as input. The emotion recognition engine analyzes this data and analyzes the user's emotions. Emotional data is generated as output, and this emotion data becomes the input to the next processing step. Specifically, the user's emotional state is identified by analyzing the choice of words read from the text data and the intonation of the voice data.

[0376] Step 3:

[0377] The server receives the emotion data generated in step 2 and the text data generated in step 1. Using these data as input, it analyzes the text data using a natural language processing engine (e.g., OpenAI's GPT-3). Based on the analysis results, it recognizes the user's intention and, taking the emotion data into consideration, sends a request to generate text to a chat generation AI (e.g., Microsoft Azure Chatbot). The generated text has a style and tone that matches the user's emotion. The appropriate generated text is obtained as output, and becomes the input for the next processing step.

[0378] Step 4:

[0379] The device receives the text data generated in step 3 and sends it to a speech synthesis engine (for example, Amazon Polly). The generated text data and emotion data are provided as input. The speech synthesis engine generates voice data while adjusting the tone and speed of the voice based on the emotion data. The generated voice data is obtained as output, and this is played back to the user. Specifically, the voice data is played back as feedback to the user, and the user confirms the content.

[0380] Step 5:

[0381] The user checks the content of the document based on the voice feedback played in step 4 and, if necessary, gives voice instructions for correction. The checked content and correction instructions are provided as input. The device uses a voice capture module to collect the user's correction instructions in real time and converts them into text data using a voice recognition engine again. The corrected text data is obtained as output and is sent again to the server for repeated processing.

[0382] Step 6:

[0383] When the user makes a final confirmation and issues the instruction "Save this content," the device sends this instruction to the server. The final confirmed text is provided as input. The server converts the confirmed text data into the final document format (e.g., PDF, Word, etc.). The specific output is a file in the converted document format, which is stored in a database or cloud storage. Finally, the device notifies the user by voice that the save is complete.

[0384] (Application example 2)

[0385] 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."

[0386] Conventional voice input systems generate text and provide voice feedback without considering the user's emotions, resulting in a flawed user experience. Furthermore, robots operating in factories lack the functionality to provide appropriate feedback based on emotions, which is insufficient for improving work efficiency and user satisfaction. To solve these problems, a system that can properly analyze the user's emotional state and reflect it in feedback is needed.

[0387] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0388] In this invention, the server includes means for capturing a user's voice, means for converting the captured voice into text data, means for analyzing the user's intention based on the converted text data, means for generating appropriate text based on the user's intention, means for converting the generated text into voice data and presenting it to the user, means for correcting and regenerating the text in accordance with the user's confirmation and correction instructions, means for saving the generated and corrected text in a final document format, means for capturing voice instructions and analyzing emotions, and means for adjusting the tone and speed of the voice data based on the emotion analysis results. This enables feedback that matches the user's emotions, thereby improving work efficiency in the factory and user satisfaction.

[0389] "Audio capture means" refers to devices or software for collecting a user's voice in real time.

[0390] "Speech recognition means" refers to the algorithms or software that convert collected voice data into text data.

[0391] "Intention analysis means" refers to algorithms or software that analyzes the converted text data and understands what the user intended.

[0392] "Text generation means" refers to algorithms or software that generate appropriate text based on the analyzed user intent.

[0393] "Speech synthesis means" refers to the algorithm or software used to convert the generated text into speech data.

[0394] "Text correction tool" refers to algorithms or software for correcting and regenerating text according to user confirmation and correction instructions.

[0395] "Document Storage" refers to the device or software used to store generated and modified text in its final document format.

[0396] "Emotion analysis means" refers to algorithms or software that analyze emotions from a user's voice or text and generate the results as data.

[0397] "Audio tone adjustment means" refers to algorithms or software for adjusting the tone or rate of audio data based on the results of emotion analysis.

[0398] The present invention aims to provide a factory robot management system that performs tasks based on voice instructions and provides feedback according to the user's emotions. The overall configuration of the system and specific embodiments thereof will be described below.

[0399] System Configuration

[0400] The system consists of the following major components:

[0401] 1. Users

[0402] 2. Terminal

[0403] 3. Server

[0404] 4. Sentiment Analysis Engine

[0405] User

[0406] The user gives voice instructions using a dedicated microphone or a microphone built into the device, providing work instructions to the factory robot in real time.

[0407] Terminal

[0408] The terminal collects the user's voice in real time using a voice capture means and transmits the voice data to the server. The terminal has the following functions.

[0409] Audio capture means: collects the user's voice in real time.

[0410] Speech recognition means: Converts collected voice data into text data.

[0411] Speech synthesis means: Converts the generated text into speech data and presents it to the user.

[0412] server

[0413] The server analyzes the user's intent and generates or modifies text and voice data using the following methods:

[0414] Intention analysis means: Analyze the converted text data to understand the user's intention.

[0415] Text generation means: Generates appropriate text based on the analyzed user intent.

[0416] Text correction means: Corrects and regenerates the text according to the user's confirmation and correction instructions.

[0417] Document storage means: storing the generated and modified text in the final document format.

[0418] Sentiment Analysis Engine

[0419] The emotion analysis engine analyzes emotions from the user's voice and text and generates the results as data. This emotion data is used as follows:

[0420] Sentiment analysis means: Analyzes emotions from the user's voice and text.

[0421] Voice tone adjustment means: Adjusts the tone and speed of voice data based on the results of emotion analysis.

[0422] System Operation

[0423] 1. Voice capture: The user dictates, "I will start the maintenance work on the machine." The device captures the voice and sends it to the server as voice data.

[0424] 2. Speech recognition: The server uses speech recognition means to convert this voice data into text data.

[0425] 3. Emotion analysis: The emotion analysis engine analyzes the text and voice data and generates data representing the user's emotions. In this case, it is analyzed that the user is nervous.

[0426] 4. Intention analysis and text generation: The server uses the intention analysis means to analyze the user's intention and generate text such as "Maintenance work has begun."

[0427] 5. Voice feedback: This text is converted into voice data using a voice synthesis means and played back to the user, with the tone and speed of the voice adjusted appropriately based on the sentiment analysis data.

[0428] 6. Confirm and correct: When the user says "Tell me the steps," the system regenerates the text and plays it back.

[0429] 7. Save document: When the user instructs "Save as this content," the server saves the generated text in the final document format.

[0430] Specific examples

[0431] Example 1:

[0432] User: "I'm going to start maintenance on the machine."

[0433] System: "Maintenance work will begin. Please advise how to proceed."

[0434] Example 2:

[0435] User: "Tell me how to do this."

[0436] System: "The first step is to power down the machine."

[0437] Prompt Sentence Examples

[0438] "I'm going to start maintenance on the machine. Please tell me the procedure."

[0439] This system enables feedback that matches the user's emotions, improving work efficiency in factories and user satisfaction.

[0440] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0441] Step 1:

[0442] Audio Capture

[0443] The user dictates, "I will start maintenance work on the machine."

[0444] Input: User's voice

[0445] The terminal collects the user's voice in real time using a voice capture means, and the collected voice data is then prepared for processing by a voice recognition engine.

[0446] Output: Audio data

[0447] Step 2:

[0448] Voice Recognition

[0449] Input: Audio data

[0450] The server converts the collected voice data into text data using a voice recognition engine. The voice recognition engine analyzes the phonemes in the voice data and generates corresponding character strings. This process provides the user's voice instructions in text format.

[0451] Output: Text data

[0452] Step 3:

[0453] Emotion analysis

[0454] Input: Text data, audio data

[0455] The server uses emotion analysis means to analyze the obtained text data and voice data and recognize the user's emotions. The emotion analysis engine analyzes the tone, speed, pitch, etc. of the sound and generates emotion data such as whether the user is tense or relaxed.

[0456] Output: Emotion data

[0457] Step 4:

[0458] Intention Analysis

[0459] Input: Text data, emotion data

[0460] The server uses intent analysis to analyze the text data and understand the user's intent. The natural language processing engine analyzes keywords and contextual information in the text data and recognizes the "intent to start maintenance work." This lays the foundation for generating an appropriate response.

[0461] Output: Intent data

[0462] Step 5:

[0463] Text Generation

[0464] Input: Intention data, emotion data

[0465] The server uses a text generation means to generate appropriate text based on the analyzed user intent. Since emotional data is also taken into account, the generated text is in a style and tone that matches the user's emotions. A response such as "Maintenance work will begin. Please tell me the procedure" is generated.

[0466] Output: Response text

[0467] Step 6:

[0468] Voice generation

[0469] Input: Response text, emotion data

[0470] The server uses a speech synthesis means to convert the generated response text into speech data. The tone and speed of the speech are appropriately adjusted based on the emotional data obtained from the emotion analysis means. For example, if the user is nervous, the speech speed is slowed down so that the speech can be played in a calmer tone.

[0471] Output: Modified audio data

[0472] Step 7:

[0473] Audio Feedback

[0474] Input: Modified audio data

[0475] The terminal plays the generated voice data to the user, and the user can confirm the system's response through this voice feedback and give the next instruction.

[0476] Output: User's understanding and next instructions

[0477] Step 8:

[0478] Check and correct

[0479] Input: The user's next voice command

[0480] If the user says "Tell me how to do it," the system recaptures this instruction and repeats the process of speech recognition, intent analysis, text generation, and speech generation. The resulting text is then reformatted.

[0481] Output: Corrected text and audio data

[0482] Step 9:

[0483] Document Storage

[0484] Input: Final confirmed text data

[0485] When the user selects "Save this content," the server saves the generated and modified text in the final document format (e.g., PDF, Word), and the saved document is stored in a database or cloud storage.

[0486] Output: Saved document

[0487] This system enables feedback that matches the user's emotions, improving work efficiency in factories and user satisfaction.

[0488] 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.

[0489] 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.

[0490] 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.

[0491] [Second embodiment]

[0492] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0493] 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.

[0494] 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).

[0495] 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.

[0496] 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.

[0497] 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).

[0498] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0499] 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.

[0500] 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.

[0501] 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.

[0502] 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.

[0503] 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."

[0504] This invention relates to a system that allows desk workers to create, check, and correct documents by voice without using their eyes. This system generates text data based on the user's voice input and allows checking and correction through voice feedback. Each component and its operation are explained below.

[0505] System Configuration

[0506] This system consists of three main components: users, terminals, and servers.

[0507] Audio capture and preprocessing

[0508] The user gives voice instructions using a dedicated microphone or a microphone built into the device, and the device collects the user's voice in real time using a voice capture module, allowing the user to start working without needing an input device.

[0509] Once the voice is captured, the device converts the voice data into text using a speech recognition engine that is highly accurate and takes into account a wide range of pronunciations and accents.

[0510] Intention Analysis and Text Generation

[0511] The converted text data is sent to the server and analyzed by a natural language processing engine. The server analyzes this text data to recognize the user's intent. For example, if a user is given an instruction such as "Please write about sales analysis," the server analyzes the intent and creates an appropriate text generation request.

[0512] Based on the analyzed intent, the server uses chat generation AI to generate text according to the instructions. The generated text reflects industry-standard formatting and structure, and is tailored to the user's needs.

[0513] Audio feedback and confirmation / correction

[0514] The generated text is sent to the device, which converts it into voice data using a speech synthesis engine. The device then plays the converted voice data back to the user, prompting them to confirm the content. The user can then use the voice feedback to confirm the content of the document and, if necessary, provide voice instructions for corrections.

[0515] For example, if a user says, "Please change the expression 'year-on-year comparison' to 'the percentage increase from the same month last year,'" the device captures this speech, converts it back into text data, and sends it to the server. The server again analyzes the intent, corrects it as necessary, and regenerates the text. This process is repeated until the user is satisfied.

[0516] Save the final document

[0517] When the user makes a final confirmation and gives the instruction to "save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., Word, PDF) and saves it in a database or cloud storage. After the saving process is complete, the device will inform the user by voice that the save is complete.

[0518] Specific examples

[0519] Example 1:

[0520] Launch and Start: When the user says "Start creating a report," the device captures the voice, converts it into text using a speech recognition engine, and sends it to the server.

[0521] Intent analysis and text generation: The server analyzes the intent of "Start creating a report," generates the text "Start creating a report. What would you like as your first item?", and sends it to the device.

[0522] Confirm and correct: If the user instructs "Write about sales analysis," the same process is followed to generate the text "This month's sales increased by 5% compared to the same month last year." The user then requests corrections, and the final content is "This month's sales increased by 5% compared to the same month last year."

[0523] Example 2:

[0524] Save: When the user says "Save this content," the server saves it as the final document and the device notifies the user by voice that the save is complete.

[0525] In this way, the system of the present invention enables desk workers to create and edit documents without using their eyes, reducing eye strain and supporting healthy working styles.

[0526] The processing flow will be explained below.

[0527] Step 1:

[0528] A user logs into a dedicated application or device and dictates "Start report creation."

[0529] Step 2:

[0530] The device captures the user's voice through the microphone and stores the voice data in a buffer.

[0531] Step 3:

[0532] The voice data captured by the device is sent to a voice recognition library and converted into text data.

[0533] Step 4:

[0534] The voice recognition library converts the voice data into text data and sends it back to the device.

[0535] Step 5:

[0536] The terminal encodes the converted text data and sends it to the server as an HTTP request.

[0537] Step 6:

[0538] The server decodes the encoded text data and analyzes it using a natural language processing engine.

[0539] Step 7:

[0540] The server recognizes the user's intention based on the analysis results and sends a request to the chat generation AI to generate text based on that intention.

[0541] Step 8:

[0542] The chat generation AI generates appropriate text based on the instructions and sends it back to the server.

[0543] Step 9:

[0544] The server receives the generated text data and sends it to the terminal as an HTTP response.

[0545] Step 10:

[0546] The text data received by the terminal is sent to a speech synthesis engine and converted into voice data.

[0547] Step 11:

[0548] The speech synthesis engine converts the text into speech data and sends it back to the device.

[0549] Step 12:

[0550] The terminal plays back the generated voice data to the user, prompting them to confirm the content.

[0551] Step 13:

[0552] The user confirms the result through voice feedback and, if necessary, gives voice instructions for correction.

[0553] Step 14:

[0554] The device captures the correction instructions and converts them into text data using a voice recognition library again.

[0555] Step 15:

[0556] The terminal transmits the corrected text data to the server.

[0557] Step 16:

[0558] The server parses the modified text data and regenerates the text as necessary.

[0559] Step 17:

[0560] The server transmits the regenerated text data to the terminal, where it is converted back into voice data and played back to the user.

[0561] Step 18:

[0562] This process is repeated, with revisions and validations being made until the user is satisfied.

[0563] Step 19:

[0564] The user issues a voice command saying, "Please save this content."

[0565] Step 20:

[0566] The terminal sends a final instruction to the server.

[0567] Step 21:

[0568] The server converts the verified text into a final document format (e.g., Word, PDF) and stores it in a database or cloud storage.

[0569] Step 22:

[0570] The server sends a notification of completion of saving to the terminal.

[0571] Step 23:

[0572] The device will notify the user by voice when the save is complete.

[0573] This allows users to create, edit, and save documents without using their eyes, reducing physical strain.

[0574] Example 1

[0575] 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."

[0576] Modern desk workers are suffering from health problems such as eye strain and vision loss due to long hours of computer work. Creating, reviewing, and correcting documents in particular places a heavy visual burden on the eyes, reducing efficiency and causing health problems. Furthermore, traditional document creation methods are difficult for visually impaired users. A system that can solve these problems and achieve a more efficient and healthy work environment is needed.

[0577] 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.

[0578] In this invention, the server includes a means for capturing a user's voice, a means for converting the captured voice into text data, and a means for analyzing the user's intent based on the converted text data. This allows the user to create, review, and modify documents using only voice instructions. Furthermore, by providing a means for creating documents using a generative AI model and a means for saving the final formatted document in data storage, efficient and accurate document creation and storage can be achieved. This reduces visual strain and supports healthy working styles.

[0579] A "means for capturing audio" is a device or system for capturing audio uttered by a user and processing it as a digital signal.

[0580] The "means for converting into text data" is a technique for analyzing the captured audio data and converting it into a corresponding text format.

[0581] "Means for analyzing user intent" refers to natural language processing technology used to understand the converted text data and identify what the user is looking for.

[0582] "Means for generating documents" refers to technology that uses a generative AI model to create text according to the specified content based on the user's intentions.

[0583] The "means for converting the text into audio data and presenting it to the user" refers to a technique for converting the generated or corrected text into audio and providing it to the user as an anti-visual interface.

[0584] The "means for correcting and regenerating a document according to confirmation and correction instructions" is a processing system for correcting the contents of a document in response to feedback from a user and regenerating the document as necessary.

[0585] "Means for saving in final format" refers to the technology for converting the completed document into a specified file format and saving it in a database, cloud storage, etc.

[0586] A "data analysis module" is a program or function that analyzes data based on the user's voice instructions and sends appropriate instructions to the generative AI model.

[0587] A "generative AI model" is an artificial intelligence technology that generates text based on a given prompt.

[0588] "Data storage" refers to a system or device for storing and managing data over the long term.

[0589] This invention relates to a system that allows desk workers to create, check, and correct documents by voice without using their eyes. This system generates text data based on the user's voice input and allows checking and correction through voice feedback. Each component and its operation are explained below.

[0590] System Configuration

[0591] This system consists of three main components: users, terminals, and servers.

[0592] Audio capture and preprocessing

[0593] The user gives voice instructions using a dedicated microphone or a microphone built into the device, and the device collects the user's voice in real time using a voice capture module, allowing the user to start working without needing an input device.

[0594] Voice Recognition

[0595] Once the audio is captured, the device converts it into text using a speech recognition engine (e.g., Google Cloud Speech-to-Text) that has high accuracy and takes into account various pronunciations and accents.

