System

Abstract

A system is provided.SOLUTION: A system comprising means for a user to input a question by voice, means for converting the voice question into text, means for transmitting the text data to a server, means for extracting a keyword by analyzing the text question, means for acquiring information from a database based on the keyword, means for generating an answer by using the acquired information, means for transmitting the generated answer to a terminal, means for converting the text answer into voice, and means for reproducing the voice answer.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] There is a need for new and junior employees to be able to easily resolve work-related questions and problems in a remote work environment. With conventional systems, the timing and means for asking questions to senior employees are limited, leading to reluctance to frequently repeated questions. This leads to problems such as reduced work efficiency and increased employee turnover. It is necessary to provide a support system that solves these issues and allows new and junior employees to work with peace of mind. [Means for solving the problem]

[0005] The present invention provides a system that includes a means for a user to input a question by voice, a means for converting the voice question into text, a means for transmitting the text data to a server, a means for analyzing the text question and extracting keywords, a means for retrieving information from a database based on the keywords, a means for generating an answer using the retrieved information, a means for transmitting the generated answer to a terminal, a means for converting the text answer into voice, and a means for playing back the voice answer. This system allows users to easily ask work-related questions and receive prompt answers, even in a remote work environment. Furthermore, by utilizing natural language processing and natural language generation technologies, it is possible to provide information in a format that is easy for users to understand.

[0006] "User" refers to the individual who operates the system and asks questions verbally.

[0007] "Means for inputting a question by voice" refers to a microphone or voice input device that a user uses to input voice.

[0008] "Means for converting voice questions into text" refers to a function that uses voice recognition technology to convert input voice questions into character data (text).

[0009] The "means for transmitting text data to a server" refers to a function for transmitting the converted text data to a server via a network using a communication protocol.

[0010] "Means for analyzing text questions and extracting keywords" refers to a function that uses natural language processing technology to identify relevant keywords and important phrases from text questions.

[0011] "Means of obtaining information from a database" refers to the function of searching an internal database based on keywords and extracting and obtaining relevant information.

[0012] "Means for generating an answer" refers to a function that uses the acquired information to generate an appropriate answer for the user in natural language.

[0013] The "means for transmitting the generated answer to the terminal" refers to a function for transferring the generated answer to the terminal.

[0014] "Means for converting text responses into voice" refers to a function that converts text responses into voice data using voice synthesis technology.

[0015] "Means for playing back audio responses" refers to a function for playing back audio data through an audio output device such as a speaker or earphones. [Brief explanation of the drawings]

[0016] [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

[0017] 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.

[0018] First, the terms used in the following description will be explained.

[0019] 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).

[0020] 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.

[0021] 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.

[0022] 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.

[0023] 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."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 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.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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."

[0037] The system of the present invention enables new and junior employees to easily solve work-related questions and problems in a remote work environment. The system includes a means for a user to input a question by voice, a means for converting the voice question into text, a means for transmitting the text data to a server, a means for analyzing the text question and extracting keywords, a means for retrieving information from a database based on the keywords, a means for generating an answer using the retrieved information, a means for transmitting the generated answer to a terminal, a means for converting the text answer into voice, and a means for playing back the voice answer.

[0038] First, the user speaks a question about work into the device's microphone. This spoken question is converted into text data by the device's voice recognition engine. For example, if a user asks, "Senior, where can I find the format for this report?", the voice is converted into the text "Where can I find the format for this report?"

[0039] The device then sends the converted text data to the server as an HTTP request, which also includes user authentication information and is transmitted over a secure communication channel.

[0040] The server passes the received text data to a natural language processing engine, which analyzes the question. Specifically, it extracts important keywords and phrases from the question. In this example, keywords such as "report," "format," and "where" are extracted. The server uses these keywords to search the company database to obtain relevant information. For example, it obtains information such as "The report format can be found in the document template section of the company portal."

[0041] Based on the acquired information, the generative AI model on the server generates an appropriate answer. This answer is then formatted using natural language generation technology into a format that is easy for the user to understand. The generated answer is then sent to the device in text format.

[0042] The device passes the received text response to a speech synthesis engine, which converts it into voice data. Finally, the converted voice data is played back through the device's speaker or earphones. This allows the user to hear the answer to their question in voice, as if it were being answered directly by a senior employee.

[0043] For example, if a user asks a work-related question, "Senior, where can I find the format for this report?", the process is as follows: The device converts this speech into text and sends it to the server. The server searches the database based on the analysis and generates the answer, "The report format can be found in the document template section of the in-house portal." The generated answer is converted into speech and played back to the user via the device.

[0044] This system utilizes natural language processing and generation technologies to provide timely answers in a format that is easy for users to understand. It also allows new and junior employees to receive support as if they were being supported by a senior employee, even in a remote work environment, improving the work efficiency of new and junior employees and contributing to a lower turnover rate.

[0045] The processing flow will be explained below.

[0046] Step 1:

[0047] The user speaks into the device's microphone to ask a work-related question. For example, "Senior, where can I find the format for this report?"

[0048] Step 2:

[0049] The device uses a speech recognition engine to convert the user's spoken question into text. The speech data is analyzed and the corresponding text is generated. For example, the speech is converted to "Where can I find the format for this report?"

[0050] Step 3:

[0051] The device sends the converted text data to the server in the form of an HTTP request, which also includes the user's authentication token. For example, a text question and the authentication token are sent.

[0052] Step 4:

[0053] The server passes the received text data to a natural language processing engine, which analyzes the question. The engine extracts keywords and important phrases from the text. For example, it extracts keywords such as "report," "format," and "where."

[0054] Step 5:

[0055] The server searches the company database based on the analysis results to retrieve relevant information. It generates a search query and queries the database. For example, it executes the query "SELECT FROM documentation_templates WHERE category="report format"" and retrieves the results.

[0056] Step 6:

[0057] The generative AI model creates an answer based on the information obtained by the server. The answer is formatted using natural language generation technology in a way that is easy for the user to understand. For example, "The report format can be found in the document template section of the internal portal."

[0058] Step 7:

[0059] The server generates a text response and sends it to the terminal as an HTTP response. For example, the response text "The report format can be found in the document template section of the internal portal" is sent.

[0060] Step 8:

[0061] The text response received by the device is passed to the speech synthesis engine, which converts the text into speech data. The synthesized speech file is saved in temporary memory. Example: "The report format can be found in the document template section of the in-house portal" is converted into speech data.

[0062] Step 9:

[0063] The device outputs the audio data to a playback device (speaker or earphones). By listening to the answer, the user feels as if a senior employee had answered them directly. Example: Play the audio "The report format can be found in the document template section of the internal portal."

[0064] By following these steps, users can quickly get answers to their business-related questions even in a remote work environment.

[0065] Example 1

[0066] 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."

[0067] In a remote work environment, new and junior employees lack the means to easily resolve work-related questions and problems. This can lead to reduced work efficiency and delayed employee growth. There are also concerns that the difficulty of communication in a remote environment could lead to increased employee turnover.

[0068] 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.

[0069] In this invention, the server includes means for allowing a user to input a question by voice, means for converting the voice question into text, means for transmitting the text data to an information processing device, means for analyzing the text question and extracting keywords, means for acquiring information from an information storage device based on the keywords, means for generating an answer using the acquired information, means for transmitting the generated answer to a display device, means for converting the text answer into voice, means for playing back the voice answer, means for generating an answer in a format that is easy for the user to understand using natural language generation technology, and means for ensuring secure communication using HTTPS. This enables smooth communication even in a remote work environment, allowing new and junior employees to quickly solve problems, which is expected to improve work efficiency and reduce turnover.

[0070] A "user" is a person who uses the system to input questions by voice and receive answers.

[0071] A "means for inputting a question by voice" is a microphone or other voice input device that allows a user to input a voice question into the system.

[0072] A "means for converting voice queries into text" is a device or software that includes speech recognition technology or algorithms that analyzes input voice data and converts it into text data.

[0073] The "means for transmitting text data to the information processing device" refers to a communication function or protocol for transferring text data to a server via a network.

[0074] An "information processing device" is a computer system or server that analyzes text data and generates answers.

[0075] "Means for analyzing text questions and extracting keywords" refers to natural language processing techniques and algorithms that identify and extract important keywords from received text data.

[0076] "Means for retrieving information from an information storage device based on keywords" refers to a technique for using extracted keywords to search for and retrieve related information from databases, knowledge bases, and other information storage devices.

[0077] An "information storage device" is a storage device for storing databases and other digital information.

[0078] The "means for generating a response using acquired information" refers to a device or software that generates a response sentence using natural language generation technology based on the acquired information.

[0079] The "means for transmitting the generated answer to the display device" refers to a communication function or protocol for transferring the generated answer to the user's terminal via a network.

[0080] The "display device" refers to a display, speaker, or earphone for displaying or playing back the answer to the user.

[0081] "Means for converting text responses to speech" refers to speech synthesis techniques and algorithms that convert text responses into speech data.

[0082] The "means for playing back audio responses" refers to a device or system for playing back audio data through a speaker or earphones.

[0083] "Natural language generation technology" refers to the technology and algorithms used to analyze text data and generate natural-sounding sentences that are easy for humans to understand.

[0084] "Means of ensuring secure communications through HTTPS" refers to protocols and technologies that ensure the confidentiality and integrity of data by encrypting and sending and receiving communication data.

[0085] A "prompt" is text containing specific instructions or questions that a generative AI model uses to generate appropriate answers.

[0086] The system of the present invention allows new and junior employees to easily resolve work-related questions and problems in a remote work environment. This system is realized using the following hardware and software.

[0087] The user speaks a question about work into the microphone on the device. For example, the user might ask, "Senior, where can I find the format for this report?" This spoken question is converted into text data using a speech recognition engine (e.g., Google (registered trademark) Speech-to-Text API) installed on the device. The converted text data is then sent from the device to the server via secure communication using HTTPS.

[0088] The server passes the received text data to a natural language processing engine (e.g., SpaCy or BERT) and analyzes the question. Specifically, it tokenizes the text and extracts keywords such as "report," "format," and "where." Next, the server searches an information storage device (database or knowledge base) based on the extracted keywords to obtain related information. For example, it can obtain information such as "The report format can be found in the document template section of the internal portal."

[0089] Based on the acquired information, an appropriate answer is generated using a generative AI model on the server (e.g., OpenAI's GPT-3 (registered trademark)). At this time, the prompt text is "Please generate an answer that is easy for the user to understand based on the information, 'The report format can be found in the document template section of the internal portal.'" The generative AI model generates a natural and easy-to-understand answer based on this prompt text.

[0090] The generated answer is sent to the device in text format. The device then passes the received text answer to a speech synthesis engine (e.g., Amazon Polly or Google Text-to-Speech) and converts it into audio data. Finally, the device plays the converted audio data through a speaker or earphones. This process allows the user to hear the answer to their question in audio, as if it were being answered directly by a senior employee.

[0091] This system's unique feature is its use of natural language processing and natural language generation technologies, allowing it to quickly provide answers in a format that is easy for users to understand. It also uses the HTTPS protocol, which ensures secure communications, ensuring high communication security. By using this system, new and junior employees can communicate smoothly even in a remote work environment, which is expected to improve work efficiency and reduce turnover.

[0092] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0093] Step 1:

[0094] The user speaks a question into the microphone of the device. Specifically, the user speaks, "Senior, where is the format for this report?" The input of this step is the user's voice data, and the output is also voice data.

[0095] Step 2:

[0096] The device passes the voice data acquired from the microphone to a voice recognition engine. Specifically, a voice recognition engine (e.g., Google Speech-to-Text API) is used to convert the voice data into text data. The input of this step is the voice data, and the output is the text data, "Where can I find the format for this report?"