[0596] Intention Analysis and Text Generation

[0597] The converted text data is sent to the server and analyzed by a natural language processing engine (e.g., spaCy). The server analyzes this text data to recognize the user's intent. For example, if an instruction such as "Please write about sales analysis" is given, the server analyzes the intent and creates an appropriate text generation request.

[0598] Based on the parsed intent, the server uses a generative AI model (e.g., OpenAI GPT-3) to generate text based on the instructions, reflecting industry-standard formatting and structure and tailored to the user's needs.

[0599] Audio feedback and confirmation / correction

[0600] The generated text is sent to the device, which converts it into voice data using a speech synthesis engine (e.g., Amazon Polly). The device then plays the converted voice data back to the user, prompting them to confirm the content. The user can then confirm the content of the document through voice feedback and, if necessary, provide voice instructions for correction.

[0601] For example, if a user says, "Please change the expression 'compared to the same month last year' to 'the percentage increase from the same month last year,'" the device captures this speech, converts it into text data using a speech recognition engine, and sends it to the server. The server again analyzes the intent, corrects the text as necessary, and regenerates it. This process is repeated until the user is satisfied.

[0602] Save the final document

[0603] When the user makes a final confirmation and gives the instruction "Save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., Word, PDF) and saves it in a database or cloud storage (e.g., Google Drive, Dropbox). After the saving process is complete, the device will inform the user by voice that the save is complete.

[0604] Specific examples

[0605] Below is a concrete example of how this system can be used.

[0606] Example 1:

[0607] Boot and start:

[0608] When a user says "start report creation," the device captures the voice, converts it into text using a speech recognition engine, and sends it to the server.

[0609] The server analyzes the intent of "Start creating a report," generates the text "Start creating a report. What would you like as your first item?", and sends it to the terminal.

[0610] The user gives the voice instruction "Write about your sales analysis."

[0611] The generative AI model generates the text, "Sales this month increased by 5% compared to the same month last year."

[0612] The user further instructs, "Please change the year-on-year comparison to the increase rate from the same month last year."

[0613] The device captures the audio again and sends correction instructions to the server.

[0614] The server generates the text "Sales this month increased 5% from the same month last year" reflecting the correction.

[0615] Example 2:

[0616] keep:

[0617] When the user says "Save this content," the device captures the audio, converts it into text, and sends it to the server.

[0618] The server saves it as the final document, and the device notifies you with an audio message that the save is complete.

[0619] As described above, the system of the present invention allows desk workers to create, check, and edit documents using only voice commands, thereby reducing visual strain and supporting healthy working styles.

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

[0621] Step 1: Capture audio

[0622] Specific description:

[0623] Input: The user gives a voice command such as "Start report creation."

[0624] Specific operation: The user's voice is captured using a dedicated microphone or the device's built-in microphone.

[0625] Data processing: The terminal collects audio data using an audio capture module.

[0626] Output: The captured voice data is passed to the speech recognition engine.

[0627] Step 2: Voice Recognition

[0628] Specific description:

[0629] Input: Captured audio data.

[0630] Specific operation: The terminal converts voice data into text data using a voice recognition engine (e.g., Google Cloud Speech-to-Text).

[0631] Data processing: The speech recognition engine analyzes the speech waveform and converts it into a corresponding string of characters.

[0632] Output: The converted text data is generated.

[0633] Step 3: Intent Analysis

[0634] Specific description:

[0635] Input: Text data output from the speech recognition engine.

[0636] Specific operation: The server receives the text data sent from the terminal and analyzes it using a natural language processing engine (e.g., spaCy).

[0637] Data processing: Natural language processing engines analyze text data to identify user intent.

[0638] Output: Analysis results that reflect the user's intentions are generated.

[0639] Step 4: Text Generation

[0640] Specific description:

[0641] Input: Analysis results obtained from intent analysis.

[0642] Specific operation: The server uses a generative AI model (e.g., OpenAI GPT-3) to generate text based on the user's intent.

[0643] Data processing: The generative AI model generates relevant text based on the input prompt.

[0644] Output: Generated text data is generated.

[0645] Step 5: Audio feedback

[0646] Specific description:

[0647] Input: Generated text data.

[0648] Specific operation: The device converts the generated text data into audio data using a speech synthesis engine (e.g., Amazon Polly) and plays it back.

[0649] Data processing: The speech synthesis engine converts text data into speech waveforms.

[0650] Output: Audio data is generated that is played back to the user.

[0651] Step 6: Accepting correction instructions

[0652] Specific description:

[0653] Input: User's spoken correction instructions (e.g., "Please change the expression 'year-over-year comparison' to 'percent increase from the same month last year'").

[0654] Specific operation: The device again captures the user's voice correction instructions and converts them into text data using a voice recognition engine.

[0655] Data processing: The new voice command is converted into text and sent to the server.

[0656] Output: Text data containing correction instructions is passed to the server.

[0657] Step 7: Fix and Regenerate

[0658] Specific description:

[0659] Input: Text data containing correction instructions.

[0660] Specific behavior: The server parses the correction instructions, corrects the text as necessary, and regenerates it.

[0661] Data processing: The data analysis module analyzes the correction instructions and sends new prompts to the generative AI model to regenerate the text.

[0662] Output: The corrected text data is generated.

[0663] Step 8: Save the final document

[0664] Specific description:

[0665] Input: Final text data after correction (e.g., "Sales this month increased 5% from the same month last year.").

[0666] Specific operation: The server converts the final text data into a document format (e.g., Word, PDF) and saves it in data storage (e.g., Google Drive, Dropbox).

[0667] Data processing: The text data is converted into the specified file format and uploaded to cloud storage.

[0668] Output: A sound is generated to notify the user that the save is complete.

[0669] (Application example 1)

[0670] 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."

[0671] The objective of this invention is to provide a system that reduces the burden of deskwork in a factory environment and allows users to create, check, modify, and finally save reports using only voice commands. This system allows users to operate it by voice without using their eyes, improving worker health and increasing productivity. Furthermore, in a voice-only work environment, rapid and accurate data processing is required.

[0672] 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.

[0673] In this invention, the server includes means for capturing a user's voice, means for converting the captured voice into text data, means for analyzing the user's intent based on the converted text data, means for generating appropriate text based on the user's intent, means for converting the generated text into voice data and presenting it to the user, means for correcting and regenerating the text according to the user's confirmation and correction instructions, means for saving the generated and corrected text in a final document format, and robot means for realizing use in a voice-specific work environment, a factory environment where reports are created by voice instructions. This allows workers to create documents using only voice instructions, reducing eye fatigue and improving work efficiency.

[0674] "User" refers to the person who operates the system and gives voice instructions.

[0675] "Audio capturing means" refers to a device or method that captures the audio produced by a user as electronic data.

[0676] "Means for converting voice into text data" refers to technology that analyzes captured voice data and converts it into corresponding text data.

[0677] "Means of intent analysis" refers to technology that understands what a user wants from text data and identifies appropriate actions.

[0678] "Means for generating appropriate text" refers to technology that automatically creates the necessary information and sentences based on the user's intentions.

[0679] The "means for converting the generated text into audio data and presenting it to the user" refers to a device or method for converting the generated text back into audio data and allowing the user to listen to it.

[0680] "Means for correcting and regenerating text" refers to a technique for revising existing text and generating new text according to user instructions.

[0681] "Means for saving in final document format" refers to a technique for saving the final verified text in a specified format.

[0682] A "voice-focused work environment" refers to a workplace where operations and instructions are primarily performed by voice.

[0683] "Use in a factory environment where reports are created using voice instructions" refers to performing a series of tasks such as creating, modifying, and saving reports using only voice within the factory.

[0684] "Robotic means" refers to automated equipment or systems that implement voice commands in a factory environment.

[0685] This invention is a system for efficiently creating reports using voice instructions in a factory environment. The system generates text data based on the user's voice input and allows confirmation and correction through voice feedback. The specific configuration and processing procedure for realizing this system will be described below.

[0686] System Configuration

[0687] The system consists of four main components: users, terminals, servers, and robots.

[0688] 1. Audio capture and preprocessing

[0689] The user gives voice instructions using a dedicated microphone or the device's built-in microphone. The device collects the user's voice in real time using a voice capture module. The collected voice data is converted into text data using a voice recognition engine (e.g., Google Speech-to-Text API).

[0690] 2. Intention Analysis and Text Generation

[0691] The converted text data is sent to a server and analyzed by a natural language processing engine (e.g., OpenAI GPT-4). The server analyzes this text data to recognize the user's intent. For example, if an instruction such as "Please write about sales analysis" is given, the server analyzes the intent and creates an appropriate text generation request. Based on the analyzed intent, the server uses a generative AI model to generate text according to the instruction.

[0692] 3. Audio feedback and confirmation / correction

[0693] The generated text is sent to the device, which converts it into audio data using a speech synthesis engine (e.g., Google Text-to-Speech API). The device then plays the converted audio data back to the user, prompting them to confirm the content. The user can then use the audio feedback to confirm the content of the document and, if necessary, provide voice instructions for corrections.

[0694] 4. Preservation of the final document

[0695] When the user makes a final confirmation and gives the instruction to "save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., PDF) and saves it in a database or cloud storage (e.g., AWS S3). After the saving process is complete, the device will inform the user by voice that the save is complete.

[0696] Specific examples

[0697] Example 1: Starting a report

[0698] User: Says "Start creating a report."

[0699] The device collects the voice using a voice capture module and converts it into text data such as "Start creating report" using a voice recognition engine.

[0700] The server receives this text, generates the text "Let's start creating a report. What would you like as the first item?" and sends it to the terminal.

[0701] The terminal uses a speech synthesis engine to convert this text into audio data and play it back to the user.

[0702] Example 2: Sales analysis instructions and corrections

[0703] User: "Write about your sales analysis."

[0704] The terminal collects voice using a voice capture module and converts it into text data using a voice recognition engine.

[0705] The server analyzes this text data, recognizes the intent of "Please write about sales analysis," and generates the text "This month's sales increased by 5% compared to the same month last year."

[0706] The terminal converts the generated text into voice data using a voice synthesis engine and plays it back to the user.

[0707] When the user instructs, "Please change it to the increase rate compared to the same month last year," the device captures this voice, converts it back into text data, and sends it to the server.

[0708] The server again parses the intent and generates the corrected text, "Sales this month increased 5% from the same month last year."

[0709] The terminal converts the corrected text into voice data using a voice synthesis engine and plays it back to the user.

[0710] Prompt Sentence Examples

[0711] "Please write a report on your sales analysis. Please include specific figures and year-on-year increases or decreases."

[0712] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0713] Step 1: User initiates voice command

[0714] The user gives voice instructions (such as "Start creating a report") using a dedicated microphone or the device's built-in microphone. The device collects voice data in real time using a voice capture module. At this time, the user's voice is input, and the captured voice data is output.

[0715] Step 2: Convert the audio data to text

[0716] The device converts the collected voice data into text data using a voice recognition engine (e.g., Google Speech-to-Text API). Specifically, the captured voice data is input and passed through an advanced voice analysis algorithm to generate corresponding text data. This text data is then output.

[0717] Step 3: Analyze the text data

[0718] The server receives the converted text data and analyzes the user's intent through a natural language processing engine (e.g., OpenAI GPT-4). For example, it interprets the purpose of an instruction such as "start creating a report." In this case, the input is text data, and the output is the analysis result (data indicating the user's intent).

[0719] Step 4: Generate the appropriate text

[0720] The server uses a generative AI model based on the results of intent analysis to generate the required text. Based on the converted text data, a response text such as "I'm starting to write a report. What would you like as the first item?" is generated. The input is the analysis result, and the output is the generated text data.

[0721] Step 5: Convert the generated text to audio data

[0722] The device receives the generated text data and converts it into voice data using a speech synthesis engine (e.g., Google Text-to-Speech API). Specifically, it takes the generated text data as input and outputs it as voice waveform data. This voice data is then played back to the user.

[0723] Step 6: Receive user confirmation and correction instructions

[0724] The user can check the generated text through voice feedback and make corrections as necessary. For example, they can say, "Please change it to the increase rate compared to the same month last year." This voice instruction is input, captured by the device, and converted back into text data through the voice recognition engine (output).

[0725] Step 7: Regenerate the text based on the correction instructions

[0726] The server receives the text data again, analyzes the intent, and makes the necessary corrections. For example, based on the instruction "Please change this to the increase rate compared to the same month last year," the corrected text "This month's sales increased by 5% compared to the same month last year" is generated. The input is the text data of the correction instruction, and the output is the corrected text data.

[0727] Step 8: Convert the corrected text back into audio data and present it

[0728] The corrected text data is again converted into voice data by the terminal using a speech synthesis engine. It is then played back to the user for reconfirmation. This process is repeated until the user is satisfied. The input is the corrected text data, and the output is the corrected voice data.

[0729] Step 9: Save in the final document format

[0730] Finally, when the user instructs "Save this content," the server converts the final confirmed text data into a specified document format (e.g., PDF) and saves it in cloud storage (e.g., AWS S3). The input is the final confirmed text data, and the output is the saved document. The device notifies the user by voice when saving is complete.

[0731] 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.

[0732] This invention relates to a system that allows users to generate text through voice input and confirm and correct it by voice. A feature of this invention is that by incorporating an emotion engine that recognizes the user's emotions, the generated text and voice feedback are controlled to match the user's emotions. This improves the user experience and supports healthier working styles.

[0733] System Configuration

[0734] This system consists of four main components: the user, the terminal, the server, and the emotion engine.

[0735] Audio capture and preprocessing

[0736] The user gives voice instructions using a dedicated microphone or a microphone built into the device. The device collects the user's voice in real time using a voice capture module. This voice data is sent to a voice recognition engine and converted into text data.

[0737] emotion recognition

[0738] The converted text data and voice data are sent by the terminal to the emotion engine, which analyzes the user's emotions from the voice and text and generates emotion data. This emotion data is used for subsequent text generation and voice synthesis.

[0739] Intention Analysis and Text Generation

[0740] The server receives the text data and emotion data and analyzes the text data using a natural language processing engine. The server recognizes the user's intention based on the analysis results, and sends a request to the chat generation AI to generate text based on that intention, taking into account the emotion data. The generated text is expected to have a style and tone that matches the user's emotion.

[0741] Audio feedback and confirmation / correction

[0742] The generated text is sent to the device and converted into voice data by a speech synthesis engine. At this time, the tone and speed of the voice are adjusted based on the emotional data obtained from the emotion engine. The device then plays the adjusted voice data back to the user, prompting them to confirm the content. The user can check the content of the document through voice feedback and, if necessary, give voice instructions for corrections.

[0743] For example, if a user instructs the server to "Please prepare a sales report for the previous month," the server analyzes the instruction and generates text such as "Last month's sales increased by 10% compared to the plan." If the emotion engine recognizes that the user is nervous, the speech synthesis engine generates speech in a slightly slower, calmer tone.

[0744] Save the final document

[0745] When the user makes a final confirmation and gives the instruction to "save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., Word, PDF) and saves it in a database or cloud storage. After the saving process is complete, the device will inform the user by voice that the save is complete.

[0746] Specific examples

[0747] Example 1:

[0748] Launch and Start: The user says "Start report creation." The device captures the voice and converts it into text data using the speech recognition engine. At the same time, the emotion engine analyzes the user's current emotion and generates emotion data.

[0749] Intention analysis and text generation: The server analyzes the intention of "Start creating a report" and generates the text "Start creating a report. What would you like to do first?" taking into account the user's relaxation state.

[0750] Confirmation and correction: If the user instructs "Please write about sales analysis," the text generated reflects that state of mind: "This month's sales increased by 5% compared to the same month last year." If the user requests a correction, such as "Please change the expression "compared to the same month last year" to "the percentage increase from the same month last year," the text is corrected based on that request.

[0751] Example 2:

[0752] Save: When the user says "Save this content," the server saves it as the final document and the device notifies the user by voice that the save is complete.

[0753] In this way, the system of the present invention allows document creation and editing to be performed while taking the user's emotions into consideration, improving the user experience while reducing eye strain.

[0754] The processing flow will be explained below.

[0755] Step 1:

[0756] A user logs into a dedicated application or device and dictates "Start report creation."

[0757] Step 2:

[0758] The device captures the user's voice through the microphone and stores the voice data in a buffer.

[0759] Step 3:

[0760] The voice data captured by the device is sent to a voice recognition library and converted into text data.

[0761] Step 4:

[0762] The voice recognition library converts the voice data into text data and sends it back to the device.

[0763] Step 5:

[0764] The terminal transmits the converted text data to an emotion engine, which analyzes the user's emotions.

[0765] Step 6:

[0766] The emotion engine analyzes the voice and text data, generates the user's emotion data, and sends it back to the terminal.

[0767] Step 7:

[0768] The device encodes the text data and emotion data and sends them to the server as an HTTP request.

[0769] Step 8:

[0770] The server decodes the encoded text data and emotion data and analyzes the text data using a natural language processing engine.

[0771] Step 9:

[0772] The server recognizes the user's intention based on the analysis results, takes emotional data into consideration, and sends a request for text generation to the chat generation AI.

[0773] Step 10:

[0774] The chat generation AI generates appropriate text based on the intent and emotion data and sends it back to the server.

[0775] Step 11:

[0776] The server receives the generated text data and sends it to the terminal as an HTTP response.

[0777] Step 12:

[0778] The text data received by the terminal is sent to a speech synthesis engine and converted into voice data based on emotional data.

[0779] Step 13:

[0780] The speech synthesis engine takes into account the emotional data, converts the text into speech data, and sends it back to the device.

[0781] Step 14:

[0782] The terminal plays back the generated voice data to the user, prompting them to confirm the content.

[0783] Step 15:

[0784] The user confirms the result through voice feedback and, if necessary, gives voice instructions for correction.

[0785] Step 16:

[0786] The device captures the correction instructions and converts them into text data using a voice recognition library again.

[0787] Step 17:

[0788] The terminal transmits the corrected text data and emotion data to the server.

[0789] Step 18:

[0790] The server parses the modified text data and regenerates the text as necessary.

[0791] Step 19:

[0792] The server transmits the regenerated text data to the terminal, where it is converted back into voice data and played back to the user.

[0793] Step 20:

[0794] This process is repeated, with revisions and validations being made until the user is satisfied.

[0795] Step 20:

[0796] The user issues a voice command saying, "Please save this content."

[0797] Step 22:

[0798] The terminal sends a final instruction to the server.

[0799] Step 23:

[0800] The server converts the verified text into a final document format (e.g., Word, PDF) and stores it in a database or cloud storage.

[0801] Step 24:

[0802] The server sends a notification of completion of saving to the terminal.

[0803] Step 25:

[0804] The device will notify the user by voice when the save is complete.

[0805] This allows users to create, edit and save documents without using their eyes, and when combined with an emotion engine, provides more relevant and emotionally tailored feedback.

[0806] Example 2

[0807] 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."

[0808] Conventional document creation systems fail to consider the user's emotions when generating text using voice instructions and then reviewing and correcting it. This prevents the system from adjusting text and voice feedback based on the user's emotional state, resulting in a lack of user experience. This also makes it difficult to adjust the tempo and mood while creating documents, hindering efficient and healthy work practices. To address these issues, a system is needed that can recognize the user's emotions in real time and generate text and provide voice feedback that adapts to those emotions.

[0809] 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.

[0810] In this invention, the server includes means for capturing a user's voice, means for converting the captured voice into text data, means for analyzing the user's emotions based on the converted text data and voice data and generating emotion data, means for analyzing the user's intention based on the emotion data and text data, means for generating appropriate text based on the user's intention and emotion data, means for converting the generated text into voice data, adjusting the tone and speed of the voice based on the emotion data and presenting it to the user, means for correcting and regenerating the text in accordance with the user's confirmation and correction instructions, and means for saving the generated and corrected text in a final document format. This enables text generation and voice feedback adapted to the user's emotions, improving the user experience and realizing an efficient and healthy way of working.

[0811] An "audio capture means" is a device or software for collecting a user's voice in real time.

[0812] "Speech recognition means" refers to a device or software for converting collected voice data into text data.

[0813] The "emotion recognition means" is a device or software that analyzes the user's emotions based on the converted text data and voice data, and generates emotion data.

[0814] The "intention analysis means" is a device or software that analyzes the user's intention based on text data and emotion data.

[0815] The "text generation means" is a device or software that generates appropriate text based on the user's intention and emotion data.

[0816] A "speech synthesis means" is a device or software that converts the generated text into speech data and adjusts the tone and speed of the speech based on emotional data.

[0817] A "verification and correction means" is a device or software that corrects and regenerates text according to a user's verification and correction instructions.

[0818] A "data storage means" is a device or software that stores the generated and modified text in its final document format.

[0819] This system allows users to generate text through voice input and confirm and correct it by voice. One of the features of this system is that it incorporates an emotion engine that recognizes the user's emotions, and controls the generated text and voice feedback to match the user's emotions. This improves the user experience and supports more efficient and healthy working styles.

[0820] This system consists of four main components: the user, the terminal, the server, and the emotion engine.

[0821] Audio capture and preprocessing

[0822] The user gives voice instructions using a dedicated microphone or the device's built-in microphone. The device collects the user's voice in real time using a voice capture module. Specifically, Python's speech_recognition library can be used. This voice data is sent to a speech recognition engine (for example, Google Cloud Speech-to-Text API) and converted into text data.

[0823] emotion recognition

[0824] The device sends the converted text data and voice data to an emotion engine (e.g., IBM Watson Tone Analyzer). The emotion engine analyzes the user's emotions from the voice and text and generates emotion data. This emotion data is used for subsequent text generation and voice synthesis.

[0825] Intention Analysis and Text Generation

[0826] The server receives the emotion data and text data and analyzes the text data using a natural language processing engine (e.g., OpenAI's GPT-3). The server recognizes the user's intent based on the analysis results and sends a request to a chat generation AI (e.g., Microsoft Azure Chatbot) to generate text based on that intent, taking the emotion data into consideration. The generated text is expected to have a style and tone that matches the user's emotions.

[0827] Audio feedback and confirmation / correction

[0828] The generated text is sent to the device and converted into voice data by a speech synthesis engine (e.g., Amazon Polly). At this time, the tone and speed of the voice are adjusted based on the emotional data obtained from the emotion engine. The device then plays the adjusted voice data back to the user, prompting them to confirm the content. The user can check the content of the document through voice feedback and, if necessary, give voice instructions for corrections.

[0829] For example, if a user instructs the server to "Please prepare a sales report for the previous month," the server analyzes the instruction and generates text such as "Last month's sales increased by 10% compared to the plan." If the emotion engine recognizes that the user is nervous, the speech synthesis engine generates a voice with a slightly slower and calmer tone.

[0830] Save the final document

[0831] When the user makes a final confirmation and gives the instruction to "save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., Word, PDF) and saves it in a database or cloud storage. After the saving process is complete, the device notifies the user by voice that the save is complete.

[0832] Specific examples

[0833] Example 1:

[0834] Launch and start: The user says "Start report creation." The device captures the voice and converts it into text data using the speech recognition engine. At the same time, the emotion engine analyzes the user's current emotion and generates emotion data.

[0835] Intention analysis and text generation: The server analyzes the intention of "Start creating a report" and generates the text "Start creating a report. What would you like to do first?" taking into account the user's relaxation state.

[0836] Confirmation and correction: If the user instructs "Write about sales analysis," the text generated reflects that state of mind: "This month's sales increased by 5% compared to the same month last year." If the user requests a correction, such as "Please change the expression "compared to the same month last year" to "the percentage increase from the same month last year," the text is corrected based on that request.

[0837] Example 2:

[0838] Save: When the user says "Save this content," the server saves it as the final document and the device gives a voice notification that the save is complete.

[0839] In this way, the system of the present invention enables document creation and editing to be performed while taking the user's emotions into consideration, thereby improving the user experience and supporting efficient work methods.

[0840] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0841] Step 1:

[0842] The device uses a dedicated or built-in microphone to capture the user's voice. The user's voice instructions are provided as input. The device uses the Python speech_recognition library to collect voice data in real time. The collected voice data is sent to a speech recognition engine (e.g., Google Cloud Speech-to-Text API) which generates text data as output. This text data becomes the input for the next processing step.

[0843] Step 2:

[0844] The device sends the text data and voice data converted in step 1 to an emotion engine (for example, IBM Watson Tone Analyzer). The text data and voice data are provided as input. The emotion recognition engine analyzes this data and analyzes the user's emotions. Emotional data is generated as output, and this emotion data becomes the input to the next processing step. Specifically, the user's emotional state is identified by analyzing the choice of words read from the text data and the intonation of the voice data.

[0845] Step 3:

[0846] The server receives the emotion data generated in step 2 and the text data generated in step 1. Using these data as input, it analyzes the text data using a natural language processing engine (e.g., OpenAI's GPT-3). Based on the analysis results, it recognizes the user's intention and, taking the emotion data into consideration, sends a request to generate text to a chat generation AI (e.g., Microsoft Azure Chatbot). The generated text has a style and tone that matches the user's emotion. The appropriate generated text is obtained as output, and becomes the input for the next processing step.

[0847] Step 4:

[0848] The device receives the text data generated in step 3 and sends it to a speech synthesis engine (for example, Amazon Polly). The generated text data and emotion data are provided as input. The speech synthesis engine generates voice data while adjusting the tone and speed of the voice based on the emotion data. The generated voice data is obtained as output, and this is played back to the user. Specifically, the voice data is played back as feedback to the user, and the user confirms the content.

[0849] Step 5:

[0850] The user checks the content of the document based on the voice feedback played in step 4 and, if necessary, gives voice instructions for correction. The checked content and correction instructions are provided as input. The device uses a voice capture module to collect the user's correction instructions in real time and converts them into text data using a voice recognition engine again. The corrected text data is obtained as output and is sent again to the server for repeated processing.

[0851] Step 6:

[0852] When the user makes a final confirmation and issues the instruction "Save this content," the device sends this instruction to the server. The final confirmed text is provided as input. The server converts the confirmed text data into the final document format (e.g., PDF, Word, etc.). The specific output is a file in the converted document format, which is stored in a database or cloud storage. Finally, the device notifies the user by voice that the save is complete.

[0853] (Application example 2)

[0854] 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."

[0855] Conventional voice input systems generate text and provide voice feedback without considering the user's emotions, resulting in a flawed user experience. Furthermore, robots operating in factories lack the functionality to provide appropriate feedback based on emotions, which is insufficient for improving work efficiency and user satisfaction. To solve these problems, a system that can properly analyze the user's emotional state and reflect it in feedback is needed.

[0856] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0857] In this invention, the server includes means for capturing a user's voice, means for converting the captured voice into text data, means for analyzing the user's intention based on the converted text data, means for generating appropriate text based on the user's intention, means for converting the generated text into voice data and presenting it to the user, means for correcting and regenerating the text in accordance with the user's confirmation and correction instructions, means for saving the generated and corrected text in a final document format, means for capturing voice instructions and analyzing emotions, and means for adjusting the tone and speed of the voice data based on the emotion analysis results. This enables feedback that matches the user's emotions, thereby improving work efficiency in the factory and user satisfaction.

[0858] "Audio capture means" refers to devices or software for collecting a user's voice in real time.

[0859] "Speech recognition means" refers to the algorithms or software that convert collected voice data into text data.

[0860] "Intention analysis means" refers to algorithms or software that analyzes the converted text data and understands what the user intended.

[0861] "Text generation means" refers to algorithms or software that generate appropriate text based on the analyzed user intent.

[0862] "Speech synthesis means" refers to the algorithm or software used to convert the generated text into speech data.

[0863] "Text correction tool" refers to algorithms or software for correcting and regenerating text according to user confirmation and correction instructions.

[0864] "Document Storage" refers to the device or software used to store generated and modified text in its final document format.

[0865] "Emotion analysis means" refers to algorithms or software that analyze emotions from a user's voice or text and generate the results as data.

[0866] "Audio tone adjustment means" refers to algorithms or software for adjusting the tone or rate of audio data based on the results of emotion analysis.

[0867] The present invention aims to provide a factory robot management system that performs tasks based on voice instructions and provides feedback according to the user's emotions. The overall configuration of the system and specific embodiments thereof will be described below.

[0868] System Configuration

[0869] The system consists of the following major components:

[0870] 1. Users

[0871] 2. Terminal

[0872] 3. Server

[0873] 4. Sentiment Analysis Engine

[0874] User

[0875] The user gives voice instructions using a dedicated microphone or a microphone built into the device, providing work instructions to the factory robot in real time.

[0876] Terminal

[0877] The terminal collects the user's voice in real time using a voice capture means and transmits the voice data to the server. The terminal has the following functions.

[0878] Audio capture means: collects the user's voice in real time.

[0879] Speech recognition means: Converts collected voice data into text data.

[0880] Speech synthesis means: Converts the generated text into speech data and presents it to the user.

[0881] server

[0882] The server analyzes the user's intent and generates or modifies text and voice data using the following methods:

[0883] Intention analysis means: Analyze the converted text data to understand the user's intention.

[0884] Text generation means: Generates appropriate text based on the analyzed user intent.

[0885] Text correction means: Corrects and regenerates the text according to the user's confirmation and correction instructions.

[0886] Document storage means: storing the generated and modified text in the final document format.

[0887] Sentiment Analysis Engine

[0888] The emotion analysis engine analyzes emotions from the user's voice and text and generates the results as data. This emotion data is used as follows:

[0889] Sentiment analysis means: Analyzes emotions from the user's voice and text.

[0890] Voice tone adjustment means: Adjusts the tone and speed of voice data based on the results of emotion analysis.

[0891] System Operation

[0892] 1. Voice capture: The user dictates, "I will start the maintenance work on the machine." The device captures the voice and sends it to the server as voice data.

[0893] 2. Speech recognition: The server uses speech recognition means to convert this voice data into text data.

[0894] 3. Emotion analysis: The emotion analysis engine analyzes the text and voice data and generates data representing the user's emotions. In this case, it is analyzed that the user is nervous.

[0895] 4. Intention analysis and text generation: The server uses the intention analysis means to analyze the user's intention and generate text such as "Maintenance work has begun."

[0896] 5. Voice feedback: This text is converted into voice data using a voice synthesis means and played back to the user, with the tone and speed of the voice adjusted appropriately based on the sentiment analysis data.

[0897] 6. Confirm and correct: When the user says "Tell me the steps," the system regenerates the text and plays it back.

[0898] 7. Save document: When the user instructs "Save as this content," the server saves the generated text in the final document format.

[0899] Specific examples

[0900] Example 1:

[0901] User: "I'm going to start maintenance on the machine."

[0902] System: "Maintenance work will begin. Please advise how to proceed."

[0903] Example 2:

[0904] User: "Tell me how to do this."

[0905] System: "The first step is to power down the machine."

[0906] Prompt Sentence Examples

[0907] "I'm going to start maintenance on the machine. Please tell me the procedure."

[0908] This system enables feedback that matches the user's emotions, improving work efficiency in factories and user satisfaction.

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

[0910] Step 1:

[0911] Audio Capture

[0912] The user dictates, "I will start maintenance work on the machine."

[0913] Input: User's voice

[0914] The terminal collects the user's voice in real time using a voice capture means, and the collected voice data is then prepared for processing by a voice recognition engine.

[0915] Output: Audio data

[0916] Step 2:

[0917] Voice Recognition

[0918] Input: Audio data

[0919] The server converts the collected voice data into text data using a voice recognition engine. The voice recognition engine analyzes the phonemes in the voice data and generates corresponding character strings. This process provides the user's voice instructions in text format.

[0920] Output: Text data

[0921] Step 3:

[0922] Emotion analysis

[0923] Input: Text data, audio data

[0924] The server uses emotion analysis means to analyze the obtained text data and voice data and recognize the user's emotions. The emotion analysis engine analyzes the tone, speed, pitch, etc. of the sound and generates emotion data such as whether the user is tense or relaxed.

[0925] Output: Emotion data

[0926] Step 4:

[0927] Intention Analysis

[0928] Input: Text data, emotion data

[0929] The server uses intent analysis to analyze the text data and understand the user's intent. The natural language processing engine analyzes keywords and contextual information in the text data and recognizes the "intent to start maintenance work." This lays the foundation for generating an appropriate response.

[0930] Output: Intent data

[0931] Step 5:

[0932] Text Generation

[0933] Input: Intention data, emotion data

[0934] The server uses a text generation means to generate appropriate text based on the analyzed user intent. Since emotional data is also taken into account, the generated text is in a style and tone that matches the user's emotions. A response such as "Maintenance work will begin. Please tell me the procedure" is generated.

[0935] Output: Response text

[0936] Step 6:

[0937] Voice generation

[0938] Input: Response text, emotion data

[0939] The server uses a speech synthesis means to convert the generated response text into speech data. The tone and speed of the speech are appropriately adjusted based on the emotional data obtained from the emotion analysis means. For example, if the user is nervous, the speech speed is slowed down so that the speech can be played in a calmer tone.

[0940] Output: Modified audio data

[0941] Step 7:

[0942] Audio Feedback

[0943] Input: Modified audio data

[0944] The terminal plays the generated voice data to the user, and the user can confirm the system's response through this voice feedback and give the next instruction.

[0945] Output: User's understanding and next instructions

[0946] Step 8:

[0947] Check and correct

[0948] Input: The user's next voice command

[0949] If the user says "Tell me how to do it," the system recaptures this instruction and repeats the process of speech recognition, intent analysis, text generation, and speech generation. The resulting text is then reformatted.

[0950] Output: Corrected text and audio data

[0951] Step 9:

[0952] Document Storage

[0953] Input: Final confirmed text data

[0954] When the user selects "Save this content," the server saves the generated and modified text in the final document format (e.g., PDF, Word), and the saved document is stored in a database or cloud storage.

[0955] Output: Saved document

[0956] This system enables feedback that matches the user's emotions, improving work efficiency in factories and user satisfaction.

[0957] 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.

[0958] 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.

[0959] 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.

[0960] [Third embodiment]

[0961] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0962] 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.

[0963] 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).

[0964] 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.

[0965] 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.

[0966] 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).

[0967] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0968] 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.

[0969] 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.

[0970] 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.

[0971] 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.

[0972] 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."

[0973] This invention relates to a system that allows desk workers to create, check, and correct documents by voice without using their eyes. This system generates text data based on the user's voice input and allows checking and correction through voice feedback. Each component and its operation are explained below.

[0974] System Configuration

[0975] This system consists of three main components: users, terminals, and servers.

[0976] Audio capture and preprocessing

[0977] The user gives voice instructions using a dedicated microphone or a microphone built into the device, and the device collects the user's voice in real time using a voice capture module, allowing the user to start working without needing an input device.

[0978] Once the voice is captured, the device converts the voice data into text using a speech recognition engine that is highly accurate and takes into account a wide range of pronunciations and accents.

[0979] Intention Analysis and Text Generation

[0980] The converted text data is sent to the server and analyzed by a natural language processing engine. The server analyzes this text data to recognize the user's intent. For example, if a user is given an instruction such as "Please write about sales analysis," the server analyzes the intent and creates an appropriate text generation request.

[0981] Based on the analyzed intent, the server uses chat generation AI to generate text according to the instructions. The generated text reflects industry-standard formatting and structure, and is tailored to the user's needs.