[0097] Step 3:

[0098] The terminal sends the converted text data to the server as an HTTPS request. Specifically, it generates a request including the text data and user authentication information and sends it to the server via a communication channel encrypted by SSL / TLS. The input of this step is the text data and authentication information, and the output is an HTTPS request to the server.

[0099] Step 4:

[0100] The server passes the received text data to a natural language processing engine. Specifically, it uses a natural language processing engine (e.g., SpaCy or BERT) to analyze the text and extract important keywords. For example, it extracts keywords such as "report," "format," and "where." The input of this step is the received text data, and the output is the extracted keywords.

[0101] Step 5:

[0102] The server searches the information storage device based on the extracted keywords. Specifically, it generates an SQL query to search a database (e.g., MySQL (registered trademark) or PostgreSQL) and retrieves related information. For example, it obtains information such as "The report format is in the document template section of the in-house portal." The input of this step is the extracted keywords, and the output is the retrieved information.

[0103] Step 6:

[0104] The server uses a generative AI model to generate an answer based on the acquired information. Specifically, the prompt "Generate an answer that is easy for the user to understand based on the information 'The report format can be found in the document template section of the internal portal'" is input into a generative AI model (e.g., OpenAI's GPT-3) to generate a natural and easy-to-understand answer. The input for this step is the acquired information and the prompt, and the output is the generated answer in text format.

[0105] Step 7:

[0106] The server sends the generated text answer to the device using HTTPS. Specifically, the text answer is sent over a communication channel that is again encrypted with SSL / TLS. The input to this step is the generated text answer, and the output is an HTTPS response to the device.

[0107] Step 8:

[0108] The device passes the received text response to a speech synthesis engine (e.g., Amazon Polly or Google Text-to-Speech) to convert the text to speech data. The input of this step is the received text response, and the output is speech data.

[0109] Step 9:

[0110] The terminal plays the audio data through a speaker or earphone. Specifically, using the terminal's audio output device, the user hears the audio response as if the senior employee were speaking directly to them. The input for this step is the audio data, and the output is the audio response provided to the user.

[0111] By following these steps, users can get quick and accurate answers even in a remote work environment.

[0112] (Application example 1)

[0113] 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."

[0114] Factory floors require a means for operators to quickly and accurately give instructions and ask questions to robots. However, existing systems mainly use text input, and few support voice input. This prevents operators from responding quickly, resulting in reduced production efficiency. Furthermore, even if systems exist that support voice input, they lack natural language processing and generation technologies, making it difficult to provide accurate answers and instructions. There is a need to provide a system that can solve these issues and enable operators to communicate smoothly with robots.

[0115] 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.

[0116] In this invention, the server includes means for a user to input a question by voice, means for converting the voice question into text, means for transmitting the text data to the server, means for analyzing the text question and extracting keywords, means for retrieving information from a database based on the keywords, means for generating an answer using the retrieved information, means for transmitting the generated answer to a terminal, means for converting the text answer into voice, means for playing back the voice answer, and means for a user to give instructions or questions to a robot in a factory by voice, and for the robot to analyze the voice and return an appropriate answer or instruction by voice. This makes it possible for an operator to give instructions or questions to the robot by voice, thereby quickly and accurately acquiring information or issuing operating instructions.

[0117] "Voice input" is a means of capturing and processing the user's voice into an electronic device.

[0118] "Text conversion" is the process of converting information captured through speech into written data.

[0119] "Server transmission" refers to the act of transmitting the converted character data to the central processing unit.

[0120] "Text analysis" is a technology that extracts important keywords from transmitted text data and analyzes its meaning.

[0121] "Database acquisition" refers to searching and acquiring related information from a database based on the analysis results.

[0122] "Answer generation" is the process of creating an appropriate answer to a user's question based on the acquired information.

[0123] "Terminal transmission" is the act of transmitting the generated answer to a terminal accessible by the user.

[0124] "Speech conversion" is a technology that converts answers from text data into voice data.

[0125] "Audio playback" refers to the act of playing audio data through a speaker or the like and providing it to the user.

[0126] A "factory floor" is a facility where various production activities take place and where employees and robots work.

[0127] A "robot" is a machine that performs programmed actions and automates tasks.

[0128] This invention is a system that supports effective communication between operators and robots in factories. This system is composed of means for speech recognition, text conversion, text analysis, database search, answer generation, speech synthesis, and speech playback.

[0129] First, the user, an operator, gives instructions or asks questions to the robot by voice. To achieve this, the robot or its connected device is equipped with a microphone. For example, when the operator says, "Robot, how do I maintain this machine?", the voice data is captured.

[0130] Next, the terminal converts the acquired voice data into text data using a voice recognition engine (for example, a SpeechRecognition library), and sends the converted text data to a central processing unit (server) as an HTTP request.

[0131] Once the text data arrives at the server, it is analyzed by a natural language processing engine (for example, Hugging Face's Transformers library) to extract important keywords. In this example, keywords such as "machine," "maintenance," and "method" are extracted.

[0132] Next, the server uses these keywords to search for and retrieve relevant information from a database. The database stores various manuals and work instructions for factory machines. For example, the server might retrieve information such as, "To maintain a machine, first turn off the power and remove the safety devices."

[0133] Based on the acquired information, the server uses a generative AI model to generate an appropriate answer. This generative AI model incorporates technology to generate answers in a format that is easy for the user to understand. The generated answer is sent to the device in text format.

[0134] Finally, the terminal passes the received text response to a speech synthesis engine (for example, the Pyttsx3 library) and converts it into voice data. This voice data is then played back to the operator through a speaker, allowing the user to hear the answer to their question aloud, as if the robot were answering them directly.

[0135] For example, if an operator asks, "Robot, what's the next step?", the system will respond, "The next step is to install part A." Or, if the operator asks, "Robot, how do you maintain this machine?", the system will respond, "The way to maintain this machine is to first turn off the power and remove the safety devices."

[0136] An example of a prompt sentence to input to the generative AI model is:

[0137] "Robot, how do you maintain this machine?"

[0138] "Robot, what's the next step?"

[0139] Let's say.

[0140] In this way, a system can be realized that improves communication between operators and robots in factories and increases work efficiency.

[0141] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0142] Step 1:

[0143] The user enters a question by voice.

[0144] Specifically, the operator speaks into the microphone and asks a question, for example, "Robot, how do you maintain this machine?" The input here is voice data.

[0145] Step 2:

[0146] The terminal sends the voice data to a voice recognition engine and converts it into text data.

[0147] Specifically, the voice data acquired by the device's microphone is converted into text data using the SpeechRecognition library, resulting in the text data "Robot, how do you maintain this machine?"

[0148] Step 3:

[0149] The terminal transmits the text data to the server.

[0150] Specifically, the converted text data is sent to the server as an HTTP request. This request also includes user authentication information. The input is text data, and the output is a request to the server.

[0151] Step 4:

[0152] The server passes the text data to a natural language processing engine to extract keywords.

[0153] Specifically, the server analyzes the received text data using Hugging Face's Transformers library and extracts the keywords "machine," "maintenance," and "method." The input is text data, and the output is keyword data.

[0154] Step 5:

[0155] The server searches the database based on the keywords to retrieve relevant information.

[0156] Specifically, the server uses the extracted keywords to search a database containing manuals and work instructions for factory machines. The information obtained is, "To maintain the machine, first turn off the power and remove the safety devices." The input is keyword data, and the output is the acquired information.

[0157] Step 6:

[0158] Based on the information acquired by the server, an appropriate answer is generated using a generative AI model.

[0159] Specifically, the server uses the acquired information to send a prompt to a generative AI model (e.g., GPT-3) to generate an answer such as "First, turn off the power and remove the safety device." The input is the acquired information, and the output is the generated answer text.

[0160] Step 7:

[0161] The server sends the generated response to the terminal.

[0162] Specifically, the generated answer text is sent from the server to the terminal as an HTTP response. The input is the generated answer text, and the output is the response to the terminal.

[0163] Step 8:

[0164] The answer received by the terminal is passed to a speech synthesis engine and converted into voice data.

[0165] Specifically, the device passes the received text response to the Pyttsx3 library and converts it into audio data. The input is the text response and the output is audio data.

[0166] Step 9:

[0167] The terminal plays the generated audio data to the user through a speaker or earphone.

[0168] Specifically, a voice message saying, "First, turn off the power and remove the safety device" is played back. The input is voice data, and the output is voice playback.

[0169] This series of steps creates a system in which the user can ask questions by voice and receive answers by voice.

[0170] 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.

[0171] The system of the present invention enables new and junior employees to easily solve work-related questions and problems in a remote work environment. The system includes a means for a user to input a question by voice, a means for converting the voice question into text, a means for transmitting the text data to a server, a means for analyzing the text question and extracting keywords, a means for retrieving information from a database based on the keywords, a means for generating an answer using the retrieved information, a means for transmitting the generated answer to a terminal, a means for converting the text answer into voice, and a means for playing back the voice answer.

[0172] Furthermore, the present invention provides a more personalized response by combining an emotion engine that recognizes emotions from the user's voice. This emotion engine can recognize the user's emotional state and incorporate that information into the answer generation process.

[0173] First, the user speaks a work-related question into the device's microphone. For example, "Senior, where is the format for this report?" This spoken question is converted into text data by the device, and the emotion engine then analyzes the user's emotional state. The emotion engine analyzes the tone, pace, emphasis, etc. of the voice, and recognizes the user's emotional state, such as "I'm wondering" or "I'm confused."

[0174] The device then sends the converted text data and the recognized emotion data to the server. The server then passes the received text data to a natural language processing engine, which analyzes the question and extracts important keywords and phrases. For example, from the question "Where can I find the format for this report?", keywords such as "report," "format," and "where" are extracted.

[0175] The server searches the company database based on these keywords to retrieve relevant information, for example, "The report format can be found in the document template section of the company portal."

[0176] The server then uses the generative AI model to generate an appropriate response, taking into account the emotional data from the emotion engine. For example, if the server detects that the user is confused, the response will be more polite and detailed. The generated response is sent to the device in text format.

[0177] The device passes the received text response to a speech synthesis engine, which converts it into voice data. This voice data is then played through the device's speaker or earphones. For example, a polite voice might say, "The report format can be found in the document template section of the internal portal."

[0178] In this way, users can receive answers to their questions via voice, as if they were being answered directly by a senior employee. Furthermore, the emotion engine recognizes the user's emotional state and responds accordingly, providing more appropriate and personalized support. For example, if a user asks a work-related question, "Senior, where can I find the format for this report?", the system will go through the steps of voice conversion, emotion recognition, information acquisition, answer generation, and voice synthesis to provide the user with an appropriate answer via voice. This is particularly beneficial for new and junior employees in remote work environments, improving work efficiency and contributing to lower employee turnover.

[0179] The processing flow will be explained below.

[0180] Step 1:

[0181] The user speaks into the device's microphone to ask a work-related question. For example, "Senior, where can I find the format for this report?"

[0182] Step 2:

[0183] The device uses a speech recognition engine to convert the user's voice question into text. The voice data is analyzed and the corresponding text is generated. For example, "Where can I find the format for this report?"

[0184] Step 3:

[0185] While the device converts the voice question into text, it also uses an emotion engine to analyze the voice data and recognize the user's emotional state, for example by extracting emotions such as "doubtful" or "confused" from the tone and pace of the voice.

[0186] Step 4:

[0187] The device sends the converted text data and the recognized emotion data to the server in the form of an HTTP request. This request also includes the user's authentication token. Example: Send a text question, emotion data, and authentication token to the server.

[0188] Step 5:

[0189] The server passes the received text data to a natural language processing engine, which analyzes the question. The engine extracts keywords and important phrases from the text. For example, it extracts keywords such as "report," "format," and "where."

[0190] Step 6:

[0191] The server searches the company database based on the analysis results to retrieve relevant information. It generates a search query and queries the database. For example, it executes the query "SELECT FROM documentation_templates WHERE category="report format"" to retrieve relevant information.

[0192] Step 7:

[0193] The generative AI model creates a response based on the information obtained by the server. It adjusts the tone and level of detail of the response taking into account emotional data from the emotion engine. For example, if the user is perceived as "confused," the response will be more detailed and polite.

[0194] Step 8:

[0195] The server generates a text response and sends it to the device as an HTTP response. The server returns the text response to the device. Example: "The report format can be found in the document template section of the internal portal."

[0196] Step 9:

[0197] The text response received by the device is passed to the speech synthesis engine, which converts the text into speech data. The synthesized speech file is saved in temporary memory. For example, it is converted to "The report format can be found in the document template section of the internal portal."

[0198] Step 10:

[0199] The device outputs the audio data to a playback device (speaker or earphones). By listening to the answer, the user feels as if a senior employee had answered them directly. Example: "The report format can be found in the document template section of the internal portal."

[0200] Through this series of steps, users can receive prompt and personalized responses even in a remote work environment. In particular, the introduction of an emotion engine allows for appropriate responses based on the user's emotional state, which is expected to improve the work efficiency of new and junior employees and reduce their psychological burden.

[0201] Example 2

[0202] 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."

[0203] In recent years, with the rapid spread of remote work environments, new and junior employees are seeking ways to quickly and efficiently resolve work-related questions and problems. Conventional systems lack support for resolving work-related questions, particularly the lack of a method that combines voice input and emotion analysis. Furthermore, there is a lack of systems that can provide personalized responses that take into account the user's emotional state. This reduces the work efficiency of new and junior employees and leads to increased turnover, so a new system is needed to solve this issue.

[0204] 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.

[0205] In this invention, the server includes a means for analyzing a user's emotions, a means for reflecting the analyzed emotion data in the answer generation process, and a means for generating an appropriate answer using a generative AI model, thereby enabling the server to quickly provide appropriate and personalized answers to questions entered by the user via voice.

[0206] "User" refers to a person who uses the system to solve a business question or problem.

[0207] "Means for inputting questions by voice" refers to a device including a microphone and an interface that a user uses to input voice.

[0208] "Means for converting voice queries into text" refers to speech recognition software or algorithms used to convert user-entered voice data into text data.

[0209] "Means for transmitting text data to a server" refers to a communication means for sending text data to a server via a network.

[0210] "Means for analyzing text questions and extracting keywords" refers to natural language processing techniques used to analyze received text data and identify important words and phrases.

[0211] "Means for retrieving information from a database based on keywords" refers to means for searching a database using extracted keywords and retrieving related information.

[0212] "Means for generating an answer using acquired information" refers to the technology or algorithm used to create an appropriate answer based on the acquired information.

[0213] The "means for transmitting the generated answer to the terminal" refers to a communication means for sending the generated answer to the terminal via a network.

[0214] "Means for converting text responses to speech" refers to a speech synthesis engine or algorithm for converting the generated text responses into speech data.

[0215] The "means for playing back an audio response" refers to a device such as a speaker or earphone for playing back audio data and providing a response to the user.

[0216] "Means for analyzing user emotions" refers to an emotion engine or algorithm for analyzing the user's emotional state from their voice or text input.

[0217] "Means for incorporating analyzed emotional data into the answer generation process" refers to technologies and algorithms for adjusting answers to take into account the user's emotional state.

[0218] "Generative AI model" refers to an artificial intelligence model for generating text using natural language generation technology.

[0219] The system of the present invention allows new and junior employees to easily solve work-related questions and problems in a remote work environment. The system includes a means for users to input questions by voice, a means for converting the voice question into text, a means for transmitting the text data to a server, a means for analyzing the text question and extracting keywords, a means for retrieving information from a database based on the keywords, a means for generating an answer using the retrieved information, a means for transmitting the generated answer to a terminal, a means for converting the text answer into voice, and a means for playing back the voice answer. Furthermore, the present invention provides more personalized responses by combining a means for analyzing the user's emotions with a means for incorporating the analyzed emotional data into the answer generation process.

[0220] The user speaks a question about work into the device's microphone. The device used for this is a computer device such as a regular PC, smartphone, or tablet. For example, a user might ask, "Senior, where can I find the format for this report?" This spoken question is converted into text data using speech recognition software on the device (for example, Google Cloud Speech-to-Text).

[0221] The text data is then analyzed by an emotion engine (e.g., IBM Watson® Tone Analyzer) to determine the user's emotional state. The emotion engine analyzes the tone, pace, and emphasis of the voice to recognize the user's emotional state (e.g., "questioning" or "confused"). The device then transmits the converted text data and the recognized emotion data to the server.

[0222] The server passes the received text data to a natural language processing engine (e.g., spaCy) and analyzes the question. Important keywords and phrases are extracted, such as "report," "format," and "where." The server then searches a database (e.g., a MySQL database) based on these keywords to retrieve relevant information. For example, it retrieves the information, "The report format can be found in the document templates section of the internal portal."

[0223] The server then generates an appropriate response using a generative AI model (e.g., OpenAI's GPT-3) while taking into account the emotional data from the emotion engine. For example, if the server recognizes that the user is "confused," the response will be more polite and detailed. The generated response is sent to the device in text format.

[0224] The device passes the received text response to a speech synthesis engine (e.g., Amazon Polly) and converts it into voice data. This voice data is played back through the device's speaker or earphones. For example, it may say in a polite tone, "The report format can be found in the document template section of the internal portal."

[0225] A unique feature of this system is that it analyzes the user's emotions and uses that information to adjust the answer generation process, allowing users to receive personalized support as if they were being answered directly by a senior employee.

[0226] As a concrete example, if a user asks a work-related question such as "Senior, where can I find the format for this report?", the system will go through the steps of voice conversion, emotion recognition, information acquisition, answer generation, and voice synthesis to provide the user with an appropriate answer via voice.An example of a prompt sentence for the generative AI model is shown below.

[0227] Example prompt sentence:

[0228] User question: "Senior, where can I find the format for this report?"

[0229] Extracted keywords: "report", "format", "where"

[0230] Emotion recognized: "I'm confused"

[0231] Database lookup result: "The report format can be found in the Document Templates section of the Intranet Portal."

[0232] Generate an appropriate response considering the recognized emotion.

[0233] These prompts can be fed into a generative AI model to generate appropriate answers.

[0234] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0235] Step 1:

[0236] The user speaks into the microphone of the terminal to input a question about the business.

[0237] As a specific operation, when a user says, "Senior, where is the format for this report?", this speech is input.

[0238] Step 2:

[0239] The device uses voice recognition software such as Google Cloud Speech-to-Text to convert the user's voice data into text data.

[0240] Input: Voice data "Senior, where can I find the format for this report?"

[0241] Output: Text data "Senior, where can I find the format for this report?"

[0242] Specifically, speech recognition software analyzes the speech waveform and converts it into corresponding text.

[0243] Step 3:

[0244] The device then sends the converted text data to an emotion engine such as IBM Watson Tone Analyzer to analyze the user's emotional state.

[0245] Input: Text data "Senior, where can I find the format for this report?"

[0246] Output: Emotion data "Confused"

[0247] Specifically, the emotion engine analyzes the tone, pace, and emphasis of the voice to recognize the user's emotions.

[0248] Step 4:

[0249] The terminal transmits the converted text data and the recognized emotion data to the server.

[0250] Input: Text data "Senior, where is the format for this report?", Emotion data "Confused"

[0251] Output: Data sent to the server

[0252] As a specific operation, data is transmitted from the terminal to the server via the network.

[0253] Step 5:

[0254] The server passes the received text data to a natural language processing engine (e.g., spaCy) to analyze the question content and extract keywords.

[0255] Input: Text data "Senior, where can I find the format for this report?"

[0256] Output: Keywords "report", "format", "where"

[0257] Specifically, a natural language processing engine analyzes the text and identifies important words and phrases.

[0258] Step 6:

[0259] Based on the extracted keywords, the server searches internal databases such as MySQL databases to obtain relevant information.

[0260] Input: Keywords "report", "format", "where"

[0261] Output: Retrieved information: "The report format can be found in the document template section of the intranet."

[0262] Specifically, a database query is performed to retrieve relevant information.

[0263] Step 7:

[0264] The server takes into account the acquired information and emotional data and generates an appropriate answer using a generative AI model such as OpenAI's GPT-3.

[0265] Input: Information retrieved from the database: "The report format can be found in the document template section of the internal portal.", Emotion data: "Confused"

[0266] Output: Generated answer "The report format can be found in the document templates section of the internal portal. Please contact us if you require more information."

[0267] Specifically, the generative AI model receives a prompt and generates a natural-sounding response based on the context.

[0268] Step 8:

[0269] The generated text response is sent to the device, which converts it into voice data using a speech synthesis engine such as Amazon Polly.

[0270] Input: Text response "The report format can be found in the document templates section of the intranet. Please contact us if you require more information."

[0271] Output: Audio data

[0272] Specifically, the speech synthesis engine converts the text into speech and generates speech data.

[0273] Step 9:

[0274] The terminal plays back the generated voice data and provides the answer to the user.

[0275] Input: Audio data

[0276] Output: The user hears the audio response

[0277] Specifically, the audio data is played through the terminal's speaker or earphones.

[0278] (Application example 2)

[0279] 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."

[0280] It is difficult for workers on factory production lines to receive prompt and appropriate support when they have questions or problems related to their work. New or junior employees in particular need real-time assistance when they have questions or are confused about their work. Traditional methods require direct interaction with superiors or senior employees, which can lead to lower productivity. Furthermore, responses that do not take into account the worker's emotional state run the risk of increasing stress. Therefore, personalized support for workers is needed to improve work efficiency within factories.

[0281] 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.

[0282] In this invention, the server includes means for a user to input a question by voice, means for converting the voice question into text, means for transmitting the text data to the server, means for analyzing the text question and extracting keywords, means for retrieving information from a database based on the keywords, means for generating an answer using the retrieved information, means for transmitting the generated answer to a terminal, means for converting the text answer into voice, means for playing back the voice answer, and means for analyzing emotions. This allows factory workers to solve their questions and problems in real time and receive appropriate support that takes their emotional state into consideration.

[0283] "Users" refer to factory workers who use the system.

[0284] "Means for inputting questions by voice" refers to a device or method that allows a worker to input questions by voice using a microphone or the like.

[0285] "Means for converting voice queries into text" refers to a device or method that uses voice recognition technology to convert an input voice query into text data.

[0286] The "means for transmitting text data to a server" refers to a device or method for transmitting the converted text data to a server via a network.

[0287] "Means for analyzing text questions and extracting keywords" refers to a device or method that uses natural language processing techniques to extract important keywords from text questions.

[0288] The term "means for retrieving information from a database based on keywords" refers to a device or method that uses the extracted keywords to search for and retrieve related information from a database.

[0289] "Means for generating an answer using acquired information" refers to a device or method that generates an appropriate answer based on information acquired from a database.

[0290] The "means for transmitting the generated answer to the terminal" refers to a device or method for transmitting the generated answer to the user's terminal via a network.

[0291] "Means for converting text responses to speech" refers to a device or method that uses speech synthesis technology to convert generated text responses into speech data.

[0292] "Means for playing back audio responses" refers to a device or method for playing back converted audio data through a speaker or earphones.

[0293] "Emotion analysis means" refers to a device or method that analyzes the emotional state of a worker from voice or text and personalizes assistance based on that information.

[0294] The system of the present invention helps factory production line workers to quickly resolve work-related questions and problems remotely. The system's main components combine functions such as voice input, text conversion, emotion analysis, data transmission, information acquisition, answer generation, and voice playback.