[0982] Audio feedback and confirmation / correction

[0983] The generated text is sent to the device, which converts it into voice data using a speech synthesis engine. The device then plays the converted voice data back to the user, prompting them to confirm the content. The user can then use the voice feedback to confirm the content of the document and, if necessary, provide voice instructions for corrections.

[0984] For example, if a user says, "Please change the expression 'year-on-year comparison' to 'the percentage increase from the same month last year,'" the device captures this speech, converts it back into text data, and sends it to the server. The server again analyzes the intent, corrects it as necessary, and regenerates the text. This process is repeated until the user is satisfied.

[0985] Save the final document

[0986] When the user makes a final confirmation and gives the instruction to "save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., Word, PDF) and saves it in a database or cloud storage. After the saving process is complete, the device will inform the user by voice that the save is complete.

[0987] Specific examples

[0988] Example 1:

[0989] Launch and Start: When the user says "Start creating a report," the device captures the voice, converts it into text using a speech recognition engine, and sends it to the server.

[0990] Intent analysis and text generation: The server analyzes the intent of "Start creating a report," generates the text "Start creating a report. What would you like as your first item?", and sends it to the device.

[0991] Confirm and correct: If the user instructs "Write about sales analysis," the same process is followed to generate the text "This month's sales increased by 5% compared to the same month last year." The user then requests corrections, and the final content is "This month's sales increased by 5% compared to the same month last year."

[0992] Example 2:

[0993] Save: When the user says "Save this content," the server saves it as the final document and the device notifies the user by voice that the save is complete.

[0994] In this way, the system of the present invention enables desk workers to create and edit documents without using their eyes, reducing eye strain and supporting healthy working styles.

[0995] The processing flow will be explained below.

[0996] Step 1:

[0997] A user logs into a dedicated application or device and dictates "Start report creation."

[0998] Step 2:

[0999] The device captures the user's voice through the microphone and stores the voice data in a buffer.

[1000] Step 3:

[1001] The voice data captured by the device is sent to a voice recognition library and converted into text data.

[1002] Step 4:

[1003] The voice recognition library converts the voice data into text data and sends it back to the device.

[1004] Step 5:

[1005] The terminal encodes the converted text data and sends it to the server as an HTTP request.

[1006] Step 6:

[1007] The server decodes the encoded text data and analyzes it using a natural language processing engine.

[1008] Step 7:

[1009] The server recognizes the user's intention based on the analysis results and sends a request to the chat generation AI to generate text based on that intention.

[1010] Step 8:

[1011] The chat generation AI generates appropriate text based on the instructions and sends it back to the server.

[1012] Step 9:

[1013] The server receives the generated text data and sends it to the terminal as an HTTP response.

[1014] Step 10:

[1015] The text data received by the terminal is sent to a speech synthesis engine and converted into voice data.

[1016] Step 11:

[1017] The speech synthesis engine converts the text into speech data and sends it back to the device.

[1018] Step 12:

[1019] The terminal plays back the generated voice data to the user, prompting them to confirm the content.

[1020] Step 13:

[1021] The user confirms the result through voice feedback and, if necessary, gives voice instructions for correction.

[1022] Step 14:

[1023] The device captures the correction instructions and converts them into text data using a voice recognition library again.

[1024] Step 15:

[1025] The terminal transmits the corrected text data to the server.

[1026] Step 16:

[1027] The server parses the modified text data and regenerates the text as necessary.

[1028] Step 17:

[1029] The server transmits the regenerated text data to the terminal, where it is converted back into voice data and played back to the user.

[1030] Step 18:

[1031] This process is repeated, with revisions and validations being made until the user is satisfied.

[1032] Step 19:

[1033] The user issues a voice command saying, "Please save this content."

[1034] Step 20:

[1035] The terminal sends a final instruction to the server.

[1036] Step 21:

[1037] The server converts the verified text into a final document format (e.g., Word, PDF) and stores it in a database or cloud storage.

[1038] Step 22:

[1039] The server sends a notification of completion of saving to the terminal.

[1040] Step 23:

[1041] The device will notify the user by voice when the save is complete.

[1042] This allows users to create, edit, and save documents without using their eyes, reducing physical strain.

[1043] Example 1

[1044] 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."

[1045] Modern desk workers are suffering from health problems such as eye strain and vision loss due to long hours of computer work. Creating, reviewing, and correcting documents in particular places a heavy visual burden on the eyes, reducing efficiency and causing health problems. Furthermore, traditional document creation methods are difficult for visually impaired users. A system that can solve these problems and achieve a more efficient and healthy work environment is needed.

[1046] 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.

[1047] In this invention, the server includes a means for capturing a user's voice, a means for converting the captured voice into text data, and a means for analyzing the user's intent based on the converted text data. This allows the user to create, review, and modify documents using only voice instructions. Furthermore, by providing a means for creating documents using a generative AI model and a means for saving the final formatted document in data storage, efficient and accurate document creation and storage can be achieved. This reduces visual strain and supports healthy working styles.

[1048] A "means for capturing audio" is a device or system for capturing audio uttered by a user and processing it as a digital signal.

[1049] The "means for converting into text data" is a technique for analyzing the captured audio data and converting it into a corresponding text format.

[1050] "Means for analyzing user intent" refers to natural language processing technology used to understand the converted text data and identify what the user is looking for.

[1051] "Means for generating documents" refers to technology that uses a generative AI model to create text according to the specified content based on the user's intentions.

[1052] The "means for converting the text into audio data and presenting it to the user" refers to a technique for converting the generated or corrected text into audio and providing it to the user as an anti-visual interface.

[1053] The "means for correcting and regenerating a document according to confirmation and correction instructions" is a processing system for correcting the contents of a document in response to feedback from a user and regenerating the document as necessary.

[1054] "Means for saving in final format" refers to the technology for converting the completed document into a specified file format and saving it in a database, cloud storage, etc.

[1055] A "data analysis module" is a program or function that analyzes data based on the user's voice instructions and sends appropriate instructions to the generative AI model.

[1056] A "generative AI model" is an artificial intelligence technology that generates text based on a given prompt.

[1057] "Data storage" refers to a system or device for storing and managing data over the long term.

[1058] This invention relates to a system that allows desk workers to create, check, and correct documents by voice without using their eyes. This system generates text data based on the user's voice input and allows checking and correction through voice feedback. Each component and its operation are explained below.

[1059] System Configuration

[1060] This system consists of three main components: users, terminals, and servers.

[1061] Audio capture and preprocessing

[1062] The user gives voice instructions using a dedicated microphone or a microphone built into the device, and the device collects the user's voice in real time using a voice capture module, allowing the user to start working without needing an input device.

[1063] Voice Recognition

[1064] Once the audio is captured, the device converts it into text using a speech recognition engine (e.g., Google Cloud Speech-to-Text) that has high accuracy and takes into account various pronunciations and accents.

[1065] Intention Analysis and Text Generation

[1066] The converted text data is sent to the server and analyzed by a natural language processing engine (e.g., spaCy). The server analyzes this text data to recognize the user's intent. For example, if an instruction such as "Please write about sales analysis" is given, the server analyzes the intent and creates an appropriate text generation request.

[1067] Based on the parsed intent, the server uses a generative AI model (e.g., OpenAI GPT-3) to generate text based on the instructions, reflecting industry-standard formatting and structure and tailored to the user's needs.

[1068] Audio feedback and confirmation / correction

[1069] The generated text is sent to the device, which converts it into voice data using a speech synthesis engine (e.g., Amazon Polly). The device then plays the converted voice data back to the user, prompting them to confirm the content. The user can then confirm the content of the document through voice feedback and, if necessary, provide voice instructions for correction.

[1070] For example, if a user says, "Please change the expression 'compared to the same month last year' to 'the percentage increase from the same month last year,'" the device captures this speech, converts it into text data using a speech recognition engine, and sends it to the server. The server again analyzes the intent, corrects the text as necessary, and regenerates it. This process is repeated until the user is satisfied.

[1071] Save the final document

[1072] When the user makes a final confirmation and gives the instruction "Save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., Word, PDF) and saves it in a database or cloud storage (e.g., Google Drive, Dropbox). After the saving process is complete, the device will inform the user by voice that the save is complete.

[1073] Specific examples

[1074] Below is a concrete example of how this system can be used.

[1075] Example 1:

[1076] Boot and start:

[1077] When a user says "start report creation," the device captures the voice, converts it into text using a speech recognition engine, and sends it to the server.

[1078] The server analyzes the intent of "Start creating a report," generates the text "Start creating a report. What would you like as your first item?", and sends it to the terminal.

[1079] The user gives the voice instruction "Write about your sales analysis."

[1080] The generative AI model generates the text, "Sales this month increased by 5% compared to the same month last year."

[1081] The user further instructs, "Please change the year-on-year comparison to the increase rate from the same month last year."

[1082] The device captures the audio again and sends correction instructions to the server.

[1083] The server generates the text "Sales this month increased 5% from the same month last year" reflecting the correction.

[1084] Example 2:

[1085] keep:

[1086] When the user says "Save this content," the device captures the audio, converts it into text, and sends it to the server.

[1087] The server saves it as the final document, and the device notifies you with an audio message that the save is complete.

[1088] As described above, the system of the present invention allows desk workers to create, check, and edit documents using only voice commands, thereby reducing visual strain and supporting healthy working styles.

[1089] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1090] Step 1: Capture audio

[1091] Specific description:

[1092] Input: The user gives a voice command such as "Start report creation."

[1093] Specific operation: The user's voice is captured using a dedicated microphone or the device's built-in microphone.

[1094] Data processing: The terminal collects audio data using an audio capture module.

[1095] Output: The captured voice data is passed to the speech recognition engine.

[1096] Step 2: Voice Recognition

[1097] Specific description:

[1098] Input: Captured audio data.

[1099] Specific operation: The terminal converts voice data into text data using a voice recognition engine (e.g., Google Cloud Speech-to-Text).

[1100] Data processing: The speech recognition engine analyzes the speech waveform and converts it into a corresponding string of characters.

[1101] Output: The converted text data is generated.

[1102] Step 3: Intent Analysis

[1103] Specific description:

[1104] Input: Text data output from the speech recognition engine.

[1105] Specific operation: The server receives the text data sent from the terminal and analyzes it using a natural language processing engine (e.g., spaCy).

[1106] Data processing: Natural language processing engines analyze text data to identify user intent.

[1107] Output: Analysis results that reflect the user's intentions are generated.

[1108] Step 4: Text Generation

[1109] Specific description:

[1110] Input: Analysis results obtained from intent analysis.

[1111] Specific operation: The server uses a generative AI model (e.g., OpenAI GPT-3) to generate text based on the user's intent.

[1112] Data processing: The generative AI model generates relevant text based on the input prompt.

[1113] Output: Generated text data is generated.

[1114] Step 5: Audio feedback

[1115] Specific description:

[1116] Input: Generated text data.

[1117] Specific operation: The device converts the generated text data into audio data using a speech synthesis engine (e.g., Amazon Polly) and plays it back.

[1118] Data processing: The speech synthesis engine converts text data into speech waveforms.

[1119] Output: Audio data is generated that is played back to the user.

[1120] Step 6: Accepting correction instructions

[1121] Specific description:

[1122] Input: User's spoken correction instructions (e.g., "Please change the expression 'year-over-year comparison' to 'percent increase from the same month last year'").

[1123] Specific operation: The device again captures the user's voice correction instructions and converts them into text data using a voice recognition engine.

[1124] Data processing: The new voice command is converted into text and sent to the server.

[1125] Output: Text data containing correction instructions is passed to the server.

[1126] Step 7: Fix and Regenerate

[1127] Specific description:

[1128] Input: Text data containing correction instructions.

[1129] Specific behavior: The server parses the correction instructions, corrects the text as necessary, and regenerates it.

[1130] Data processing: The data analysis module analyzes the correction instructions and sends new prompts to the generative AI model to regenerate the text.

[1131] Output: The corrected text data is generated.

[1132] Step 8: Save the final document

[1133] Specific description:

[1134] Input: Final text data after correction (e.g., "Sales this month increased 5% from the same month last year.").

[1135] Specific operation: The server converts the final text data into a document format (e.g., Word, PDF) and saves it in data storage (e.g., Google Drive, Dropbox).

[1136] Data processing: The text data is converted into the specified file format and uploaded to cloud storage.

[1137] Output: A sound is generated to notify the user that the save is complete.

[1138] (Application example 1)

[1139] 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."

[1140] The objective of this invention is to provide a system that reduces the burden of deskwork in a factory environment and allows users to create, check, modify, and finally save reports using only voice commands. This system allows users to operate it by voice without using their eyes, improving worker health and increasing productivity. Furthermore, in a voice-only work environment, rapid and accurate data processing is required.

[1141] 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.

[1142] In this invention, the server includes means for capturing a user's voice, means for converting the captured voice into text data, means for analyzing the user's intent based on the converted text data, means for generating appropriate text based on the user's intent, means for converting the generated text into voice data and presenting it to the user, means for correcting and regenerating the text according to the user's confirmation and correction instructions, means for saving the generated and corrected text in a final document format, and robot means for realizing use in a voice-specific work environment, a factory environment where reports are created by voice instructions. This allows workers to create documents using only voice instructions, reducing eye fatigue and improving work efficiency.

[1143] "User" refers to the person who operates the system and gives voice instructions.

[1144] "Audio capturing means" refers to a device or method that captures the audio produced by a user as electronic data.

[1145] "Means for converting voice into text data" refers to technology that analyzes captured voice data and converts it into corresponding text data.

[1146] "Means of intent analysis" refers to technology that understands what a user wants from text data and identifies appropriate actions.

[1147] "Means for generating appropriate text" refers to technology that automatically creates the necessary information and sentences based on the user's intentions.

[1148] The "means for converting the generated text into audio data and presenting it to the user" refers to a device or method for converting the generated text back into audio data and allowing the user to listen to it.

[1149] "Means for correcting and regenerating text" refers to a technique for revising existing text and generating new text according to user instructions.

[1150] "Means for saving in final document format" refers to a technique for saving the final verified text in a specified format.

[1151] A "voice-focused work environment" refers to a workplace where operations and instructions are primarily performed by voice.

[1152] "Use in a factory environment where reports are created using voice instructions" refers to performing a series of tasks such as creating, modifying, and saving reports using only voice within the factory.

[1153] "Robotic means" refers to automated equipment or systems that implement voice commands in a factory environment.

[1154] This invention is a system for efficiently creating reports using voice instructions in a factory environment. The system generates text data based on the user's voice input and allows confirmation and correction through voice feedback. The specific configuration and processing procedure for realizing this system will be described below.

[1155] System Configuration

[1156] The system consists of four main components: users, terminals, servers, and robots.

[1157] 1. Audio capture and preprocessing

[1158] The user gives voice instructions using a dedicated microphone or the device's built-in microphone. The device collects the user's voice in real time using a voice capture module. The collected voice data is converted into text data using a voice recognition engine (e.g., Google Speech-to-Text API).

[1159] 2. Intention Analysis and Text Generation

[1160] The converted text data is sent to a server and analyzed by a natural language processing engine (e.g., OpenAI GPT-4). The server analyzes this text data to recognize the user's intent. For example, if an instruction such as "Please write about sales analysis" is given, the server analyzes the intent and creates an appropriate text generation request. Based on the analyzed intent, the server uses a generative AI model to generate text according to the instruction.

[1161] 3. Audio feedback and confirmation / correction

[1162] The generated text is sent to the device, which converts it into audio data using a speech synthesis engine (e.g., Google Text-to-Speech API). The device then plays the converted audio data back to the user, prompting them to confirm the content. The user can then use the audio feedback to confirm the content of the document and, if necessary, provide voice instructions for corrections.

[1163] 4. Preservation of the final document

[1164] When the user makes a final confirmation and gives the instruction to "save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., PDF) and saves it in a database or cloud storage (e.g., AWS S3). After the saving process is complete, the device will inform the user by voice that the save is complete.

[1165] Specific examples

[1166] Example 1: Starting a report

[1167] User: Says "Start creating a report."

[1168] The device collects the voice using a voice capture module and converts it into text data such as "Start creating report" using a voice recognition engine.

[1169] The server receives this text, generates the text "Let's start creating a report. What would you like as the first item?" and sends it to the terminal.

[1170] The terminal uses a speech synthesis engine to convert this text into audio data and play it back to the user.

[1171] Example 2: Sales analysis instructions and corrections

[1172] User: "Write about your sales analysis."

[1173] The terminal collects voice using a voice capture module and converts it into text data using a voice recognition engine.

[1174] The server analyzes this text data, recognizes the intent of "Please write about sales analysis," and generates the text "This month's sales increased by 5% compared to the same month last year."

[1175] The terminal converts the generated text into voice data using a voice synthesis engine and plays it back to the user.

[1176] When the user instructs, "Please change it to the increase rate compared to the same month last year," the device captures this voice, converts it back into text data, and sends it to the server.

[1177] The server again parses the intent and generates the corrected text, "Sales this month increased 5% from the same month last year."

[1178] The terminal converts the corrected text into voice data using a voice synthesis engine and plays it back to the user.

[1179] Prompt Sentence Examples

[1180] "Please write a report on your sales analysis. Please include specific figures and year-on-year increases or decreases."

[1181] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1182] Step 1: User initiates voice command

[1183] The user gives voice instructions (such as "Start creating a report") using a dedicated microphone or the device's built-in microphone. The device collects voice data in real time using a voice capture module. At this time, the user's voice is input, and the captured voice data is output.

[1184] Step 2: Convert the audio data to text

[1185] The device converts the collected voice data into text data using a voice recognition engine (e.g., Google Speech-to-Text API). Specifically, the captured voice data is input and passed through an advanced voice analysis algorithm to generate corresponding text data. This text data is then output.