[0295] In order to implement the present invention, the following hardware and software are used.

[0296] 1. Voice input

[0297] Hardware: Microphone

[0298] Software: Google Cloud Speech-to-Text API

[0299] Description: A user speaks into a microphone to ask a question about a task. For example, a question could be, "What is the maintenance procedure for this machine?"

[0300] 2. Text Conversion

[0301] Hardware: Devices (smartphones, tablets, PCs, etc.)

[0302] Software: Google Cloud Speech-to-Text API

[0303] Description: The input speech is converted to text data. The converted text is "What is the maintenance procedure for this machine?"

[0304] 3. Emotion analysis

[0305] Hardware: Built-in microphone

[0306] Software: Microsoft® Azure® Emotion API

[0307] Description: Analyzes a user's emotional state from speech and text. For example, recognizes the emotion "confused" from a user's voice.

[0308] 4. Data Transmission

[0309] Hardware: Terminal, Wi-Fi module

[0310] Software: HTTPS communication

[0311] Description: Sends the converted text data and analyzed emotion data to the server.

[0312] 5. Information acquisition

[0313] Hardware: Server

[0314] Software: Natural language processing engine, database search algorithm

[0315] Description: The server extracts keywords from the received text data and searches the database for related information based on these keywords. For example, it retrieves related procedure manuals based on the keyword "machine maintenance procedures."

[0316] 6. Answer generation

[0317] Hardware: Server

[0318] Software: Generative AI model (OpenAI GPT-4 (registered trademark))

[0319] Description: Generate an appropriate answer based on the acquired information and emotion data. The generated answer will be in text format, such as "Details of the machine maintenance procedure can be found on page 5 of the procedure manual."

[0320] 7. Audio playback

[0321] Hardware: Device, speakers or earphones

[0322] Software: Google Cloud Text-to-Speech API

[0323] Description: The generated text response is converted to audio data and played back. The user hears, "The machine maintenance procedure is detailed on page 5 of the procedure manual."

[0324] As a result, workers feel as if they are being directly answered by a senior employee, allowing them to solve problems quickly and accurately.Furthermore, emotion analysis makes it possible to provide personalized responses based on the user's emotions, which is expected to improve work efficiency and reduce stress.

[0325] For example, if a user asks, "What is the maintenance procedure for this machine?", the system goes through the steps of voice conversion, emotion analysis, information acquisition, answer generation, and voice synthesis, and provides a spoken answer such as, "Details of the machine's maintenance procedure are on page 5 of the instruction manual."

[0326] An example of a prompt sentence is as follows:

[0327] Question: "What is the maintenance procedure for this machine?" Emotion: "I'm confused."

[0328] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0329] Step 1:

[0330] Users speak into a microphone to ask questions about their work. This voice becomes input to the system. For example, they might ask, "What is the maintenance procedure for this machine?"

[0331] Step 2:

[0332] The device receives voice input and converts it to text using the Google Cloud Speech-to-Text API. The converted text becomes the input for the next step in the system. The output is the text, "What are the maintenance procedures for this machine?"

[0333] Step 3:

[0334] The device then passes the converted text data to the Microsoft Azure Emotion API for emotion analysis. The analysis identifies the emotion the user expressed while asking the question. For example, emotional information such as "confused" is output. This emotional data is then used as input for the next step.

[0335] Step 4:

[0336] The device sends text data and emotion data to the server. The sent data is analyzed by the server's natural language processing engine, and important keywords are extracted. For example, from the question "What is the maintenance procedure for this machine?" keywords such as "machine," "maintenance," and "procedure" are extracted.

[0337] Step 5:

[0338] The server retrieves related information from the database based on the extracted keywords. For example, it searches using the keywords "machine," "maintenance," and "procedure" to retrieve procedure manuals and related technical documents. This retrieved information becomes the input for the next step.

[0339] Step 6:

[0340] The server generates an appropriate answer using a generative AI model (e.g., OpenAI GPT-4) based on the acquired information and emotion data. The generated answer is output as text data. For example, a response such as "Details of the machine maintenance procedure are on page 5 of the instruction manual" may be generated.

[0341] Step 7:

[0342] The server sends the generated text response to the device. The device passes it to the Google Cloud Text-to-Speech API and converts it into voice data. The converter outputs the voice data, "Details of the machine maintenance procedure can be found on page 5 of the procedure manual."

[0343] Step 8:

[0344] The terminal plays the generated voice data through a speaker or earphone, allowing the user to receive a voice response in real time. Specifically, the speaker will say, "Details of the machine maintenance procedure can be found on page 5 of the procedure manual."

[0345] 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.

[0346] 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.

[0347] 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.

[0348] [Second embodiment]

[0349] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0350] 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.

[0351] 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).

[0352] 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.

[0353] 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.

[0354] 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).

[0355] 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.

[0356] 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.

[0357] 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.

[0358] 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.

[0359] 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.

[0360] 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."

[0361] The system of the present invention enables new and junior employees to easily solve work-related questions and problems in a remote work environment. The system includes a means for a user to input a question by voice, a means for converting the voice question into text, a means for transmitting the text data to a server, a means for analyzing the text question and extracting keywords, a means for retrieving information from a database based on the keywords, a means for generating an answer using the retrieved information, a means for transmitting the generated answer to a terminal, a means for converting the text answer into voice, and a means for playing back the voice answer.

[0362] First, the user speaks a question about work into the device's microphone. This spoken question is converted into text data by the device's voice recognition engine. For example, if a user asks, "Senior, where can I find the format for this report?", the voice is converted into the text "Where can I find the format for this report?"

[0363] The device then sends the converted text data to the server as an HTTP request, which also includes user authentication information and is transmitted over a secure communication channel.

[0364] The server passes the received text data to a natural language processing engine, which analyzes the question. Specifically, it extracts important keywords and phrases from the question. In this example, keywords such as "report," "format," and "where" are extracted. The server uses these keywords to search the company database to obtain relevant information. For example, it obtains information such as "The report format can be found in the document template section of the company portal."

[0365] Based on the acquired information, the generative AI model on the server generates an appropriate answer. This answer is then formatted using natural language generation technology into a format that is easy for the user to understand. The generated answer is then sent to the device in text format.

[0366] The device passes the received text response to a speech synthesis engine, which converts it into voice data. Finally, the converted voice data is played back through the device's speaker or earphones. This allows the user to hear the answer to their question in voice, as if it were being answered directly by a senior employee.

[0367] For example, if a user asks a work-related question, "Senior, where can I find the format for this report?", the process is as follows: The device converts this speech into text and sends it to the server. The server searches the database based on the analysis and generates the answer, "The report format can be found in the document template section of the in-house portal." The generated answer is converted into speech and played back to the user via the device.

[0368] This system utilizes natural language processing and generation technologies to provide timely answers in a format that is easy for users to understand. It also allows new and junior employees to receive support as if they were being supported by a senior employee, even in a remote work environment, improving the work efficiency of new and junior employees and contributing to a lower turnover rate.

[0369] The processing flow will be explained below.

[0370] Step 1:

[0371] The user speaks into the device's microphone to ask a work-related question. For example, "Senior, where can I find the format for this report?"

[0372] Step 2:

[0373] The device uses a speech recognition engine to convert the user's spoken question into text. The speech data is analyzed and the corresponding text is generated. For example, the speech is converted to "Where can I find the format for this report?"

[0374] Step 3:

[0375] The device sends the converted text data to the server in the form of an HTTP request, which also includes the user's authentication token. For example, a text question and the authentication token are sent.

[0376] Step 4:

[0377] The server passes the received text data to a natural language processing engine, which analyzes the question. The engine extracts keywords and important phrases from the text. For example, it extracts keywords such as "report," "format," and "where."

[0378] Step 5:

[0379] The server searches the company database based on the analysis results to retrieve relevant information. It generates a search query and queries the database. For example, it executes the query "SELECT FROM documentation_templates WHERE category="report format"" and retrieves the results.

[0380] Step 6:

[0381] The generative AI model creates an answer based on the information obtained by the server. The answer is formatted using natural language generation technology in a way that is easy for the user to understand. For example, "The report format can be found in the document template section of the internal portal."

[0382] Step 7:

[0383] The server generates a text response and sends it to the terminal as an HTTP response. For example, the response text "The report format can be found in the document template section of the internal portal" is sent.

[0384] Step 8:

[0385] The text response received by the device is passed to the speech synthesis engine, which converts the text into speech data. The synthesized speech file is saved in temporary memory. Example: "The report format can be found in the document template section of the in-house portal" is converted into speech data.

[0386] Step 9:

[0387] The device outputs the audio data to a playback device (speaker or earphones). By listening to the answer, the user feels as if a senior employee had answered them directly. Example: Play the audio "The report format can be found in the document template section of the internal portal."

[0388] By following these steps, users can quickly get answers to their business-related questions even in a remote work environment.

[0389] Example 1

[0390] 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."

[0391] In a remote work environment, new and junior employees lack the means to easily resolve work-related questions and problems. This can lead to reduced work efficiency and delayed employee growth. There are also concerns that the difficulty of communication in a remote environment could lead to increased employee turnover.

[0392] 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.

[0393] In this invention, the server includes means for allowing a user to input a question by voice, means for converting the voice question into text, means for transmitting the text data to an information processing device, means for analyzing the text question and extracting keywords, means for acquiring information from an information storage device based on the keywords, means for generating an answer using the acquired information, means for transmitting the generated answer to a display device, means for converting the text answer into voice, means for playing back the voice answer, means for generating an answer in a format that is easy for the user to understand using natural language generation technology, and means for ensuring secure communication using HTTPS. This enables smooth communication even in a remote work environment, allowing new and junior employees to quickly solve problems, which is expected to improve work efficiency and reduce turnover.

[0394] A "user" is a person who uses the system to input questions by voice and receive answers.

[0395] A "means for inputting a question by voice" is a microphone or other voice input device that allows a user to input a voice question into the system.

[0396] A "means for converting voice queries into text" is a device or software that includes speech recognition technology or algorithms that analyzes input voice data and converts it into text data.

[0397] The "means for transmitting text data to the information processing device" refers to a communication function or protocol for transferring text data to a server via a network.

[0398] An "information processing device" is a computer system or server that analyzes text data and generates answers.

[0399] "Means for analyzing text questions and extracting keywords" refers to natural language processing techniques and algorithms that identify and extract important keywords from received text data.

[0400] "Means for retrieving information from an information storage device based on keywords" refers to a technique for using extracted keywords to search for and retrieve related information from databases, knowledge bases, and other information storage devices.

[0401] An "information storage device" is a storage device for storing databases and other digital information.

[0402] The "means for generating a response using acquired information" refers to a device or software that generates a response sentence using natural language generation technology based on the acquired information.

[0403] The "means for transmitting the generated answer to the display device" refers to a communication function or protocol for transferring the generated answer to the user's terminal via a network.

[0404] The "display device" refers to a display, speaker, or earphone for displaying or playing back the answer to the user.

[0405] "Means for converting text responses to speech" refers to speech synthesis techniques and algorithms that convert text responses into speech data.

[0406] The "means for playing back audio responses" refers to a device or system for playing back audio data through a speaker or earphones.

[0407] "Natural language generation technology" refers to the technology and algorithms used to analyze text data and generate natural-sounding sentences that are easy for humans to understand.

[0408] "Means of ensuring secure communications through HTTPS" refers to protocols and technologies that ensure the confidentiality and integrity of data by encrypting and sending and receiving communication data.

[0409] A "prompt" is text containing specific instructions or questions that a generative AI model uses to generate appropriate answers.

[0410] The system of the present invention allows new and junior employees to easily resolve work-related questions and problems in a remote work environment. This system is realized using the following hardware and software.