[1186] Step 3: Analyze the text data

[1187] The server receives the converted text data and analyzes the user's intent through a natural language processing engine (e.g., OpenAI GPT-4). For example, it interprets the purpose of an instruction such as "start creating a report." In this case, the input is text data, and the output is the analysis result (data indicating the user's intent).

[1188] Step 4: Generate the appropriate text

[1189] The server uses a generative AI model based on the results of intent analysis to generate the required text. Based on the converted text data, a response text such as "I'm starting to write a report. What would you like as the first item?" is generated. The input is the analysis result, and the output is the generated text data.

[1190] Step 5: Convert the generated text to audio data

[1191] The device receives the generated text data and converts it into voice data using a speech synthesis engine (e.g., Google Text-to-Speech API). Specifically, it takes the generated text data as input and outputs it as voice waveform data. This voice data is then played back to the user.

[1192] Step 6: Receive user confirmation and correction instructions

[1193] The user can check the generated text through voice feedback and make corrections as necessary. For example, they can say, "Please change it to the increase rate compared to the same month last year." This voice instruction is input, captured by the device, and converted back into text data through the voice recognition engine (output).

[1194] Step 7: Regenerate the text based on the correction instructions

[1195] The server receives the text data again, analyzes the intent, and makes the necessary corrections. For example, based on the instruction "Please change this to the increase rate compared to the same month last year," the corrected text "This month's sales increased by 5% compared to the same month last year" is generated. The input is the text data of the correction instruction, and the output is the corrected text data.

[1196] Step 8: Convert the corrected text back into audio data and present it

[1197] The corrected text data is again converted into voice data by the terminal using a speech synthesis engine. It is then played back to the user for reconfirmation. This process is repeated until the user is satisfied. The input is the corrected text data, and the output is the corrected voice data.

[1198] Step 9: Save in the final document format

[1199] Finally, when the user instructs "Save this content," the server converts the final confirmed text data into a specified document format (e.g., PDF) and saves it in cloud storage (e.g., AWS S3). The input is the final confirmed text data, and the output is the saved document. The device notifies the user by voice when saving is complete.

[1200] 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.

[1201] This invention relates to a system that allows users to generate text through voice input and confirm and correct it by voice. A feature of this invention is that by incorporating an emotion engine that recognizes the user's emotions, the generated text and voice feedback are controlled to match the user's emotions. This improves the user experience and supports healthier working styles.

[1202] System Configuration

[1203] This system consists of four main components: the user, the terminal, the server, and the emotion engine.

[1204] Audio capture and preprocessing

[1205] The user gives voice instructions using a dedicated microphone or a microphone built into the device. The device collects the user's voice in real time using a voice capture module. This voice data is sent to a voice recognition engine and converted into text data.

[1206] emotion recognition

[1207] The converted text data and voice data are sent by the terminal to the emotion engine, which analyzes the user's emotions from the voice and text and generates emotion data. This emotion data is used for subsequent text generation and voice synthesis.

[1208] Intention Analysis and Text Generation

[1209] The server receives the text data and emotion data and analyzes the text data using a natural language processing engine. The server recognizes the user's intention based on the analysis results, and sends a request to the chat generation AI to generate text based on that intention, taking into account the emotion data. The generated text is expected to have a style and tone that matches the user's emotion.

[1210] Audio feedback and confirmation / correction

[1211] The generated text is sent to the device and converted into voice data by a speech synthesis engine. At this time, the tone and speed of the voice are adjusted based on the emotional data obtained from the emotion engine. The device then plays the adjusted voice data back to the user, prompting them to confirm the content. The user can check the content of the document through voice feedback and, if necessary, give voice instructions for corrections.

[1212] For example, if a user instructs the server to "Please prepare a sales report for the previous month," the server analyzes the instruction and generates text such as "Last month's sales increased by 10% compared to the plan." If the emotion engine recognizes that the user is nervous, the speech synthesis engine generates speech in a slightly slower, calmer tone.

[1213] Save the final document

[1214] When the user makes a final confirmation and gives the instruction to "save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., Word, PDF) and saves it in a database or cloud storage. After the saving process is complete, the device will inform the user by voice that the save is complete.

[1215] Specific examples

[1216] Example 1:

[1217] Launch and Start: The user says "Start report creation." The device captures the voice and converts it into text data using the speech recognition engine. At the same time, the emotion engine analyzes the user's current emotion and generates emotion data.

[1218] Intention analysis and text generation: The server analyzes the intention of "Start creating a report" and generates the text "Start creating a report. What would you like to do first?" taking into account the user's relaxation state.

[1219] Confirmation and correction: If the user instructs "Please write about sales analysis," the text generated reflects that state of mind: "This month's sales increased by 5% compared to the same month last year." If the user requests a correction, such as "Please change the expression "compared to the same month last year" to "the percentage increase from the same month last year," the text is corrected based on that request.

[1220] Example 2:

[1221] Save: When the user says "Save this content," the server saves it as the final document and the device notifies the user by voice that the save is complete.

[1222] In this way, the system of the present invention allows document creation and editing to be performed while taking the user's emotions into consideration, improving the user experience while reducing eye strain.

[1223] The processing flow will be explained below.

[1224] Step 1:

[1225] A user logs into a dedicated application or device and dictates "Start report creation."

[1226] Step 2:

[1227] The device captures the user's voice through the microphone and stores the voice data in a buffer.

[1228] Step 3:

[1229] The voice data captured by the device is sent to a voice recognition library and converted into text data.

[1230] Step 4:

[1231] The voice recognition library converts the voice data into text data and sends it back to the device.

[1232] Step 5:

[1233] The terminal transmits the converted text data to an emotion engine, which analyzes the user's emotions.

[1234] Step 6:

[1235] The emotion engine analyzes the voice and text data, generates the user's emotion data, and sends it back to the terminal.

[1236] Step 7:

[1237] The device encodes the text data and emotion data and sends them to the server as an HTTP request.

[1238] Step 8:

[1239] The server decodes the encoded text data and emotion data and analyzes the text data using a natural language processing engine.

[1240] Step 9:

[1241] The server recognizes the user's intention based on the analysis results, takes emotional data into consideration, and sends a request for text generation to the chat generation AI.

[1242] Step 10:

[1243] The chat generation AI generates appropriate text based on the intent and emotion data and sends it back to the server.

[1244] Step 11:

[1245] The server receives the generated text data and sends it to the terminal as an HTTP response.

[1246] Step 12:

[1247] The text data received by the terminal is sent to a speech synthesis engine and converted into voice data based on emotional data.

[1248] Step 13:

[1249] The speech synthesis engine takes into account the emotional data, converts the text into speech data, and sends it back to the device.

[1250] Step 14:

[1251] The terminal plays back the generated voice data to the user, prompting them to confirm the content.

[1252] Step 15:

[1253] The user confirms the result through voice feedback and, if necessary, gives voice instructions for correction.

[1254] Step 16:

[1255] The device captures the correction instructions and converts them into text data using a voice recognition library again.

[1256] Step 17:

[1257] The terminal transmits the corrected text data and emotion data to the server.

[1258] Step 18:

[1259] The server parses the modified text data and regenerates the text as necessary.

[1260] Step 19:

[1261] The server transmits the regenerated text data to the terminal, where it is converted back into voice data and played back to the user.

[1262] Step 20:

[1263] This process is repeated, with revisions and validations being made until the user is satisfied.

[1264] Step 20:

[1265] The user issues a voice command saying, "Please save this content."

[1266] Step 22:

[1267] The terminal sends a final instruction to the server.

[1268] Step 23:

[1269] The server converts the verified text into a final document format (e.g., Word, PDF) and stores it in a database or cloud storage.

[1270] Step 24:

[1271] The server sends a notification of completion of saving to the terminal.

[1272] Step 25:

[1273] The device will notify the user by voice when the save is complete.

[1274] This allows users to create, edit and save documents without using their eyes, and when combined with an emotion engine, provides more relevant and emotionally tailored feedback.

[1275] Example 2

[1276] 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."

[1277] Conventional document creation systems fail to consider the user's emotions when generating text using voice instructions and then reviewing and correcting it. This prevents the system from adjusting text and voice feedback based on the user's emotional state, resulting in a lack of user experience. This also makes it difficult to adjust the tempo and mood while creating documents, hindering efficient and healthy work practices. To address these issues, a system is needed that can recognize the user's emotions in real time and generate text and provide voice feedback that adapts to those emotions.

[1278] 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.

[1279] In this invention, the server includes means for capturing a user's voice, means for converting the captured voice into text data, means for analyzing the user's emotions based on the converted text data and voice data and generating emotion data, means for analyzing the user's intention based on the emotion data and text data, means for generating appropriate text based on the user's intention and emotion data, means for converting the generated text into voice data, adjusting the tone and speed of the voice based on the emotion data and presenting it to the user, means for correcting and regenerating the text in accordance with the user's confirmation and correction instructions, and means for saving the generated and corrected text in a final document format. This enables text generation and voice feedback adapted to the user's emotions, improving the user experience and realizing an efficient and healthy way of working.

[1280] An "audio capture means" is a device or software for collecting a user's voice in real time.

[1281] "Speech recognition means" refers to a device or software for converting collected voice data into text data.

[1282] The "emotion recognition means" is a device or software that analyzes the user's emotions based on the converted text data and voice data, and generates emotion data.

[1283] The "intention analysis means" is a device or software that analyzes the user's intention based on text data and emotion data.

[1284] The "text generation means" is a device or software that generates appropriate text based on the user's intention and emotion data.

[1285] A "speech synthesis means" is a device or software that converts the generated text into speech data and adjusts the tone and speed of the speech based on emotional data.

[1286] A "verification and correction means" is a device or software that corrects and regenerates text according to a user's verification and correction instructions.

[1287] A "data storage means" is a device or software that stores the generated and modified text in its final document format.

[1288] This system allows users to generate text through voice input and confirm and correct it by voice. One of the features of this system is that it incorporates an emotion engine that recognizes the user's emotions, and controls the generated text and voice feedback to match the user's emotions. This improves the user experience and supports more efficient and healthy working styles.

[1289] This system consists of four main components: the user, the terminal, the server, and the emotion engine.

[1290] Audio capture and preprocessing

[1291] The user gives voice instructions using a dedicated microphone or the device's built-in microphone. The device collects the user's voice in real time using a voice capture module. Specifically, Python's speech_recognition library can be used. This voice data is sent to a speech recognition engine (for example, Google Cloud Speech-to-Text API) and converted into text data.

[1292] emotion recognition

[1293] The device sends the converted text data and voice data to an emotion engine (e.g., IBM Watson Tone Analyzer). The emotion engine analyzes the user's emotions from the voice and text and generates emotion data. This emotion data is used for subsequent text generation and voice synthesis.

[1294] Intention Analysis and Text Generation

[1295] The server receives the emotion data and text data and analyzes the text data using a natural language processing engine (e.g., OpenAI's GPT-3). The server recognizes the user's intent based on the analysis results and sends a request to a chat generation AI (e.g., Microsoft Azure Chatbot) to generate text based on that intent, taking the emotion data into consideration. The generated text is expected to have a style and tone that matches the user's emotions.

[1296] Audio feedback and confirmation / correction

[1297] The generated text is sent to the device and converted into voice data by a speech synthesis engine (e.g., Amazon Polly). At this time, the tone and speed of the voice are adjusted based on the emotional data obtained from the emotion engine. The device then plays the adjusted voice data back to the user, prompting them to confirm the content. The user can check the content of the document through voice feedback and, if necessary, give voice instructions for corrections.

[1298] For example, if a user instructs the server to "Please prepare a sales report for the previous month," the server analyzes the instruction and generates text such as "Last month's sales increased by 10% compared to the plan." If the emotion engine recognizes that the user is nervous, the speech synthesis engine generates a voice with a slightly slower and calmer tone.

[1299] Save the final document

[1300] When the user makes a final confirmation and gives the instruction to "save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., Word, PDF) and saves it in a database or cloud storage. After the saving process is complete, the device notifies the user by voice that the save is complete.

[1301] Specific examples

[1302] Example 1:

[1303] Launch and start: The user says "Start report creation." The device captures the voice and converts it into text data using the speech recognition engine. At the same time, the emotion engine analyzes the user's current emotion and generates emotion data.

[1304] Intention analysis and text generation: The server analyzes the intention of "Start creating a report" and generates the text "Start creating a report. What would you like to do first?" taking into account the user's relaxation state.

[1305] Confirmation and correction: If the user instructs "Write about sales analysis," the text generated reflects that state of mind: "This month's sales increased by 5% compared to the same month last year." If the user requests a correction, such as "Please change the expression "compared to the same month last year" to "the percentage increase from the same month last year," the text is corrected based on that request.

[1306] Example 2:

[1307] Save: When the user says "Save this content," the server saves it as the final document and the device gives a voice notification that the save is complete.

[1308] In this way, the system of the present invention enables document creation and editing to be performed while taking the user's emotions into consideration, thereby improving the user experience and supporting efficient work methods.

[1309] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1310] Step 1:

[1311] The device uses a dedicated or built-in microphone to capture the user's voice. The user's voice instructions are provided as input. The device uses the Python speech_recognition library to collect voice data in real time. The collected voice data is sent to a speech recognition engine (e.g., Google Cloud Speech-to-Text API) which generates text data as output. This text data becomes the input for the next processing step.

[1312] Step 2:

[1313] The device sends the text data and voice data converted in step 1 to an emotion engine (for example, IBM Watson Tone Analyzer). The text data and voice data are provided as input. The emotion recognition engine analyzes this data and analyzes the user's emotions. Emotional data is generated as output, and this emotion data becomes the input to the next processing step. Specifically, the user's emotional state is identified by analyzing the choice of words read from the text data and the intonation of the voice data.

[1314] Step 3:

[1315] The server receives the emotion data generated in step 2 and the text data generated in step 1. Using these data as input, it analyzes the text data using a natural language processing engine (e.g., OpenAI's GPT-3). Based on the analysis results, it recognizes the user's intention and, taking the emotion data into consideration, sends a request to generate text to a chat generation AI (e.g., Microsoft Azure Chatbot). The generated text has a style and tone that matches the user's emotion. The appropriate generated text is obtained as output, and becomes the input for the next processing step.

[1316] Step 4:

[1317] The device receives the text data generated in step 3 and sends it to a speech synthesis engine (for example, Amazon Polly). The generated text data and emotion data are provided as input. The speech synthesis engine generates voice data while adjusting the tone and speed of the voice based on the emotion data. The generated voice data is obtained as output, and this is played back to the user. Specifically, the voice data is played back as feedback to the user, and the user confirms the content.

[1318] Step 5:

[1319] The user checks the content of the document based on the voice feedback played in step 4 and, if necessary, gives voice instructions for correction. The checked content and correction instructions are provided as input. The device uses a voice capture module to collect the user's correction instructions in real time and converts them into text data using a voice recognition engine again. The corrected text data is obtained as output and is sent again to the server for repeated processing.

[1320] Step 6:

[1321] When the user makes a final confirmation and issues the instruction "Save this content," the device sends this instruction to the server. The final confirmed text is provided as input. The server converts the confirmed text data into the final document format (e.g., PDF, Word, etc.). The specific output is a file in the converted document format, which is stored in a database or cloud storage. Finally, the device notifies the user by voice that the save is complete.

[1322] (Application example 2)

[1323] 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."

[1324] Conventional voice input systems generate text and provide voice feedback without considering the user's emotions, resulting in a flawed user experience. Furthermore, robots operating in factories lack the functionality to provide appropriate feedback based on emotions, which is insufficient for improving work efficiency and user satisfaction. To solve these problems, a system that can properly analyze the user's emotional state and reflect it in feedback is needed.

[1325] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1326] In this invention, the server includes means for capturing a user's voice, means for converting the captured voice into text data, means for analyzing the user's intention based on the converted text data, means for generating appropriate text based on the user's intention, means for converting the generated text into voice data and presenting it to the user, means for correcting and regenerating the text in accordance with the user's confirmation and correction instructions, means for saving the generated and corrected text in a final document format, means for capturing voice instructions and analyzing emotions, and means for adjusting the tone and speed of the voice data based on the emotion analysis results. This enables feedback that matches the user's emotions, thereby improving work efficiency in the factory and user satisfaction.

[1327] "Audio capture means" refers to devices or software for collecting a user's voice in real time.

[1328] "Speech recognition means" refers to the algorithms or software that convert collected voice data into text data.

[1329] "Intention analysis means" refers to algorithms or software that analyzes the converted text data and understands what the user intended.

[1330] "Text generation means" refers to algorithms or software that generate appropriate text based on the analyzed user intent.

[1331] "Speech synthesis means" refers to the algorithm or software used to convert the generated text into speech data.

[1332] "Text correction tool" refers to algorithms or software for correcting and regenerating text according to user confirmation and correction instructions.

[1333] "Document Storage" refers to the device or software used to store generated and modified text in its final document format.

[1334] "Emotion analysis means" refers to algorithms or software that analyze emotions from a user's voice or text and generate the results as data.

[1335] "Audio tone adjustment means" refers to algorithms or software for adjusting the tone or rate of audio data based on the results of emotion analysis.

[1336] The present invention aims to provide a factory robot management system that performs tasks based on voice instructions and provides feedback according to the user's emotions. The overall configuration of the system and specific embodiments thereof will be described below.

[1337] System Configuration

[1338] The system consists of the following major components:

[1339] 1. Users

[1340] 2. Terminal

[1341] 3. Server

[1342] 4. Sentiment Analysis Engine

[1343] User

[1344] The user gives voice instructions using a dedicated microphone or a microphone built into the device, providing work instructions to the factory robot in real time.

[1345] Terminal

[1346] The terminal collects the user's voice in real time using a voice capture means and transmits the voice data to the server. The terminal has the following functions.