[0411] The user speaks a question about work into the device's microphone. For example, they might ask, "Senior, where can I find the format for this report?" This spoken question is converted into text data using the device's built-in speech recognition engine (e.g., Google Speech-to-Text API). The converted text data is then sent from the device to the server via secure communication using HTTPS.

[0412] The server passes the received text data to a natural language processing engine (e.g., SpaCy or BERT) and analyzes the question. Specifically, it tokenizes the text and extracts keywords such as "report," "format," and "where." Next, the server searches an information storage device (database or knowledge base) based on the extracted keywords to obtain related information. For example, it can obtain information such as "The report format can be found in the document template section of the internal portal."

[0413] Based on the acquired information, an appropriate answer is generated using a generative AI model on the server (e.g., OpenAI's GPT-3). At this time, the prompt text is "Please generate an answer that is easy for the user to understand based on the information, 'The report format can be found in the document template section of the internal portal.'" The generative AI model generates a natural and easy-to-understand answer based on this prompt text.

[0414] The generated answer is sent to the device in text format. The device then passes the received text answer to a speech synthesis engine (e.g., Amazon Polly or Google Text-to-Speech) and converts it into audio data. Finally, the device plays the converted audio data through a speaker or earphones. This process allows the user to hear the answer to their question in audio, as if it were being answered directly by a senior employee.

[0415] This system's unique feature is its use of natural language processing and natural language generation technologies, allowing it to quickly provide answers in a format that is easy for users to understand. It also uses the HTTPS protocol, which ensures secure communications, ensuring high communication security. By using this system, new and junior employees can communicate smoothly even in a remote work environment, which is expected to improve work efficiency and reduce turnover.

[0416] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0417] Step 1:

[0418] The user speaks a question into the microphone of the device. Specifically, the user speaks, "Senior, where is the format for this report?" The input of this step is the user's voice data, and the output is also voice data.

[0419] Step 2:

[0420] The device passes the voice data acquired from the microphone to a voice recognition engine. Specifically, a voice recognition engine (e.g., Google Speech-to-Text API) is used to convert the voice data into text data. The input of this step is the voice data, and the output is the text data, "Where can I find the format for this report?"

[0421] Step 3:

[0422] The terminal sends the converted text data to the server as an HTTPS request. Specifically, it generates a request including the text data and user authentication information and sends it to the server via a communication channel encrypted by SSL / TLS. The input of this step is the text data and authentication information, and the output is an HTTPS request to the server.

[0423] Step 4:

[0424] The server passes the received text data to a natural language processing engine. Specifically, it uses a natural language processing engine (e.g., SpaCy or BERT) to analyze the text and extract important keywords. For example, it extracts keywords such as "report," "format," and "where." The input of this step is the received text data, and the output is the extracted keywords.

[0425] Step 5:

[0426] The server searches the information storage device based on the extracted keywords. Specifically, it generates an SQL query to search a database (e.g., MySQL or PostgreSQL) and retrieves related information. For example, it obtains information such as "The report format is in the document template section of the in-house portal." The input of this step is the extracted keywords, and the output is the retrieved information.

[0427] Step 6:

[0428] The server uses a generative AI model to generate an answer based on the acquired information. Specifically, the prompt "Generate an answer that is easy for the user to understand based on the information 'The report format can be found in the document template section of the internal portal'" is input into a generative AI model (e.g., OpenAI's GPT-3) to generate a natural and easy-to-understand answer. The input for this step is the acquired information and the prompt, and the output is the generated answer in text format.

[0429] Step 7:

[0430] The server sends the generated text answer to the device using HTTPS. Specifically, the text answer is sent over a communication channel that is again encrypted with SSL / TLS. The input to this step is the generated text answer, and the output is an HTTPS response to the device.

[0431] Step 8:

[0432] The device passes the received text response to a speech synthesis engine (e.g., Amazon Polly or Google Text-to-Speech) to convert the text to speech data. The input of this step is the received text response, and the output is speech data.

[0433] Step 9:

[0434] The terminal plays the audio data through a speaker or earphone. Specifically, using the terminal's audio output device, the user hears the audio response as if the senior employee were speaking directly to them. The input for this step is the audio data, and the output is the audio response provided to the user.

[0435] By following these steps, users can get quick and accurate answers even in a remote work environment.

[0436] (Application example 1)

[0437] 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."

[0438] Factory floors require a means for operators to quickly and accurately give instructions and ask questions to robots. However, existing systems mainly use text input, and few support voice input. This prevents operators from responding quickly, resulting in reduced production efficiency. Furthermore, even if systems exist that support voice input, they lack natural language processing and generation technologies, making it difficult to provide accurate answers and instructions. There is a need to provide a system that can solve these issues and enable operators to communicate smoothly with robots.

[0439] 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.

[0440] In this invention, the server includes means for a user to input a question by voice, means for converting the voice question into text, means for transmitting the text data to the server, means for analyzing the text question and extracting keywords, means for retrieving information from a database based on the keywords, means for generating an answer using the retrieved information, means for transmitting the generated answer to a terminal, means for converting the text answer into voice, means for playing back the voice answer, and means for a user to give instructions or questions to a robot in a factory by voice, and for the robot to analyze the voice and return an appropriate answer or instruction by voice. This makes it possible for an operator to give instructions or questions to the robot by voice, thereby quickly and accurately acquiring information or issuing operating instructions.

[0441] "Voice input" is a means of capturing and processing the user's voice into an electronic device.

[0442] "Text conversion" is the process of converting information captured through speech into written data.

[0443] "Server transmission" refers to the act of transmitting the converted character data to the central processing unit.

[0444] "Text analysis" is a technology that extracts important keywords from transmitted text data and analyzes its meaning.

[0445] "Database acquisition" refers to searching and acquiring related information from a database based on the analysis results.

[0446] "Answer generation" is the process of creating an appropriate answer to a user's question based on the acquired information.

[0447] "Terminal transmission" is the act of transmitting the generated answer to a terminal accessible by the user.

[0448] "Speech conversion" is a technology that converts answers from text data into voice data.

[0449] "Audio playback" refers to the act of playing audio data through a speaker or the like and providing it to the user.

[0450] A "factory floor" is a facility where various production activities take place and where employees and robots work.

[0451] A "robot" is a machine that performs programmed actions and automates tasks.

[0452] This invention is a system that supports effective communication between operators and robots in factories. This system is composed of means for speech recognition, text conversion, text analysis, database search, answer generation, speech synthesis, and speech playback.

[0453] First, the user, an operator, gives instructions or asks questions to the robot by voice. To achieve this, the robot or its connected device is equipped with a microphone. For example, when the operator says, "Robot, how do I maintain this machine?", the voice data is captured.

[0454] Next, the terminal converts the acquired voice data into text data using a voice recognition engine (for example, a SpeechRecognition library), and sends the converted text data to a central processing unit (server) as an HTTP request.

[0455] Once the text data arrives at the server, it is analyzed by a natural language processing engine (for example, Hugging Face's Transformers library) to extract important keywords. In this example, keywords such as "machine," "maintenance," and "method" are extracted.

[0456] Next, the server uses these keywords to search for and retrieve relevant information from a database. The database stores various manuals and work instructions for factory machines. For example, the server might retrieve information such as, "To maintain a machine, first turn off the power and remove the safety devices."

[0457] Based on the acquired information, the server uses a generative AI model to generate an appropriate answer. This generative AI model incorporates technology to generate answers in a format that is easy for the user to understand. The generated answer is sent to the device in text format.

[0458] Finally, the terminal passes the received text response to a speech synthesis engine (for example, the Pyttsx3 library) and converts it into voice data. This voice data is then played back to the operator through a speaker, allowing the user to hear the answer to their question aloud, as if the robot were answering them directly.

[0459] For example, if an operator asks, "Robot, what's the next step?", the system will respond, "The next step is to install part A." Or, if the operator asks, "Robot, how do you maintain this machine?", the system will respond, "The way to maintain this machine is to first turn off the power and remove the safety devices."

[0460] An example of a prompt sentence to input to the generative AI model is:

[0461] "Robot, how do you maintain this machine?"

[0462] "Robot, what's the next step?"

[0463] Let's say.

[0464] In this way, a system can be realized that improves communication between operators and robots in factories and increases work efficiency.

[0465] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0466] Step 1:

[0467] The user enters a question by voice.

[0468] Specifically, the operator speaks into the microphone and asks a question, for example, "Robot, how do you maintain this machine?" The input here is voice data.

[0469] Step 2:

[0470] The terminal sends the voice data to a voice recognition engine and converts it into text data.

[0471] Specifically, the voice data acquired by the device's microphone is converted into text data using the SpeechRecognition library, resulting in the text data "Robot, how do you maintain this machine?"

[0472] Step 3:

[0473] The terminal transmits the text data to the server.

[0474] Specifically, the converted text data is sent to the server as an HTTP request. This request also includes user authentication information. The input is text data, and the output is a request to the server.

[0475] Step 4:

[0476] The server passes the text data to a natural language processing engine to extract keywords.

[0477] Specifically, the server analyzes the received text data using Hugging Face's Transformers library and extracts the keywords "machine," "maintenance," and "method." The input is text data, and the output is keyword data.

[0478] Step 5:

[0479] The server searches the database based on the keywords to retrieve relevant information.

[0480] Specifically, the server uses the extracted keywords to search a database containing manuals and work instructions for factory machines. The information obtained is, "To maintain the machine, first turn off the power and remove the safety devices." The input is keyword data, and the output is the acquired information.

[0481] Step 6:

[0482] Based on the information acquired by the server, an appropriate answer is generated using a generative AI model.

[0483] Specifically, the server uses the acquired information to send a prompt to a generative AI model (e.g., GPT-3) to generate an answer such as "First, turn off the power and remove the safety device." The input is the acquired information, and the output is the generated answer text.

[0484] Step 7:

[0485] The server sends the generated response to the terminal.

[0486] Specifically, the generated answer text is sent from the server to the terminal as an HTTP response. The input is the generated answer text, and the output is the response to the terminal.

[0487] Step 8:

[0488] The answer received by the terminal is passed to a speech synthesis engine and converted into voice data.

[0489] Specifically, the device passes the received text response to the Pyttsx3 library and converts it into audio data. The input is the text response and the output is audio data.

[0490] Step 9:

[0491] The terminal plays the generated audio data to the user through a speaker or earphone.

[0492] Specifically, a voice message saying, "First, turn off the power and remove the safety device" is played back. The input is voice data, and the output is voice playback.

[0493] This series of steps creates a system in which the user can ask questions by voice and receive answers by voice.

[0494] 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.

[0495] The system of the present invention enables new and junior employees to easily solve work-related questions and problems in a remote work environment. The system includes a means for a user to input a question by voice, a means for converting the voice question into text, a means for transmitting the text data to a server, a means for analyzing the text question and extracting keywords, a means for retrieving information from a database based on the keywords, a means for generating an answer using the retrieved information, a means for transmitting the generated answer to a terminal, a means for converting the text answer into voice, and a means for playing back the voice answer.

[0496] Furthermore, the present invention provides a more personalized response by combining an emotion engine that recognizes emotions from the user's voice. This emotion engine can recognize the user's emotional state and incorporate that information into the answer generation process.

[0497] First, the user speaks a work-related question into the device's microphone. For example, "Senior, where is the format for this report?" This spoken question is converted into text data by the device, and the emotion engine then analyzes the user's emotional state. The emotion engine analyzes the tone, pace, emphasis, etc. of the voice, and recognizes the user's emotional state, such as "I'm wondering" or "I'm confused."

[0498] The device then sends the converted text data and the recognized emotion data to the server. The server then passes the received text data to a natural language processing engine, which analyzes the question and extracts important keywords and phrases. For example, from the question "Where can I find the format for this report?", keywords such as "report," "format," and "where" are extracted.

[0499] The server searches the company database based on these keywords to retrieve relevant information, for example, "The report format can be found in the document template section of the company portal."

[0500] The server then uses the generative AI model to generate an appropriate response, taking into account the emotional data from the emotion engine. For example, if the server detects that the user is confused, the response will be more polite and detailed. The generated response is sent to the device in text format.

[0501] The device passes the received text response to a speech synthesis engine, which converts it into voice data. This voice data is then played through the device's speaker or earphones. For example, a polite voice might say, "The report format can be found in the document template section of the internal portal."

[0502] In this way, users can receive answers to their questions via voice, as if they were being answered directly by a senior employee. Furthermore, the emotion engine recognizes the user's emotional state and responds accordingly, providing more appropriate and personalized support. For example, if a user asks a work-related question, "Senior, where can I find the format for this report?", the system will go through the steps of voice conversion, emotion recognition, information acquisition, answer generation, and voice synthesis to provide the user with an appropriate answer via voice. This is particularly beneficial for new and junior employees in remote work environments, improving work efficiency and contributing to lower employee turnover.

[0503] The processing flow will be explained below.

[0504] Step 1:

[0505] The user speaks into the device's microphone to ask a work-related question. For example, "Senior, where can I find the format for this report?"

[0506] Step 2:

[0507] The device uses a speech recognition engine to convert the user's voice question into text. The voice data is analyzed and the corresponding text is generated. For example, "Where can I find the format for this report?"

[0508] Step 3:

[0509] While the device converts the voice question into text, it also uses an emotion engine to analyze the voice data and recognize the user's emotional state, for example by extracting emotions such as "doubtful" or "confused" from the tone and pace of the voice.

[0510] Step 4:

[0511] The device sends the converted text data and the recognized emotion data to the server in the form of an HTTP request. This request also includes the user's authentication token. Example: Send a text question, emotion data, and authentication token to the server.

[0512] Step 5:

[0513] The server passes the received text data to a natural language processing engine, which analyzes the question. The engine extracts keywords and important phrases from the text. For example, it extracts keywords such as "report," "format," and "where."

[0514] Step 6:

[0515] The server searches the company database based on the analysis results to retrieve relevant information. It generates a search query and queries the database. For example, it executes the query "SELECT FROM documentation_templates WHERE category="report format"" to retrieve relevant information.

[0516] Step 7:

[0517] The generative AI model creates a response based on the information obtained by the server. It adjusts the tone and level of detail of the response taking into account emotional data from the emotion engine. For example, if the user is perceived as "confused," the response will be more detailed and polite.

[0518] Step 8:

[0519] The server generates a text response and sends it to the device as an HTTP response. The server returns the text response to the device. Example: "The report format can be found in the document template section of the internal portal."

[0520] Step 9:

[0521] The text response received by the device is passed to the speech synthesis engine, which converts the text into speech data. The synthesized speech file is saved in temporary memory. For example, it is converted to "The report format can be found in the document template section of the internal portal."

[0522] Step 10:

[0523] The device outputs the audio data to a playback device (speaker or earphones). By listening to the answer, the user feels as if a senior employee had answered them directly. Example: "The report format can be found in the document template section of the internal portal."

[0524] Through this series of steps, users can receive prompt and personalized responses even in a remote work environment. In particular, the introduction of an emotion engine allows for appropriate responses based on the user's emotional state, which is expected to improve the work efficiency of new and junior employees and reduce their psychological burden.

[0525] Example 2

[0526] 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."

[0527] In recent years, with the rapid spread of remote work environments, new and junior employees are seeking ways to quickly and efficiently resolve work-related questions and problems. Conventional systems lack support for resolving work-related questions, particularly the lack of a method that combines voice input and emotion analysis. Furthermore, there is a lack of systems that can provide personalized responses that take into account the user's emotional state. This reduces the work efficiency of new and junior employees and leads to increased turnover, so a new system is needed to solve this issue.

[0528] 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.

[0529] In this invention, the server includes a means for analyzing a user's emotions, a means for reflecting the analyzed emotion data in the answer generation process, and a means for generating an appropriate answer using a generative AI model, thereby enabling the server to quickly provide appropriate and personalized answers to questions entered by the user via voice.

[0530] "User" refers to a person who uses the system to solve a business question or problem.

[0531] "Means for inputting questions by voice" refers to a device including a microphone and an interface that a user uses to input voice.

[0532] "Means for converting voice queries into text" refers to speech recognition software or algorithms used to convert user-entered voice data into text data.

[0533] "Means for transmitting text data to a server" refers to a communication means for sending text data to a server via a network.

[0534] "Means for analyzing text questions and extracting keywords" refers to natural language processing techniques used to analyze received text data and identify important words and phrases.

[0535] "Means for retrieving information from a database based on keywords" refers to means for searching a database using extracted keywords and retrieving related information.

[0536] "Means for generating an answer using acquired information" refers to the technology or algorithm used to create an appropriate answer based on the acquired information.

[0537] The "means for transmitting the generated answer to the terminal" refers to a communication means for sending the generated answer to the terminal via a network.

[0538] "Means for converting text responses to speech" refers to a speech synthesis engine or algorithm for converting the generated text responses into speech data.

[0539] The "means for playing back an audio response" refers to a device such as a speaker or earphone for playing back audio data and providing a response to the user.

[0540] "Means for analyzing user emotions" refers to an emotion engine or algorithm for analyzing the user's emotional state from their voice or text input.

[0541] "Means for incorporating analyzed emotional data into the answer generation process" refers to technologies and algorithms for adjusting answers to take into account the user's emotional state.

[0542] "Generative AI model" refers to an artificial intelligence model for generating text using natural language generation technology.

[0543] The system of the present invention allows new and junior employees to easily solve work-related questions and problems in a remote work environment. The system includes a means for users to input questions by voice, a means for converting the voice question into text, a means for transmitting the text data to a server, a means for analyzing the text question and extracting keywords, a means for retrieving information from a database based on the keywords, a means for generating an answer using the retrieved information, a means for transmitting the generated answer to a terminal, a means for converting the text answer into voice, and a means for playing back the voice answer. Furthermore, the present invention provides more personalized responses by combining a means for analyzing the user's emotions with a means for incorporating the analyzed emotional data into the answer generation process.

[0544] The user speaks a question about work into the device's microphone. The device used for this is a computer device such as a regular PC, smartphone, or tablet. For example, a user might ask, "Senior, where can I find the format for this report?" This spoken question is converted into text data using speech recognition software on the device (for example, Google Cloud Speech-to-Text).

[0545] The text data is then analyzed by an emotion engine (e.g., IBM Watson Tone Analyzer) to determine the user's emotional state. The emotion engine analyzes the tone, pace, and emphasis of the voice to recognize the user's emotional state (e.g., "I'm wondering," "I'm confused," etc.). The device then sends the converted text data and the recognized emotion data to the server.

[0546] The server passes the received text data to a natural language processing engine (e.g., spaCy) and analyzes the question. Important keywords and phrases are extracted, such as "report," "format," and "where." The server then searches a database (e.g., a MySQL database) based on these keywords to retrieve relevant information. For example, it retrieves the information, "The report format can be found in the document templates section of the internal portal."

[0547] The server then generates an appropriate response using a generative AI model (e.g., OpenAI's GPT-3) while taking into account the emotional data from the emotion engine. For example, if the server recognizes that the user is "confused," the response will be more polite and detailed. The generated response is sent to the device in text format.

[0548] The device passes the received text response to a speech synthesis engine (e.g., Amazon Polly) and converts it into voice data. This voice data is played back through the device's speaker or earphones. For example, it may say in a polite tone, "The report format can be found in the document template section of the internal portal."

[0549] A unique feature of this system is that it analyzes the user's emotions and uses that information to adjust the answer generation process, allowing users to receive personalized support as if they were being answered directly by a senior employee.

[0550] As a concrete example, if a user asks a work-related question such as "Senior, where can I find the format for this report?", the system will go through the steps of voice conversion, emotion recognition, information acquisition, answer generation, and voice synthesis to provide the user with an appropriate answer via voice.An example of a prompt sentence for the generative AI model is shown below.

[0551] Example prompt sentence:

[0552] User question: "Senior, where can I find the format for this report?"

[0553] Extracted keywords: "report", "format", "where"

[0554] Emotion recognized: "I'm confused"

[0555] Database lookup result: "The report format can be found in the Document Templates section of the Intranet Portal."

[0556] Generate an appropriate response considering the recognized emotion.

[0557] These prompts can be fed into a generative AI model to generate appropriate answers.

[0558] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0559] Step 1:

[0560] The user speaks into the microphone of the terminal to input a question about the business.

[0561] As a specific operation, when a user says, "Senior, where is the format for this report?", this speech is input.

[0562] Step 2:

[0563] The device uses voice recognition software such as Google Cloud Speech-to-Text to convert the user's voice data into text data.

[0564] Input: Voice data "Senior, where can I find the format for this report?"

[0565] Output: Text data "Senior, where can I find the format for this report?"

[0566] Specifically, speech recognition software analyzes the speech waveform and converts it into corresponding text.

[0567] Step 3:

[0568] The device then sends the converted text data to an emotion engine such as IBM Watson Tone Analyzer to analyze the user's emotional state.

[0569] Input: Text data "Senior, where can I find the format for this report?"

[0570] Output: Emotion data "Confused"

[0571] Specifically, the emotion engine analyzes the tone, pace, and emphasis of the voice to recognize the user's emotions.

[0572] Step 4:

[0573] The terminal transmits the converted text data and the recognized emotion data to the server.

[0574] Input: Text data "Senior, where is the format for this report?", Emotion data "Confused"

[0575] Output: Data sent to the server

[0576] As a specific operation, data is transmitted from the terminal to the server via the network.

[0577] Step 5:

[0578] The server passes the received text data to a natural language processing engine (e.g., spaCy) to analyze the question content and extract keywords.

[0579] Input: Text data "Senior, where can I find the format for this report?"

[0580] Output: Keywords "report", "format", "where"

[0581] Specifically, a natural language processing engine analyzes the text and identifies important words and phrases.

[0582] Step 6:

[0583] Based on the extracted keywords, the server searches internal databases such as MySQL databases to obtain relevant information.

[0584] Input: Keywords "report", "format", "where"

[0585] Output: Retrieved information: "The report format can be found in the document template section of the intranet."

[0586] Specifically, a database query is performed to retrieve relevant information.

[0587] Step 7:

[0588] The server takes into account the acquired information and emotional data and generates an appropriate answer using a generative AI model such as OpenAI's GPT-3.

[0589] Input: Information retrieved from the database: "The report format can be found in the document template section of the internal portal.", Emotion data: "Confused"

[0590] Output: Generated answer "The report format can be found in the document templates section of the internal portal. Please contact us if you require more information."

[0591] Specifically, the generative AI model receives a prompt and generates a natural-sounding response based on the context.

[0592] Step 8:

[0593] The generated text response is sent to the device, which converts it into voice data using a speech synthesis engine such as Amazon Polly.

[0594] Input: Text response "The report format can be found in the document templates section of the intranet. Please contact us if you require more information."

[0595] Output: Audio data

[0596] Specifically, the speech synthesis engine converts the text into speech and generates speech data.

[0597] Step 9:

[0598] The terminal plays back the generated voice data and provides the answer to the user.

[0599] Input: Audio data

[0600] Output: The user hears the audio response

[0601] Specifically, the audio data is played through the terminal's speaker or earphones.

[0602] (Application example 2)

[0603] 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."

[0604] It is difficult for workers on factory production lines to receive prompt and appropriate support when they have questions or problems related to their work. New or junior employees in particular need real-time assistance when they have questions or are confused about their work. Traditional methods require direct interaction with superiors or senior employees, which can lead to lower productivity. Furthermore, responses that do not take into account the worker's emotional state run the risk of increasing stress. Therefore, personalized support for workers is needed to improve work efficiency within factories.