[1347] Audio capture means: collects the user's voice in real time.

[1348] Speech recognition means: Converts collected voice data into text data.

[1349] Speech synthesis means: Converts the generated text into speech data and presents it to the user.

[1350] server

[1351] The server analyzes the user's intent and generates or modifies text and voice data using the following methods:

[1352] Intention analysis means: Analyze the converted text data to understand the user's intention.

[1353] Text generation means: Generates appropriate text based on the analyzed user intent.

[1354] Text correction means: Corrects and regenerates the text according to the user's confirmation and correction instructions.

[1355] Document storage means: storing the generated and modified text in the final document format.

[1356] Sentiment Analysis Engine

[1357] The emotion analysis engine analyzes emotions from the user's voice and text and generates the results as data. This emotion data is used as follows:

[1358] Sentiment analysis means: Analyzes emotions from the user's voice and text.

[1359] Voice tone adjustment means: Adjusts the tone and speed of voice data based on the results of emotion analysis.

[1360] System Operation

[1361] 1. Voice capture: The user dictates, "I will start the maintenance work on the machine." The device captures the voice and sends it to the server as voice data.

[1362] 2. Speech recognition: The server uses speech recognition means to convert this voice data into text data.

[1363] 3. Emotion analysis: The emotion analysis engine analyzes the text and voice data and generates data representing the user's emotions. In this case, it is analyzed that the user is nervous.

[1364] 4. Intention analysis and text generation: The server uses the intention analysis means to analyze the user's intention and generate text such as "Maintenance work has begun."

[1365] 5. Voice feedback: This text is converted into voice data using a voice synthesis means and played back to the user, with the tone and speed of the voice adjusted appropriately based on the sentiment analysis data.

[1366] 6. Confirm and correct: When the user says "Tell me the steps," the system regenerates the text and plays it back.

[1367] 7. Save document: When the user instructs "Save as this content," the server saves the generated text in the final document format.

[1368] Specific examples

[1369] Example 1:

[1370] User: "I'm going to start maintenance on the machine."

[1371] System: "Maintenance work will begin. Please advise how to proceed."

[1372] Example 2:

[1373] User: "Tell me how to do this."

[1374] System: "The first step is to power down the machine."

[1375] Prompt Sentence Examples

[1376] "I'm going to start maintenance on the machine. Please tell me the procedure."

[1377] This system enables feedback that matches the user's emotions, improving work efficiency in factories and user satisfaction.

[1378] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1379] Step 1:

[1380] Audio Capture

[1381] The user dictates, "I will start maintenance work on the machine."

[1382] Input: User's voice

[1383] The terminal collects the user's voice in real time using a voice capture means, and the collected voice data is then prepared for processing by a voice recognition engine.

[1384] Output: Audio data

[1385] Step 2:

[1386] Voice Recognition

[1387] Input: Audio data

[1388] The server converts the collected voice data into text data using a voice recognition engine. The voice recognition engine analyzes the phonemes in the voice data and generates corresponding character strings. This process provides the user's voice instructions in text format.

[1389] Output: Text data

[1390] Step 3:

[1391] Emotion analysis

[1392] Input: Text data, audio data

[1393] The server uses emotion analysis means to analyze the obtained text data and voice data and recognize the user's emotions. The emotion analysis engine analyzes the tone, speed, pitch, etc. of the sound and generates emotion data such as whether the user is tense or relaxed.

[1394] Output: Emotion data

[1395] Step 4:

[1396] Intention Analysis

[1397] Input: Text data, emotion data

[1398] The server uses intent analysis to analyze the text data and understand the user's intent. The natural language processing engine analyzes keywords and contextual information in the text data and recognizes the "intent to start maintenance work." This lays the foundation for generating an appropriate response.

[1399] Output: Intent data

[1400] Step 5:

[1401] Text Generation

[1402] Input: Intention data, emotion data

[1403] The server uses a text generation means to generate appropriate text based on the analyzed user intent. Since emotional data is also taken into account, the generated text is in a style and tone that matches the user's emotions. A response such as "Maintenance work will begin. Please tell me the procedure" is generated.

[1404] Output: Response text

[1405] Step 6:

[1406] Voice generation

[1407] Input: Response text, emotion data

[1408] The server uses a speech synthesis means to convert the generated response text into speech data. The tone and speed of the speech are appropriately adjusted based on the emotional data obtained from the emotion analysis means. For example, if the user is nervous, the speech speed is slowed down so that the speech can be played in a calmer tone.

[1409] Output: Modified audio data

[1410] Step 7:

[1411] Audio Feedback

[1412] Input: Modified audio data

[1413] The terminal plays the generated voice data to the user, and the user can confirm the system's response through this voice feedback and give the next instruction.

[1414] Output: User's understanding and next instructions

[1415] Step 8:

[1416] Check and correct

[1417] Input: The user's next voice command

[1418] If the user says "Tell me how to do it," the system recaptures this instruction and repeats the process of speech recognition, intent analysis, text generation, and speech generation. The resulting text is then reformatted.

[1419] Output: Corrected text and audio data

[1420] Step 9:

[1421] Document Storage

[1422] Input: Final confirmed text data

[1423] When the user selects "Save this content," the server saves the generated and modified text in the final document format (e.g., PDF, Word), and the saved document is stored in a database or cloud storage.

[1424] Output: Saved document

[1425] This system enables feedback that matches the user's emotions, improving work efficiency in factories and user satisfaction.

[1426] 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.

[1427] 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.

[1428] 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.

[1429] [Fourth embodiment]

[1430] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1431] 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.

[1432] 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).

[1433] 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.

[1434] 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.

[1435] 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).

[1436] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1437] 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.

[1438] 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.

[1439] 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.

[1440] 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.

[1441] 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.

[1442] 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."

[1443] This invention relates to a system that allows desk workers to create, check, and correct documents by voice without using their eyes. This system generates text data based on the user's voice input and allows checking and correction through voice feedback. Each component and its operation are explained below.

[1444] System Configuration

[1445] This system consists of three main components: users, terminals, and servers.

[1446] Audio capture and preprocessing

[1447] The user gives voice instructions using a dedicated microphone or a microphone built into the device, and the device collects the user's voice in real time using a voice capture module, allowing the user to start working without needing an input device.

[1448] Once the voice is captured, the device converts the voice data into text using a speech recognition engine that is highly accurate and takes into account a wide range of pronunciations and accents.

[1449] Intention Analysis and Text Generation

[1450] The converted text data is sent to the server and analyzed by a natural language processing engine. The server analyzes this text data to recognize the user's intent. For example, if a user is given an instruction such as "Please write about sales analysis," the server analyzes the intent and creates an appropriate text generation request.

[1451] Based on the analyzed intent, the server uses chat generation AI to generate text according to the instructions. The generated text reflects industry-standard formatting and structure, and is tailored to the user's needs.

[1452] Audio feedback and confirmation / correction

[1453] The generated text is sent to the device, which converts it into voice data using a speech synthesis engine. The device then plays the converted voice data back to the user, prompting them to confirm the content. The user can then use the voice feedback to confirm the content of the document and, if necessary, provide voice instructions for corrections.

[1454] For example, if a user says, "Please change the expression 'year-on-year comparison' to 'the percentage increase from the same month last year,'" the device captures this speech, converts it back into text data, and sends it to the server. The server again analyzes the intent, corrects it as necessary, and regenerates the text. This process is repeated until the user is satisfied.

[1455] Save the final document

[1456] When the user makes a final confirmation and gives the instruction to "save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., Word, PDF) and saves it in a database or cloud storage. After the saving process is complete, the device will inform the user by voice that the save is complete.

[1457] Specific examples

[1458] Example 1:

[1459] Launch and Start: When the user says "Start creating a report," the device captures the voice, converts it into text using a speech recognition engine, and sends it to the server.

[1460] Intent analysis and text generation: The server analyzes the intent of "Start creating a report," generates the text "Start creating a report. What would you like as your first item?", and sends it to the device.

[1461] Confirm and correct: If the user instructs "Write about sales analysis," the same process is followed to generate the text "This month's sales increased by 5% compared to the same month last year." The user then requests corrections, and the final content is "This month's sales increased by 5% compared to the same month last year."

[1462] Example 2:

[1463] Save: When the user says "Save this content," the server saves it as the final document and the device notifies the user by voice that the save is complete.

[1464] In this way, the system of the present invention enables desk workers to create and edit documents without using their eyes, reducing eye strain and supporting healthy working styles.

[1465] The processing flow will be explained below.

[1466] Step 1:

[1467] A user logs into a dedicated application or device and dictates "Start report creation."

[1468] Step 2:

[1469] The device captures the user's voice through the microphone and stores the voice data in a buffer.

[1470] Step 3:

[1471] The voice data captured by the device is sent to a voice recognition library and converted into text data.

[1472] Step 4:

[1473] The voice recognition library converts the voice data into text data and sends it back to the device.

[1474] Step 5:

[1475] The terminal encodes the converted text data and sends it to the server as an HTTP request.

[1476] Step 6:

[1477] The server decodes the encoded text data and analyzes it using a natural language processing engine.

[1478] Step 7:

[1479] The server recognizes the user's intention based on the analysis results and sends a request to the chat generation AI to generate text based on that intention.

[1480] Step 8:

[1481] The chat generation AI generates appropriate text based on the instructions and sends it back to the server.

[1482] Step 9:

[1483] The server receives the generated text data and sends it to the terminal as an HTTP response.

[1484] Step 10:

[1485] The text data received by the terminal is sent to a speech synthesis engine and converted into voice data.

[1486] Step 11:

[1487] The speech synthesis engine converts the text into speech data and sends it back to the device.

[1488] Step 12:

[1489] The terminal plays back the generated voice data to the user, prompting them to confirm the content.

[1490] Step 13:

[1491] The user confirms the result through voice feedback and, if necessary, gives voice instructions for correction.

[1492] Step 14:

[1493] The device captures the correction instructions and converts them into text data using a voice recognition library again.

[1494] Step 15:

[1495] The terminal transmits the corrected text data to the server.

[1496] Step 16:

[1497] The server parses the modified text data and regenerates the text as necessary.

[1498] Step 17:

[1499] The server transmits the regenerated text data to the terminal, where it is converted back into voice data and played back to the user.

[1500] Step 18:

[1501] This process is repeated, with revisions and validations being made until the user is satisfied.

[1502] Step 19:

[1503] The user issues a voice command saying, "Please save this content."

[1504] Step 20:

[1505] The terminal sends a final instruction to the server.

[1506] Step 21:

[1507] The server converts the verified text into a final document format (e.g., Word, PDF) and stores it in a database or cloud storage.

[1508] Step 22:

[1509] The server sends a notification of completion of saving to the terminal.

[1510] Step 23:

[1511] The device will notify the user by voice when the save is complete.

[1512] This allows users to create, edit, and save documents without using their eyes, reducing physical strain.

[1513] Example 1

[1514] 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."

[1515] Modern desk workers are suffering from health problems such as eye strain and vision loss due to long hours of computer work. Creating, reviewing, and correcting documents in particular places a heavy visual burden on the eyes, reducing efficiency and causing health problems. Furthermore, traditional document creation methods are difficult for visually impaired users. A system that can solve these problems and achieve a more efficient and healthy work environment is needed.

[1516] 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.

[1517] In this invention, the server includes a means for capturing a user's voice, a means for converting the captured voice into text data, and a means for analyzing the user's intent based on the converted text data. This allows the user to create, review, and modify documents using only voice instructions. Furthermore, by providing a means for creating documents using a generative AI model and a means for saving the final formatted document in data storage, efficient and accurate document creation and storage can be achieved. This reduces visual strain and supports healthy working styles.

[1518] A "means for capturing audio" is a device or system for capturing audio uttered by a user and processing it as a digital signal.

[1519] The "means for converting into text data" is a technique for analyzing the captured audio data and converting it into a corresponding text format.

[1520] "Means for analyzing user intent" refers to natural language processing technology used to understand the converted text data and identify what the user is looking for.

[1521] "Means for generating documents" refers to technology that uses a generative AI model to create text according to the specified content based on the user's intentions.

[1522] The "means for converting the text into audio data and presenting it to the user" refers to a technique for converting the generated or corrected text into audio and providing it to the user as an anti-visual interface.

[1523] The "means for correcting and regenerating a document according to confirmation and correction instructions" is a processing system for correcting the contents of a document in response to feedback from a user and regenerating the document as necessary.

[1524] "Means for saving in final format" refers to the technology for converting the completed document into a specified file format and saving it in a database, cloud storage, etc.

[1525] A "data analysis module" is a program or function that analyzes data based on the user's voice instructions and sends appropriate instructions to the generative AI model.

[1526] A "generative AI model" is an artificial intelligence technology that generates text based on a given prompt.

[1527] "Data storage" refers to a system or device for storing and managing data over the long term.

[1528] This invention relates to a system that allows desk workers to create, check, and correct documents by voice without using their eyes. This system generates text data based on the user's voice input and allows checking and correction through voice feedback. Each component and its operation are explained below.

[1529] System Configuration

[1530] This system consists of three main components: users, terminals, and servers.

[1531] Audio capture and preprocessing

[1532] The user gives voice instructions using a dedicated microphone or a microphone built into the device, and the device collects the user's voice in real time using a voice capture module, allowing the user to start working without needing an input device.

[1533] Voice Recognition

[1534] Once the audio is captured, the device converts it into text using a speech recognition engine (e.g., Google Cloud Speech-to-Text) that has high accuracy and takes into account various pronunciations and accents.

[1535] Intention Analysis and Text Generation

[1536] The converted text data is sent to the server and analyzed by a natural language processing engine (e.g., spaCy). The server analyzes this text data to recognize the user's intent. For example, if an instruction such as "Please write about sales analysis" is given, the server analyzes the intent and creates an appropriate text generation request.

[1537] Based on the parsed intent, the server uses a generative AI model (e.g., OpenAI GPT-3) to generate text based on the instructions, reflecting industry-standard formatting and structure and tailored to the user's needs.

[1538] Audio feedback and confirmation / correction

[1539] The generated text is sent to the device, which converts it into voice data using a speech synthesis engine (e.g., Amazon Polly). The device then plays the converted voice data back to the user, prompting them to confirm the content. The user can then confirm the content of the document through voice feedback and, if necessary, provide voice instructions for correction.

[1540] For example, if a user says, "Please change the expression 'compared to the same month last year' to 'the percentage increase from the same month last year,'" the device captures this speech, converts it into text data using a speech recognition engine, and sends it to the server. The server again analyzes the intent, corrects the text as necessary, and regenerates it. This process is repeated until the user is satisfied.

[1541] Save the final document

[1542] When the user makes a final confirmation and gives the instruction "Save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., Word, PDF) and saves it in a database or cloud storage (e.g., Google Drive, Dropbox). After the saving process is complete, the device will inform the user by voice that the save is complete.

[1543] Specific examples

[1544] Below is a concrete example of how this system can be used.

[1545] Example 1:

[1546] Boot and start:

[1547] When a user says "start report creation," the device captures the voice, converts it into text using a speech recognition engine, and sends it to the server.

[1548] The server analyzes the intent of "Start creating a report," generates the text "Start creating a report. What would you like as your first item?", and sends it to the terminal.

[1549] The user gives the voice instruction "Write about your sales analysis."

[1550] The generative AI model generates the text, "Sales this month increased by 5% compared to the same month last year."

[1551] The user further instructs, "Please change the year-on-year comparison to the increase rate from the same month last year."

[1552] The device captures the audio again and sends correction instructions to the server.

[1553] The server generates the text "Sales this month increased 5% from the same month last year" reflecting the correction.

[1554] Example 2:

[1555] keep:

[1556] When the user says "Save this content," the device captures the audio, converts it into text, and sends it to the server.

[1557] The server saves it as the final document, and the device notifies you with an audio message that the save is complete.

[1558] As described above, the system of the present invention allows desk workers to create, check, and edit documents using only voice commands, thereby reducing visual strain and supporting healthy working styles.

[1559] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1560] Step 1: Capture audio

[1561] Specific description:

[1562] Input: The user gives a voice command such as "Start report creation."

[1563] Specific operation: The user's voice is captured using a dedicated microphone or the device's built-in microphone.

[1564] Data processing: The terminal collects audio data using an audio capture module.

[1565] Output: The captured voice data is passed to the speech recognition engine.

[1566] Step 2: Voice Recognition

[1567] Specific description:

[1568] Input: Captured audio data.

[1569] Specific operation: The terminal converts voice data into text data using a voice recognition engine (e.g., Google Cloud Speech-to-Text).

[1570] Data processing: The speech recognition engine analyzes the speech waveform and converts it into a corresponding string of characters.

[1571] Output: The converted text data is generated.

[1572] Step 3: Intent Analysis

[1573] Specific description:

[1574] Input: Text data output from the speech recognition engine.

[1575] Specific operation: The server receives the text data sent from the terminal and analyzes it using a natural language processing engine (e.g., spaCy).

[1576] Data processing: Natural language processing engines analyze text data to identify user intent.

[1577] Output: Analysis results that reflect the user's intentions are generated.

[1578] Step 4: Text Generation

[1579] Specific description:

[1580] Input: Analysis results obtained from intent analysis.

[1581] Specific operation: The server uses a generative AI model (e.g., OpenAI GPT-3) to generate text based on the user's intent.

[1582] Data processing: The generative AI model generates relevant text based on the input prompt.

[1583] Output: Generated text data is generated.

[1584] Step 5: Audio feedback

[1585] Specific description:

[1586] Input: Generated text data.

[1587] Specific operation: The device converts the generated text data into audio data using a speech synthesis engine (e.g., Amazon Polly) and plays it back.

[1588] Data processing: The speech synthesis engine converts text data into speech waveforms.