[0605] 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.

[0606] In this invention, the server includes means for a user to input a question by voice, means for converting the voice question into text, means for transmitting the text data to the server, means for analyzing the text question and extracting keywords, means for retrieving information from a database based on the keywords, means for generating an answer using the retrieved information, means for transmitting the generated answer to a terminal, means for converting the text answer into voice, means for playing back the voice answer, and means for analyzing emotions. This allows factory workers to solve their questions and problems in real time and receive appropriate support that takes their emotional state into consideration.

[0607] "Users" refer to factory workers who use the system.

[0608] "Means for inputting questions by voice" refers to a device or method that allows a worker to input questions by voice using a microphone or the like.

[0609] "Means for converting voice queries into text" refers to a device or method that uses voice recognition technology to convert an input voice query into text data.

[0610] The "means for transmitting text data to a server" refers to a device or method for transmitting the converted text data to a server via a network.

[0611] "Means for analyzing text questions and extracting keywords" refers to a device or method that uses natural language processing techniques to extract important keywords from text questions.

[0612] The term "means for retrieving information from a database based on keywords" refers to a device or method that uses the extracted keywords to search for and retrieve related information from a database.

[0613] "Means for generating an answer using acquired information" refers to a device or method that generates an appropriate answer based on information acquired from a database.

[0614] The "means for transmitting the generated answer to the terminal" refers to a device or method for transmitting the generated answer to the user's terminal via a network.

[0615] "Means for converting text responses to speech" refers to a device or method that uses speech synthesis technology to convert generated text responses into speech data.

[0616] "Means for playing back audio responses" refers to a device or method for playing back converted audio data through a speaker or earphones.

[0617] "Emotion analysis means" refers to a device or method that analyzes the emotional state of a worker from voice or text and personalizes assistance based on that information.

[0618] The system of the present invention helps factory production line workers to quickly resolve work-related questions and problems remotely. The system's main components combine functions such as voice input, text conversion, emotion analysis, data transmission, information acquisition, answer generation, and voice playback.

[0619] In order to implement the present invention, the following hardware and software are used.

[0620] 1. Voice input

[0621] Hardware: Microphone

[0622] Software: Google Cloud Speech-to-Text API

[0623] Description: A user speaks into a microphone to ask a question about a task. For example, a question could be, "What is the maintenance procedure for this machine?"

[0624] 2. Text Conversion

[0625] Hardware: Devices (smartphones, tablets, PCs, etc.)

[0626] Software: Google Cloud Speech-to-Text API

[0627] Description: The input speech is converted to text data. The converted text is "What is the maintenance procedure for this machine?"

[0628] 3. Emotion analysis

[0629] Hardware: Built-in microphone

[0630] Software: Microsoft Azure Emotion API

[0631] Description: Analyzes a user's emotional state from speech and text. For example, recognizes the emotion "confused" from a user's voice.

[0632] 4. Data Transmission

[0633] Hardware: Terminal, Wi-Fi module

[0634] Software: HTTPS communication

[0635] Description: Sends the converted text data and analyzed emotion data to the server.

[0636] 5. Information acquisition

[0637] Hardware: Server

[0638] Software: Natural language processing engine, database search algorithm

[0639] Description: The server extracts keywords from the received text data and searches the database for related information based on these keywords. For example, it retrieves related procedure manuals based on the keyword "machine maintenance procedures."

[0640] 6. Answer generation

[0641] Hardware: Server

[0642] Software: Generative AI model (OpenAI GPT-4)

[0643] Description: Generate an appropriate answer based on the acquired information and emotion data. The generated answer will be in text format, such as "Details of the machine maintenance procedure can be found on page 5 of the procedure manual."

[0644] 7. Audio playback

[0645] Hardware: Device, speakers or earphones

[0646] Software: Google Cloud Text-to-Speech API

[0647] Description: The generated text response is converted to audio data and played back. The user hears, "The machine maintenance procedure is detailed on page 5 of the procedure manual."

[0648] As a result, workers feel as if they are being directly answered by a senior employee, allowing them to solve problems quickly and accurately.Furthermore, emotion analysis makes it possible to provide personalized responses based on the user's emotions, which is expected to improve work efficiency and reduce stress.

[0649] For example, if a user asks, "What is the maintenance procedure for this machine?", the system goes through the steps of voice conversion, emotion analysis, information acquisition, answer generation, and voice synthesis, and provides a spoken answer such as, "Details of the machine's maintenance procedure are on page 5 of the instruction manual."

[0650] An example of a prompt sentence is as follows:

[0651] Question: "What is the maintenance procedure for this machine?" Emotion: "I'm confused."

[0652] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0653] Step 1:

[0654] Users speak into a microphone to ask questions about their work. This voice becomes input to the system. For example, they might ask, "What is the maintenance procedure for this machine?"

[0655] Step 2:

[0656] The device receives voice input and converts it to text using the Google Cloud Speech-to-Text API. The converted text becomes the input for the next step in the system. The output is the text, "What are the maintenance procedures for this machine?"

[0657] Step 3:

[0658] The device then passes the converted text data to the Microsoft Azure Emotion API for emotion analysis. The analysis identifies the emotion the user expressed while asking the question. For example, emotional information such as "confused" is output. This emotional data is then used as input for the next step.

[0659] Step 4:

[0660] The device sends text data and emotion data to the server. The sent data is analyzed by the server's natural language processing engine, and important keywords are extracted. For example, from the question "What is the maintenance procedure for this machine?" keywords such as "machine," "maintenance," and "procedure" are extracted.

[0661] Step 5:

[0662] The server retrieves related information from the database based on the extracted keywords. For example, it searches using the keywords "machine," "maintenance," and "procedure" to retrieve procedure manuals and related technical documents. This retrieved information becomes the input for the next step.

[0663] Step 6:

[0664] The server generates an appropriate answer using a generative AI model (e.g., OpenAI GPT-4) based on the acquired information and emotion data. The generated answer is output as text data. For example, a response such as "Details of the machine maintenance procedure are on page 5 of the instruction manual" may be generated.

[0665] Step 7:

[0666] The server sends the generated text response to the device. The device passes it to the Google Cloud Text-to-Speech API and converts it into voice data. The converter outputs the voice data, "Details of the machine maintenance procedure can be found on page 5 of the procedure manual."

[0667] Step 8:

[0668] The terminal plays the generated voice data through a speaker or earphone, allowing the user to receive a voice response in real time. Specifically, the speaker will say, "Details of the machine maintenance procedure can be found on page 5 of the procedure manual."

[0669] 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.

[0670] 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.

[0671] 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.

[0672] [Third embodiment]

[0673] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0674] 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.

[0675] 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).

[0676] 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.

[0677] 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.

[0678] 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).

[0679] 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.

[0680] 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.

[0681] 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.

[0682] 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.

[0683] 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.

[0684] 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."

[0685] The system of the present invention enables new and junior employees to easily solve work-related questions and problems in a remote work environment. The system includes a means for a user to input a question by voice, a means for converting the voice question into text, a means for transmitting the text data to a server, a means for analyzing the text question and extracting keywords, a means for retrieving information from a database based on the keywords, a means for generating an answer using the retrieved information, a means for transmitting the generated answer to a terminal, a means for converting the text answer into voice, and a means for playing back the voice answer.

[0686] First, the user speaks a question about work into the device's microphone. This spoken question is converted into text data by the device's voice recognition engine. For example, if a user asks, "Senior, where can I find the format for this report?", the voice is converted into the text "Where can I find the format for this report?"

[0687] The device then sends the converted text data to the server as an HTTP request, which also includes user authentication information and is transmitted over a secure communication channel.

[0688] The server passes the received text data to a natural language processing engine, which analyzes the question. Specifically, it extracts important keywords and phrases from the question. In this example, keywords such as "report," "format," and "where" are extracted. The server uses these keywords to search the company database to obtain relevant information. For example, it obtains information such as "The report format can be found in the document template section of the company portal."

[0689] Based on the acquired information, the generative AI model on the server generates an appropriate answer. This answer is then formatted using natural language generation technology into a format that is easy for the user to understand. The generated answer is then sent to the device in text format.

[0690] The device passes the received text response to a speech synthesis engine, which converts it into voice data. Finally, the converted voice data is played back through the device's speaker or earphones. This allows the user to hear the answer to their question in voice, as if it were being answered directly by a senior employee.

[0691] For example, if a user asks a work-related question, "Senior, where can I find the format for this report?", the process is as follows: The device converts this speech into text and sends it to the server. The server searches the database based on the analysis and generates the answer, "The report format can be found in the document template section of the in-house portal." The generated answer is converted into speech and played back to the user via the device.

[0692] This system utilizes natural language processing and generation technologies to provide timely answers in a format that is easy for users to understand. It also allows new and junior employees to receive support as if they were being supported by a senior employee, even in a remote work environment, improving the work efficiency of new and junior employees and contributing to a lower turnover rate.

[0693] The processing flow will be explained below.

[0694] Step 1:

[0695] The user speaks into the device's microphone to ask a work-related question. For example, "Senior, where can I find the format for this report?"

[0696] Step 2:

[0697] The device uses a speech recognition engine to convert the user's spoken question into text. The speech data is analyzed and the corresponding text is generated. For example, the speech is converted to "Where can I find the format for this report?"

[0698] Step 3:

[0699] The device sends the converted text data to the server in the form of an HTTP request, which also includes the user's authentication token. For example, a text question and the authentication token are sent.

[0700] Step 4:

[0701] The server passes the received text data to a natural language processing engine, which analyzes the question. The engine extracts keywords and important phrases from the text. For example, it extracts keywords such as "report," "format," and "where."

[0702] Step 5:

[0703] The server searches the company database based on the analysis results to retrieve relevant information. It generates a search query and queries the database. For example, it executes the query "SELECT FROM documentation_templates WHERE category="report format"" and retrieves the results.

[0704] Step 6:

[0705] The generative AI model creates an answer based on the information obtained by the server. The answer is formatted using natural language generation technology in a way that is easy for the user to understand. For example, "The report format can be found in the document template section of the internal portal."

[0706] Step 7:

[0707] The server generates a text response and sends it to the terminal as an HTTP response. For example, the response text "The report format can be found in the document template section of the internal portal" is sent.

[0708] Step 8:

[0709] The text response received by the device is passed to the speech synthesis engine, which converts the text into speech data. The synthesized speech file is saved in temporary memory. Example: "The report format can be found in the document template section of the in-house portal" is converted into speech data.

[0710] Step 9:

[0711] The device outputs the audio data to a playback device (speaker or earphones). By listening to the answer, the user feels as if a senior employee had answered them directly. Example: Play the audio "The report format can be found in the document template section of the internal portal."

[0712] By following these steps, users can quickly get answers to their business-related questions even in a remote work environment.

[0713] Example 1

[0714] 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."

[0715] In a remote work environment, new and junior employees lack the means to easily resolve work-related questions and problems. This can lead to reduced work efficiency and delayed employee growth. There are also concerns that the difficulty of communication in a remote environment could lead to increased employee turnover.

[0716] 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.

[0717] In this invention, the server includes means for allowing a user to input a question by voice, means for converting the voice question into text, means for transmitting the text data to an information processing device, means for analyzing the text question and extracting keywords, means for acquiring information from an information storage device based on the keywords, means for generating an answer using the acquired information, means for transmitting the generated answer to a display device, means for converting the text answer into voice, means for playing back the voice answer, means for generating an answer in a format that is easy for the user to understand using natural language generation technology, and means for ensuring secure communication using HTTPS. This enables smooth communication even in a remote work environment, allowing new and junior employees to quickly solve problems, which is expected to improve work efficiency and reduce turnover.