[1589] Output: Audio data is generated that is played back to the user.

[1590] Step 6: Accepting correction instructions

[1591] Specific description:

[1592] Input: User's spoken correction instructions (e.g., "Please change the expression 'year-over-year comparison' to 'percent increase from the same month last year'").

[1593] Specific operation: The device again captures the user's voice correction instructions and converts them into text data using a voice recognition engine.

[1594] Data processing: The new voice command is converted into text and sent to the server.

[1595] Output: Text data containing correction instructions is passed to the server.

[1596] Step 7: Fix and Regenerate

[1597] Specific description:

[1598] Input: Text data containing correction instructions.

[1599] Specific behavior: The server parses the correction instructions, corrects the text as necessary, and regenerates it.

[1600] Data processing: The data analysis module analyzes the correction instructions and sends new prompts to the generative AI model to regenerate the text.

[1601] Output: The corrected text data is generated.

[1602] Step 8: Save the final document

[1603] Specific description:

[1604] Input: Final text data after correction (e.g., "Sales this month increased 5% from the same month last year.").

[1605] Specific operation: The server converts the final text data into a document format (e.g., Word, PDF) and saves it in data storage (e.g., Google Drive, Dropbox).

[1606] Data processing: The text data is converted into the specified file format and uploaded to cloud storage.

[1607] Output: A sound is generated to notify the user that the save is complete.

[1608] (Application example 1)

[1609] 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."

[1610] The objective of this invention is to provide a system that reduces the burden of deskwork in a factory environment and allows users to create, check, modify, and finally save reports using only voice commands. This system allows users to operate it by voice without using their eyes, improving worker health and increasing productivity. Furthermore, in a voice-only work environment, rapid and accurate data processing is required.

[1611] 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.

[1612] In this invention, the server includes means for capturing a user's voice, means for converting the captured voice into text data, means for analyzing the user's intent based on the converted text data, means for generating appropriate text based on the user's intent, means for converting the generated text into voice data and presenting it to the user, means for correcting and regenerating the text according to the user's confirmation and correction instructions, means for saving the generated and corrected text in a final document format, and robot means for realizing use in a voice-specific work environment, a factory environment where reports are created by voice instructions. This allows workers to create documents using only voice instructions, reducing eye fatigue and improving work efficiency.

[1613] "User" refers to the person who operates the system and gives voice instructions.

[1614] "Audio capturing means" refers to a device or method that captures the audio produced by a user as electronic data.

[1615] "Means for converting voice into text data" refers to technology that analyzes captured voice data and converts it into corresponding text data.

[1616] "Means of intent analysis" refers to technology that understands what a user wants from text data and identifies appropriate actions.

[1617] "Means for generating appropriate text" refers to technology that automatically creates the necessary information and sentences based on the user's intentions.

[1618] The "means for converting the generated text into audio data and presenting it to the user" refers to a device or method for converting the generated text back into audio data and allowing the user to listen to it.

[1619] "Means for correcting and regenerating text" refers to a technique for revising existing text and generating new text according to user instructions.

[1620] "Means for saving in final document format" refers to a technique for saving the final verified text in a specified format.

[1621] A "voice-focused work environment" refers to a workplace where operations and instructions are primarily performed by voice.

[1622] "Use in a factory environment where reports are created using voice instructions" refers to performing a series of tasks such as creating, modifying, and saving reports using only voice within the factory.

[1623] "Robotic means" refers to automated equipment or systems that implement voice commands in a factory environment.

[1624] This invention is a system for efficiently creating reports using voice instructions in a factory environment. The system generates text data based on the user's voice input and allows confirmation and correction through voice feedback. The specific configuration and processing procedure for realizing this system will be described below.

[1625] System Configuration

[1626] The system consists of four main components: users, terminals, servers, and robots.

[1627] 1. Audio capture and preprocessing

[1628] The user gives voice instructions using a dedicated microphone or the device's built-in microphone. The device collects the user's voice in real time using a voice capture module. The collected voice data is converted into text data using a voice recognition engine (e.g., Google Speech-to-Text API).

[1629] 2. Intention Analysis and Text Generation

[1630] The converted text data is sent to a server and analyzed by a natural language processing engine (e.g., OpenAI GPT-4). The server analyzes this text data to recognize the user's intent. For example, if an instruction such as "Please write about sales analysis" is given, the server analyzes the intent and creates an appropriate text generation request. Based on the analyzed intent, the server uses a generative AI model to generate text according to the instruction.

[1631] 3. Audio feedback and confirmation / correction

[1632] The generated text is sent to the device, which converts it into audio data using a speech synthesis engine (e.g., Google Text-to-Speech API). The device then plays the converted audio data back to the user, prompting them to confirm the content. The user can then use the audio feedback to confirm the content of the document and, if necessary, provide voice instructions for corrections.

[1633] 4. Preservation of the final document

[1634] When the user makes a final confirmation and gives the instruction to "save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., PDF) and saves it in a database or cloud storage (e.g., AWS S3). After the saving process is complete, the device will inform the user by voice that the save is complete.

[1635] Specific examples

[1636] Example 1: Starting a report

[1637] User: Says "Start creating a report."

[1638] The device collects the voice using a voice capture module and converts it into text data such as "Start creating report" using a voice recognition engine.

[1639] The server receives this text, generates the text "Let's start creating a report. What would you like as the first item?" and sends it to the terminal.

[1640] The terminal uses a speech synthesis engine to convert this text into audio data and play it back to the user.

[1641] Example 2: Sales analysis instructions and corrections

[1642] User: "Write about your sales analysis."

[1643] The terminal collects voice using a voice capture module and converts it into text data using a voice recognition engine.

[1644] The server analyzes this text data, recognizes the intent of "Please write about sales analysis," and generates the text "This month's sales increased by 5% compared to the same month last year."

[1645] The terminal converts the generated text into voice data using a voice synthesis engine and plays it back to the user.

[1646] When the user instructs, "Please change it to the increase rate compared to the same month last year," the device captures this voice, converts it back into text data, and sends it to the server.

[1647] The server again parses the intent and generates the corrected text, "Sales this month increased 5% from the same month last year."

[1648] The terminal converts the corrected text into voice data using a voice synthesis engine and plays it back to the user.

[1649] Prompt Sentence Examples

[1650] "Please write a report on your sales analysis. Please include specific figures and year-on-year increases or decreases."

[1651] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1652] Step 1: User initiates voice command

[1653] The user gives voice instructions (such as "Start creating a report") using a dedicated microphone or the device's built-in microphone. The device collects voice data in real time using a voice capture module. At this time, the user's voice is input, and the captured voice data is output.

[1654] Step 2: Convert the audio data to text

[1655] The device converts the collected voice data into text data using a voice recognition engine (e.g., Google Speech-to-Text API). Specifically, the captured voice data is input and passed through an advanced voice analysis algorithm to generate corresponding text data. This text data is then output.

[1656] Step 3: Analyze the text data

[1657] The server receives the converted text data and analyzes the user's intent through a natural language processing engine (e.g., OpenAI GPT-4). For example, it interprets the purpose of an instruction such as "start creating a report." In this case, the input is text data, and the output is the analysis result (data indicating the user's intent).

[1658] Step 4: Generate the appropriate text

[1659] The server uses a generative AI model based on the results of intent analysis to generate the required text. Based on the converted text data, a response text such as "I'm starting to write a report. What would you like as the first item?" is generated. The input is the analysis result, and the output is the generated text data.

[1660] Step 5: Convert the generated text to audio data

[1661] The device receives the generated text data and converts it into voice data using a speech synthesis engine (e.g., Google Text-to-Speech API). Specifically, it takes the generated text data as input and outputs it as voice waveform data. This voice data is then played back to the user.

[1662] Step 6: Receive user confirmation and correction instructions

[1663] The user can check the generated text through voice feedback and make corrections as necessary. For example, they can say, "Please change it to the increase rate compared to the same month last year." This voice instruction is input, captured by the device, and converted back into text data through the voice recognition engine (output).

[1664] Step 7: Regenerate the text based on the correction instructions

[1665] The server receives the text data again, analyzes the intent, and makes the necessary corrections. For example, based on the instruction "Please change this to the increase rate compared to the same month last year," the corrected text "This month's sales increased by 5% compared to the same month last year" is generated. The input is the text data of the correction instruction, and the output is the corrected text data.

[1666] Step 8: Convert the corrected text back into audio data and present it

[1667] The corrected text data is again converted into voice data by the terminal using a speech synthesis engine. It is then played back to the user for reconfirmation. This process is repeated until the user is satisfied. The input is the corrected text data, and the output is the corrected voice data.

[1668] Step 9: Save in the final document format

[1669] Finally, when the user instructs "Save this content," the server converts the final confirmed text data into a specified document format (e.g., PDF) and saves it in cloud storage (e.g., AWS S3). The input is the final confirmed text data, and the output is the saved document. The device notifies the user by voice when saving is complete.

[1670] 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.

[1671] This invention relates to a system that allows users to generate text through voice input and confirm and correct it by voice. A feature of this invention is that by incorporating an emotion engine that recognizes the user's emotions, the generated text and voice feedback are controlled to match the user's emotions. This improves the user experience and supports healthier working styles.

[1672] System Configuration

[1673] This system consists of four main components: the user, the terminal, the server, and the emotion engine.

[1674] Audio capture and preprocessing

[1675] The user gives voice instructions using a dedicated microphone or a microphone built into the device. The device collects the user's voice in real time using a voice capture module. This voice data is sent to a voice recognition engine and converted into text data.

[1676] emotion recognition

[1677] The converted text data and voice data are sent by the terminal to the emotion engine, which analyzes the user's emotions from the voice and text and generates emotion data. This emotion data is used for subsequent text generation and voice synthesis.

[1678] Intention Analysis and Text Generation

[1679] The server receives the text data and emotion data and analyzes the text data using a natural language processing engine. The server recognizes the user's intention based on the analysis results, and sends a request to the chat generation AI to generate text based on that intention, taking into account the emotion data. The generated text is expected to have a style and tone that matches the user's emotion.

[1680] Audio feedback and confirmation / correction

[1681] The generated text is sent to the device and converted into voice data by a speech synthesis engine. At this time, the tone and speed of the voice are adjusted based on the emotional data obtained from the emotion engine. The device then plays the adjusted voice data back to the user, prompting them to confirm the content. The user can check the content of the document through voice feedback and, if necessary, give voice instructions for corrections.

[1682] For example, if a user instructs the server to "Please prepare a sales report for the previous month," the server analyzes the instruction and generates text such as "Last month's sales increased by 10% compared to the plan." If the emotion engine recognizes that the user is nervous, the speech synthesis engine generates speech in a slightly slower, calmer tone.

[1683] Save the final document

[1684] When the user makes a final confirmation and gives the instruction to "save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., Word, PDF) and saves it in a database or cloud storage. After the saving process is complete, the device will inform the user by voice that the save is complete.

[1685] Specific examples

[1686] Example 1:

[1687] Launch and Start: The user says "Start report creation." The device captures the voice and converts it into text data using the speech recognition engine. At the same time, the emotion engine analyzes the user's current emotion and generates emotion data.

[1688] Intention analysis and text generation: The server analyzes the intention of "Start creating a report" and generates the text "Start creating a report. What would you like to do first?" taking into account the user's relaxation state.

[1689] Confirmation and correction: If the user instructs "Please write about sales analysis," the text generated reflects that state of mind: "This month's sales increased by 5% compared to the same month last year." If the user requests a correction, such as "Please change the expression "compared to the same month last year" to "the percentage increase from the same month last year," the text is corrected based on that request.

[1690] Example 2:

[1691] Save: When the user says "Save this content," the server saves it as the final document and the device notifies the user by voice that the save is complete.

[1692] In this way, the system of the present invention allows document creation and editing to be performed while taking the user's emotions into consideration, improving the user experience while reducing eye strain.

[1693] The processing flow will be explained below.

[1694] Step 1:

[1695] A user logs into a dedicated application or device and dictates "Start report creation."

[1696] Step 2:

[1697] The device captures the user's voice through the microphone and stores the voice data in a buffer.

[1698] Step 3:

[1699] The voice data captured by the device is sent to a voice recognition library and converted into text data.

[1700] Step 4:

[1701] The voice recognition library converts the voice data into text data and sends it back to the device.

[1702] Step 5:

[1703] The terminal transmits the converted text data to an emotion engine, which analyzes the user's emotions.

[1704] Step 6:

[1705] The emotion engine analyzes the voice and text data, generates the user's emotion data, and sends it back to the terminal.

[1706] Step 7:

[1707] The device encodes the text data and emotion data and sends them to the server as an HTTP request.

[1708] Step 8:

[1709] The server decodes the encoded text data and emotion data and analyzes the text data using a natural language processing engine.

[1710] Step 9:

[1711] The server recognizes the user's intention based on the analysis results, takes emotional data into consideration, and sends a request for text generation to the chat generation AI.

[1712] Step 10:

[1713] The chat generation AI generates appropriate text based on the intent and emotion data and sends it back to the server.

[1714] Step 11:

[1715] The server receives the generated text data and sends it to the terminal as an HTTP response.

[1716] Step 12:

[1717] The text data received by the terminal is sent to a speech synthesis engine and converted into voice data based on emotional data.

[1718] Step 13:

[1719] The speech synthesis engine takes into account the emotional data, converts the text into speech data, and sends it back to the device.

[1720] Step 14:

[1721] The terminal plays back the generated voice data to the user, prompting them to confirm the content.

[1722] Step 15:

[1723] The user confirms the result through voice feedback and, if necessary, gives voice instructions for correction.

[1724] Step 16:

[1725] The device captures the correction instructions and converts them into text data using a voice recognition library again.

[1726] Step 17:

[1727] The terminal transmits the corrected text data and emotion data to the server.

[1728] Step 18:

[1729] The server parses the modified text data and regenerates the text as necessary.

[1730] Step 19:

[1731] The server transmits the regenerated text data to the terminal, where it is converted back into voice data and played back to the user.

[1732] Step 20:

[1733] This process is repeated, with revisions and validations being made until the user is satisfied.

[1734] Step 20:

[1735] The user issues a voice command saying, "Please save this content."

[1736] Step 22:

[1737] The terminal sends a final instruction to the server.

[1738] Step 23:

[1739] The server converts the verified text into a final document format (e.g., Word, PDF) and stores it in a database or cloud storage.

[1740] Step 24:

[1741] The server sends a notification of completion of saving to the terminal.

[1742] Step 25:

[1743] The device will notify the user by voice when the save is complete.

[1744] This allows users to create, edit and save documents without using their eyes, and when combined with an emotion engine, provides more relevant and emotionally tailored feedback.

[1745] Example 2

[1746] 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."

[1747] Conventional document creation systems fail to consider the user's emotions when generating text using voice instructions and then reviewing and correcting it. This prevents the system from adjusting text and voice feedback based on the user's emotional state, resulting in a lack of user experience. This also makes it difficult to adjust the tempo and mood while creating documents, hindering efficient and healthy work practices. To address these issues, a system is needed that can recognize the user's emotions in real time and generate text and provide voice feedback that adapts to those emotions.

[1748] 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.

[1749] In this invention, the server includes means for capturing a user's voice, means for converting the captured voice into text data, means for analyzing the user's emotions based on the converted text data and voice data and generating emotion data, means for analyzing the user's intention based on the emotion data and text data, means for generating appropriate text based on the user's intention and emotion data, means for converting the generated text into voice data, adjusting the tone and speed of the voice based on the emotion data and presenting it to the user, means for correcting and regenerating the text in accordance with the user's confirmation and correction instructions, and means for saving the generated and corrected text in a final document format. This enables text generation and voice feedback adapted to the user's emotions, improving the user experience and realizing an efficient and healthy way of working.

[1750] An "audio capture means" is a device or software for collecting a user's voice in real time.

[1751] "Speech recognition means" refers to a device or software for converting collected voice data into text data.

[1752] The "emotion recognition means" is a device or software that analyzes the user's emotions based on the converted text data and voice data, and generates emotion data.

[1753] The "intention analysis means" is a device or software that analyzes the user's intention based on text data and emotion data.

[1754] The "text generation means" is a device or software that generates appropriate text based on the user's intention and emotion data.

[1755] A "speech synthesis means" is a device or software that converts the generated text into speech data and adjusts the tone and speed of the speech based on emotional data.

[1756] A "verification and correction means" is a device or software that corrects and regenerates text according to a user's verification and correction instructions.

[1757] A "data storage means" is a device or software that stores the generated and modified text in its final document format.

[1758] This system allows users to generate text through voice input and confirm and correct it by voice. One of the features of this system is that it incorporates an emotion engine that recognizes the user's emotions, and controls the generated text and voice feedback to match the user's emotions. This improves the user experience and supports more efficient and healthy working styles.

[1759] This system consists of four main components: the user, the terminal, the server, and the emotion engine.

[1760] Audio capture and preprocessing

[1761] The user gives voice instructions using a dedicated microphone or the device's built-in microphone. The device collects the user's voice in real time using a voice capture module. Specifically, Python's speech_recognition library can be used. This voice data is sent to a speech recognition engine (for example, Google Cloud Speech-to-Text API) and converted into text data.

[1762] emotion recognition

[1763] The device sends the converted text data and voice data to an emotion engine (e.g., IBM Watson Tone Analyzer). The emotion engine analyzes the user's emotions from the voice and text and generates emotion data. This emotion data is used for subsequent text generation and voice synthesis.

[1764] Intention Analysis and Text Generation

[1765] The server receives the emotion data and text data and analyzes the text data using a natural language processing engine (e.g., OpenAI's GPT-3). The server recognizes the user's intent based on the analysis results and sends a request to a chat generation AI (e.g., Microsoft Azure Chatbot) to generate text based on that intent, taking the emotion data into consideration. The generated text is expected to have a style and tone that matches the user's emotions.

[1766] Audio feedback and confirmation / correction

[1767] The generated text is sent to the device and converted into voice data by a speech synthesis engine (e.g., Amazon Polly). At this time, the tone and speed of the voice are adjusted based on the emotional data obtained from the emotion engine. The device then plays the adjusted voice data back to the user, prompting them to confirm the content. The user can check the content of the document through voice feedback and, if necessary, give voice instructions for corrections.

[1768] For example, if a user instructs the server to "Please prepare a sales report for the previous month," the server analyzes the instruction and generates text such as "Last month's sales increased by 10% compared to the plan." If the emotion engine recognizes that the user is nervous, the speech synthesis engine generates a voice with a slightly slower and calmer tone.

[1769] Save the final document

[1770] When the user makes a final confirmation and gives the instruction to "save this content," the device sends this to the server. The server converts the confirmed text into the final document format (e.g., Word, PDF) and saves it in a database or cloud storage. After the saving process is complete, the device notifies the user by voice that the save is complete.

[1771] Specific examples

[1772] Example 1:

[1773] Launch and start: The user says "Start report creation." The device captures the voice and converts it into text data using the speech recognition engine. At the same time, the emotion engine analyzes the user's current emotion and generates emotion data.

[1774] Intention analysis and text generation: The server analyzes the intention of "Start creating a report" and generates the text "Start creating a report. What would you like to do first?" taking into account the user's relaxation state.

[1775] Confirmation and correction: If the user instructs "Write about sales analysis," the text generated reflects that state of mind: "This month's sales increased by 5% compared to the same month last year." If the user requests a correction, such as "Please change the expression "compared to the same month last year" to "the percentage increase from the same month last year," the text is corrected based on that request.

[1776] Example 2:

[1777] Save: When the user says "Save this content," the server saves it as the final document and the device gives a voice notification that the save is complete.

[1778] In this way, the system of the present invention enables document creation and editing to be performed while taking the user's emotions into consideration, thereby improving the user experience and supporting efficient work methods.

[1779] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1780] Step 1:

[1781] The device uses a dedicated or built-in microphone to capture the user's voice. The user's voice instructions are provided as input. The device uses the Python speech_recognition library to collect voice data in real time. The collected voice data is sent to a speech recognition engine (e.g., Google Cloud Speech-to-Text API) which generates text data as output. This text data becomes the input for the next processing step.

[1782] Step 2:

[1783] The device sends the text data and voice data converted in step 1 to an emotion engine (for example, IBM Watson Tone Analyzer). The text data and voice data are provided as input. The emotion recognition engine analyzes this data and analyzes the user's emotions. Emotional data is generated as output, and this emotion data becomes the input to the next processing step. Specifically, the user's emotional state is identified by analyzing the choice of words read from the text data and the intonation of the voice data.

[1784] Step 3:

[1785] The server receives the emotion data generated in step 2 and the text data generated in step 1. Using these data as input, it analyzes the text data using a natural language processing engine (e.g., OpenAI's GPT-3). Based on the analysis results, it recognizes the user's intention and, taking the emotion data into consideration, sends a request to generate text to a chat generation AI (e.g., Microsoft Azure Chatbot). The generated text has a style and tone that matches the user's emotion. The appropriate generated text is obtained as output, and becomes the input for the next processing step.

[1786] Step 4:

[1787] The device receives the text data generated in step 3 and sends it to a speech synthesis engine (for example, Amazon Polly). The generated text data and emotion data are provided as input. The speech synthesis engine generates voice data while adjusting the tone and speed of the voice based on the emotion data. The generated voice data is obtained as output, and this is played back to the user. Specifically, the voice data is played back as feedback to the user, and the user confirms the content.

[1788] Step 5:

[1789] The user checks the content of the document based on the voice feedback played in step 4 and, if necessary, gives voice instructions for correction. The checked content and correction instructions are provided as input. The device uses a voice capture module to collect the user's correction instructions in real time and converts them into text data using a voice recognition engine again. The corrected text data is obtained as output and is sent again to the server for repeated processing.

[1790] Step 6:

[1791] When the user makes a final confirmation and issues the instruction "Save this content," the device sends this instruction to the server. The final confirmed text is provided as input. The server converts the confirmed text data into the final document format (e.g., PDF, Word, etc.). The specific output is a file in the converted document format, which is stored in a database or cloud storage. Finally, the device notifies the user by voice that the save is complete.

[1792] (Application example 2)

[1793] 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."

[1794] Conventional voice input systems generate text and provide voice feedback without considering the user's emotions, resulting in a flawed user experience. Furthermore, robots operating in factories lack the functionality to provide appropriate feedback based on emotions, which is insufficient for improving work efficiency and user satisfaction. To solve these problems, a system that can properly analyze the user's emotional state and reflect it in feedback is needed.

[1795] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1796] In this invention, the server includes means for capturing a user's voice, means for converting the captured voice into text data, means for analyzing the user's intention based on the converted text data, means for generating appropriate text based on the user's intention, means for converting the generated text into voice data and presenting it to the user, means for correcting and regenerating the text in accordance with the user's confirmation and correction instructions, means for saving the generated and corrected text in a final document format, means for capturing voice instructions and analyzing emotions, and means for adjusting the tone and speed of the voice data based on the emotion analysis results. This enables feedback that matches the user's emotions, thereby improving work efficiency in the factory and user satisfaction.

[1797] "Audio capture means" refers to devices or software for collecting a user's voice in real time.

[1798] "Speech recognition means" refers to the algorithms or software that convert collected voice data into text data.

[1799] "Intention analysis means" refers to algorithms or software that analyzes the converted text data and understands what the user intended.

[1800] "Text generation means" refers to algorithms or software that generate appropriate text based on the analyzed user intent.

[1801] "Speech synthesis means" refers to the algorithm or software used to convert the generated text into speech data.

[1802] "Text correction tool" refers to algorithms or software for correcting and regenerating text according to user confirmation and correction instructions.

[1803] "Document Storage" refers to the device or software used to store generated and modified text in its final document format.

[1804] "Emotion analysis means" refers to algorithms or software that analyze emotions from a user's voice or text and generate the results as data.

[1805] "Audio tone adjustment means" refers to algorithms or software for adjusting the tone or rate of audio data based on the results of emotion analysis.

[1806] The present invention aims to provide a factory robot management system that performs tasks based on voice instructions and provides feedback according to the user's emotions. The overall configuration of the system and specific embodiments thereof will be described below.

[1807] System Configuration

[1808] The system consists of the following major components:

[1809] 1. Users

[1810] 2. Terminal

[1811] 3. Server

[1812] 4. Sentiment Analysis Engine

[1813] User

[1814] The user gives voice instructions using a dedicated microphone or a microphone built into the device, providing work instructions to the factory robot in real time.

[1815] Terminal

[1816] The terminal collects the user's voice in real time using a voice capture means and transmits the voice data to the server. The terminal has the following functions.

[1817] Audio capture means: collects the user's voice in real time.

[1818] Speech recognition means: Converts collected voice data into text data.

[1819] Speech synthesis means: Converts the generated text into speech data and presents it to the user.

[1820] server

[1821] The server analyzes the user's intent and generates or modifies text and voice data using the following methods:

[1822] Intention analysis means: Analyze the converted text data to understand the user's intention.

[1823] Text generation means: Generates appropriate text based on the analyzed user intent.

[1824] Text correction means: Corrects and regenerates the text according to the user's confirmation and correction instructions.

[1825] Document storage means: storing the generated and modified text in the final document format.

[1826] Sentiment Analysis Engine

[1827] The emotion analysis engine analyzes emotions from the user's voice and text and generates the results as data. This emotion data is used as follows:

[1828] Sentiment analysis means: Analyzes emotions from the user's voice and text.

[1829] Voice tone adjustment means: Adjusts the tone and speed of voice data based on the results of emotion analysis.

[1830] System Operation

[1831] 1. Voice capture: The user dictates, "I will start the maintenance work on the machine." The device captures the voice and sends it to the server as voice data.

[1832] 2. Speech recognition: The server uses speech recognition means to convert this voice data into text data.

[1833] 3. Emotion analysis: The emotion analysis engine analyzes the text and voice data and generates data representing the user's emotions. In this case, it is analyzed that the user is nervous.

[1834] 4. Intention analysis and text generation: The server uses the intention analysis means to analyze the user's intention and generate text such as "Maintenance work has begun."

[1835] 5. Voice feedback: This text is converted into voice data using a voice synthesis means and played back to the user, with the tone and speed of the voice adjusted appropriately based on the sentiment analysis data.

[1836] 6. Confirm and correct: When the user says "Tell me the steps," the system regenerates the text and plays it back.

[1837] 7. Save document: When the user instructs "Save as this content," the server saves the generated text in the final document format.

[1838] Specific examples

[1839] Example 1:

[1840] User: "I'm going to start maintenance on the machine."

[1841] System: "Maintenance work will begin. Please advise how to proceed."

[1842] Example 2:

[1843] User: "Tell me how to do this."

[1844] System: "The first step is to power down the machine."

[1845] Prompt Sentence Examples

[1846] "I'm going to start maintenance on the machine. Please tell me the procedure."

[1847] This system enables feedback that matches the user's emotions, improving work efficiency in factories and user satisfaction.

[1848] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1849] Step 1:

[1850] Audio Capture

[1851] The user dictates, "I will start maintenance work on the machine."

[1852] Input: User's voice

[1853] The terminal collects the user's voice in real time using a voice capture means, and the collected voice data is then prepared for processing by a voice recognition engine.

[1854] Output: Audio data

[1855] Step 2:

[1856] Voice Recognition

[1857] Input: Audio data

[1858] The server converts the collected voice data into text data using a voice recognition engine. The voice recognition engine analyzes the phonemes in the voice data and generates corresponding character strings. This process provides the user's voice instructions in text format.

[1859] Output: Text data

[1860] Step 3:

[1861] Emotion analysis

[1862] Input: Text data, audio data

[1863] The server uses emotion analysis means to analyze the obtained text data and voice data and recognize the user's emotions. The emotion analysis engine analyzes the tone, speed, pitch, etc. of the sound and generates emotion data such as whether the user is tense or relaxed.

[1864] Output: Emotion data

[1865] Step 4:

[1866] Intention Analysis

[1867] Input: Text data, emotion data

[1868] The server uses intent analysis to analyze the text data and understand the user's intent. The natural language processing engine analyzes keywords and contextual information in the text data and recognizes the "intent to start maintenance work." This lays the foundation for generating an appropriate response.

[1869] Output: Intent data

[1870] Step 5:

[1871] Text Generation

[1872] Input: Intention data, emotion data

[1873] The server uses a text generation means to generate appropriate text based on the analyzed user intent. Since emotional data is also taken into account, the generated text is in a style and tone that matches the user's emotions. A response such as "Maintenance work will begin. Please tell me the procedure" is generated.

[1874] Output: Response text

[1875] Step 6:

[1876] Voice generation

[1877] Input: Response text, emotion data

[1878] The server uses a speech synthesis means to convert the generated response text into speech data. The tone and speed of the speech are appropriately adjusted based on the emotional data obtained from the emotion analysis means. For example, if the user is nervous, the speech speed is slowed down so that the speech can be played in a calmer tone.

[1879] Output: Modified audio data

[1880] Step 7:

[1881] Audio Feedback

[1882] Input: Modified audio data

[1883] The terminal plays the generated voice data to the user, and the user can confirm the system's response through this voice feedback and give the next instruction.

[1884] Output: User's understanding and next instructions

[1885] Step 8:

[1886] Check and correct

[1887] Input: The user's next voice command

[1888] If the user says "Tell me how to do it," the system recaptures this instruction and repeats the process of speech recognition, intent analysis, text generation, and speech generation. The resulting text is then reformatted.

[1889] Output: Corrected text and audio data

[1890] Step 9:

[1891] Document Storage

[1892] Input: Final confirmed text data

[1893] When the user selects "Save this content," the server saves the generated and modified text in the final document format (e.g., PDF, Word), and the saved document is stored in a database or cloud storage.

[1894] Output: Saved document

[1895] This system enables feedback that matches the user's emotions, improving work efficiency in factories and user satisfaction.

[1896] 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.

[1897] 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.

[1898] 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.

[1899] 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.

[1900] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1901] 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.

[1902] 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).

[1903] 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.

[1904] 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."

[1905] 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.

[1906] 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).

[1907] 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.

[1908] 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.

[1909] 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.

[1910] 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.

[1911] 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.

[1912] 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.

[1913] 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.

[1914] 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.

[1915] 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.

[1916] 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.

[1917] The following is further disclosed regarding the above embodiment.

[1918] (Claim 1)

[1919] means for capturing the user's voice;

[1920] A means for converting the captured audio into text data;

[1921] means for analyzing a user's intent based on the converted text data;

[1922] A means for generating appropriate text based on the user's intent;

[1923] means for converting the generated text into audio data and presenting the audio data to a user;

[1924] means for modifying and regenerating the text according to user confirmation and modification instructions;

[1925] means for saving the generated and modified text in a final document format;

[1926] A system including:

[1927] (Claim 2)

[1928] 10. The system of claim 1, further comprising speech recognition means for converting user dictation into text data.

[1929] (Claim 3)

[1930] 10. The system of claim 1, further comprising a speech synthesis means for converting text data into speech data.

[1931] "Example 1"

[1932] (Claim 1)

[1933] means for capturing the user's voice;

[1934] A means for converting the captured audio into text data;

[1935] means for analyzing a user's intent based on the converted text data;

[1936] A means for generating an appropriate document based on the user's intention;

[1937] means for converting the generated document into voice data and presenting the voice data to a user;

[1938] means for modifying and regenerating the document according to the user's review and modification instructions;

[1939] a means for saving the generated and modified documents in their final format;

[1940] means for transmitting a user's voice instructions to a data analysis module;

[1941] a means for the data analysis module to generate documents using the generative AI model;

[1942] means for storing the final formatted document in data storage;

[1943] A system including:

[1944] (Claim 2)

[1945] 10. The system of claim 1, further comprising speech recognition means for converting user dictation into text data.

[1946] (Claim 3)

[1947] 10. The system of claim 1, further comprising a speech synthesis means for converting the document into audio data.

[1948] "Application Example 1"

[1949] (Claim 1)

[1950] means for capturing the user's voice;

[1951] A means for converting the captured audio into text data;

[1952] means for analyzing a user's intent based on the converted text data;

[1953] A means for generating appropriate text based on the user's intent;

[1954] means for converting the generated text into audio data and presenting the audio data to a user;

[1955] means for modifying and regenerating the text according to user confirmation and modification instructions;

[1956] means for saving the generated and modified text in a final document format;

[1957] A robot means for realizing use in a factory environment in which a report is created by voice instructions in a voice-specific work environment;

[1958] A system including:

[1959] (Claim 2)

[1960] 10. The system of claim 1, further comprising speech recognition means for converting user dictation into text data.

[1961] (Claim 3)

[1962] 10. The system of claim 1, further comprising a speech synthesis means for converting text data into speech data.

[1963] "Example 2: Combining Emotion Engines"

[1964] (Claim 1)

[1965] means for capturing the user's voice;

[1966] A means for converting the captured audio into text data;

[1967] means for analyzing a user's emotions based on the converted text data and voice data and generating emotion data;

[1968] means for analyzing a user's intention based on emotion data and text data;

[1969] means for generating appropriate text based on user intent and emotion data;

[1970] means for converting the generated text into speech data, adjusting the tone and speed of the speech based on the emotion data, and presenting the speech to the user;

[1971] means for modifying and regenerating the text according to user confirmation and modification instructions;

[1972] means for saving the generated and modified text in a final document format;

[1973] A system including:

[1974] (Claim 2)

[1975] 10. The system of claim 1, further comprising speech recognition means for converting user dictation into text data.

[1976] (Claim 3)

[1977] 10. The system of claim 1, further comprising emotion recognition means for analyzing a user's emotion from the text data and the voice data and generating emotion data.

[1978] (Claim 4)

[1979] 10. The system of claim 1, further comprising a speech synthesis means for converting the text data into speech data and adjusting the tone and rate of the speech based on the emotion data.

[1980] (Claim 5)

[1981] 10. The system of claim 1, further comprising data storage means for storing the generated and modified text in a final document format.

[1982] "Application example 2 when combining emotion engines"

[1983] (Claim 1)

[1984] means for capturing the user's voice;

[1985] A means for converting the captured audio into text data;

[1986] means for analyzing a user's intent based on the converted text data;

[1987] A means for generating appropriate text based on the user's intent;

[1988] means for converting the generated text into audio data and presenting the audio data to a user;

[1989] means for modifying and regenerating the text according to user confirmation and modification instructions;

[1990] means for saving the generated and modified text in a final document format;

[1991] a means for capturing voice instructions and analyzing emotions;

[1992] a means for adjusting the tone and speed of the voice data based on the emotion analysis results;

[1993] A system including:

[1994] (Claim 2)

[1995] 10. The system of claim 1, further comprising speech recognition means for converting user dictation into text data.

[1996] (Claim 3)

[1997] 10. The system of claim 1, further comprising a speech synthesis means for converting text data into speech data. [Explanation of symbols]

[1998] 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 capturing the user's voice; A means for converting the captured audio into text data; means for analyzing a user's intent based on the converted text data; A means for generating appropriate text based on the user's intent; means for converting the generated text into audio data and presenting the audio data to a user; means for modifying and regenerating the text according to user confirmation and modification instructions; means for saving the generated and modified text in a final document format; A system including:

2. 10. The system of claim 1, further comprising speech recognition means for converting user dictation into text data.

3. 10. The system of claim 1, further comprising speech synthesis means for converting text data into speech data.

Images