[0718] A "user" is a person who uses the system to input questions by voice and receive answers.

[0719] A "means for inputting a question by voice" is a microphone or other voice input device that allows a user to input a voice question into the system.

[0720] A "means for converting voice queries into text" is a device or software that includes speech recognition technology or algorithms that analyzes input voice data and converts it into text data.

[0721] The "means for transmitting text data to the information processing device" refers to a communication function or protocol for transferring text data to a server via a network.

[0722] An "information processing device" is a computer system or server that analyzes text data and generates answers.

[0723] "Means for analyzing text questions and extracting keywords" refers to natural language processing techniques and algorithms that identify and extract important keywords from received text data.

[0724] "Means for retrieving information from an information storage device based on keywords" refers to a technique for using extracted keywords to search for and retrieve related information from databases, knowledge bases, and other information storage devices.

[0725] An "information storage device" is a storage device for storing databases and other digital information.

[0726] The "means for generating a response using acquired information" refers to a device or software that generates a response sentence using natural language generation technology based on the acquired information.

[0727] The "means for transmitting the generated answer to the display device" refers to a communication function or protocol for transferring the generated answer to the user's terminal via a network.

[0728] The "display device" refers to a display, speaker, or earphone for displaying or playing back the answer to the user.

[0729] "Means for converting text responses to speech" refers to speech synthesis techniques and algorithms that convert text responses into speech data.

[0730] The "means for playing back audio responses" refers to a device or system for playing back audio data through a speaker or earphones.

[0731] "Natural language generation technology" refers to the technology and algorithms used to analyze text data and generate natural-sounding sentences that are easy for humans to understand.

[0732] "Means of ensuring secure communications through HTTPS" refers to protocols and technologies that ensure the confidentiality and integrity of data by encrypting and sending and receiving communication data.

[0733] A "prompt" is text containing specific instructions or questions that a generative AI model uses to generate appropriate answers.

[0734] The system of the present invention allows new and junior employees to easily resolve work-related questions and problems in a remote work environment. This system is realized using the following hardware and software.

[0735] The user speaks a question about work into the device's microphone. For example, they might ask, "Senior, where can I find the format for this report?" This spoken question is converted into text data using the device's built-in speech recognition engine (e.g., Google Speech-to-Text API). The converted text data is then sent from the device to the server via secure communication using HTTPS.

[0736] The server passes the received text data to a natural language processing engine (e.g., SpaCy or BERT) and analyzes the question. Specifically, it tokenizes the text and extracts keywords such as "report," "format," and "where." Next, the server searches an information storage device (database or knowledge base) based on the extracted keywords to obtain related information. For example, it can obtain information such as "The report format can be found in the document template section of the internal portal."

[0737] Based on the acquired information, an appropriate answer is generated using a generative AI model on the server (e.g., OpenAI's GPT-3). At this time, the prompt text is "Please generate an answer that is easy for the user to understand based on the information, 'The report format can be found in the document template section of the internal portal.'" The generative AI model generates a natural and easy-to-understand answer based on this prompt text.

[0738] The generated answer is sent to the device in text format. The device then passes the received text answer to a speech synthesis engine (e.g., Amazon Polly or Google Text-to-Speech) and converts it into audio data. Finally, the device plays the converted audio data through a speaker or earphones. This process allows the user to hear the answer to their question in audio, as if it were being answered directly by a senior employee.

[0739] This system's unique feature is its use of natural language processing and natural language generation technologies, allowing it to quickly provide answers in a format that is easy for users to understand. It also uses the HTTPS protocol, which ensures secure communications, ensuring high communication security. By using this system, new and junior employees can communicate smoothly even in a remote work environment, which is expected to improve work efficiency and reduce turnover.

[0740] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0741] Step 1:

[0742] The user speaks a question into the microphone of the device. Specifically, the user speaks, "Senior, where is the format for this report?" The input of this step is the user's voice data, and the output is also voice data.

[0743] Step 2:

[0744] The device passes the voice data acquired from the microphone to a voice recognition engine. Specifically, a voice recognition engine (e.g., Google Speech-to-Text API) is used to convert the voice data into text data. The input of this step is the voice data, and the output is the text data, "Where can I find the format for this report?"

[0745] Step 3:

[0746] The terminal sends the converted text data to the server as an HTTPS request. Specifically, it generates a request including the text data and user authentication information and sends it to the server via a communication channel encrypted by SSL / TLS. The input of this step is the text data and authentication information, and the output is an HTTPS request to the server.

[0747] Step 4:

[0748] The server passes the received text data to a natural language processing engine. Specifically, it uses a natural language processing engine (e.g., SpaCy or BERT) to analyze the text and extract important keywords. For example, it extracts keywords such as "report," "format," and "where." The input of this step is the received text data, and the output is the extracted keywords.

[0749] Step 5:

[0750] The server searches the information storage device based on the extracted keywords. Specifically, it generates an SQL query to search a database (e.g., MySQL or PostgreSQL) and retrieves related information. For example, it obtains information such as "The report format is in the document template section of the in-house portal." The input of this step is the extracted keywords, and the output is the retrieved information.

[0751] Step 6:

[0752] The server uses a generative AI model to generate an answer based on the acquired information. Specifically, the prompt "Generate an answer that is easy for the user to understand based on the information 'The report format can be found in the document template section of the internal portal'" is input into a generative AI model (e.g., OpenAI's GPT-3) to generate a natural and easy-to-understand answer. The input for this step is the acquired information and the prompt, and the output is the generated answer in text format.

[0753] Step 7:

[0754] The server sends the generated text answer to the device using HTTPS. Specifically, the text answer is sent over a communication channel that is again encrypted with SSL / TLS. The input to this step is the generated text answer, and the output is an HTTPS response to the device.

[0755] Step 8:

[0756] The device passes the received text response to a speech synthesis engine (e.g., Amazon Polly or Google Text-to-Speech) to convert the text to speech data. The input of this step is the received text response, and the output is speech data.

[0757] Step 9:

[0758] The terminal plays the audio data through a speaker or earphone. Specifically, using the terminal's audio output device, the user hears the audio response as if the senior employee were speaking directly to them. The input for this step is the audio data, and the output is the audio response provided to the user.

[0759] By following these steps, users can get quick and accurate answers even in a remote work environment.

[0760] (Application example 1)

[0761] 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."

[0762] Factory floors require a means for operators to quickly and accurately give instructions and ask questions to robots. However, existing systems mainly use text input, and few support voice input. This prevents operators from responding quickly, resulting in reduced production efficiency. Furthermore, even if systems exist that support voice input, they lack natural language processing and generation technologies, making it difficult to provide accurate answers and instructions. There is a need to provide a system that can solve these issues and enable operators to communicate smoothly with robots.

[0763] 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.

[0764] In this invention, the server includes means for a user to input a question by voice, means for converting the voice question into text, means for transmitting the text data to the server, means for analyzing the text question and extracting keywords, means for retrieving information from a database based on the keywords, means for generating an answer using the retrieved information, means for transmitting the generated answer to a terminal, means for converting the text answer into voice, means for playing back the voice answer, and means for a user to give instructions or questions to a robot in a factory by voice, and for the robot to analyze the voice and return an appropriate answer or instruction by voice. This makes it possible for an operator to give instructions or questions to the robot by voice, thereby quickly and accurately acquiring information or issuing operating instructions.

[0765] "Voice input" is a means of capturing and processing the user's voice into an electronic device.

[0766] "Text conversion" is the process of converting information captured through speech into written data.

[0767] "Server transmission" refers to the act of transmitting the converted character data to the central processing unit.

[0768] "Text analysis" is a technology that extracts important keywords from transmitted text data and analyzes its meaning.

[0769] "Database acquisition" refers to searching and acquiring related information from a database based on the analysis results.

[0770] "Answer generation" is the process of creating an appropriate answer to a user's question based on the acquired information.

[0771] "Terminal transmission" is the act of transmitting the generated answer to a terminal accessible by the user.

[0772] "Speech conversion" is a technology that converts answers from text data into voice data.

[0773] "Audio playback" refers to the act of playing audio data through a speaker or the like and providing it to the user.

[0774] A "factory floor" is a facility where various production activities take place and where employees and robots work.

[0775] A "robot" is a machine that performs programmed actions and automates tasks.

[0776] This invention is a system that supports effective communication between operators and robots in factories. This system is composed of means for speech recognition, text conversion, text analysis, database search, answer generation, speech synthesis, and speech playback.

[0777] First, the user, an operator, gives instructions or asks questions to the robot by voice. To achieve this, the robot or its connected device is equipped with a microphone. For example, when the operator says, "Robot, how do I maintain this machine?", the voice data is captured.

[0778] Next, the terminal converts the acquired voice data into text data using a voice recognition engine (for example, a SpeechRecognition library), and sends the converted text data to a central processing unit (server) as an HTTP request.

[0779] Once the text data arrives at the server, it is analyzed by a natural language processing engine (for example, Hugging Face's Transformers library) to extract important keywords. In this example, keywords such as "machine," "maintenance," and "method" are extracted.

[0780] Next, the server uses these keywords to search for and retrieve relevant information from a database. The database stores various manuals and work instructions for factory machines. For example, the server might retrieve information such as, "To maintain a machine, first turn off the power and remove the safety devices."

[0781] Based on the acquired information, the server uses a generative AI model to generate an appropriate answer. This generative AI model incorporates technology to generate answers in a format that is easy for the user to understand. The generated answer is sent to the device in text format.

[0782] Finally, the terminal passes the received text response to a speech synthesis engine (for example, the Pyttsx3 library) and converts it into voice data. This voice data is then played back to the operator through a speaker, allowing the user to hear the answer to their question aloud, as if the robot were answering them directly.

[0783] For example, if an operator asks, "Robot, what's the next step?", the system will respond, "The next step is to install part A." Or, if the operator asks, "Robot, how do you maintain this machine?", the system will respond, "The way to maintain this machine is to first turn off the power and remove the safety devices."

[0784] An example of a prompt sentence to input to the generative AI model is:

[0785] "Robot, how do you maintain this machine?"

[0786] "Robot, what's the next step?"

[0787] Let's say.

[0788] In this way, a system can be realized that improves communication between operators and robots in factories and increases work efficiency.

[0789] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0790] Step 1:

[0791] The user enters a question by voice.

[0792] Specifically, the operator speaks into the microphone and asks a question, for example, "Robot, how do you maintain this machine?" The input here is voice data.

[0793] Step 2:

[0794] The terminal sends the voice data to a voice recognition engine and converts it into text data.

[0795] Specifically, the voice data acquired by the device's microphone is converted into text data using the SpeechRecognition library, resulting in the text data "Robot, how do you maintain this machine?"

[0796] Step 3:

[0797] The terminal transmits the text data to the server.

[0798] Specifically, the converted text data is sent to the server as an HTTP request. This request also includes user authentication information. The input is text data, and the output is a request to the server.

[0799] Step 4:

[0800] The server passes the text data to a natural language processing engine to extract keywords.

[0801] Specifically, the server analyzes the received text data using Hugging Face's Transformers library and extracts the keywords "machine," "maintenance," and "method." The input is text data, and the output is keyword data.

[0802] Step 5:

[0803] The server searches the database based on the keywords to retrieve relevant information.

[0804] Specifically, the server uses the extracted keywords to search a database containing manuals and work instructions for factory machines. The information obtained is, "To maintain the machine, first turn off the power and remove the safety devices." The input is keyword data, and the output is the acquired information.

[0805] Step 6:

[0806] Based on the information acquired by the server, an appropriate answer is generated using a generative AI model.

[0807] Specifically, the server uses the acquired information to send a prompt to a generative AI model (e.g., GPT-3) to generate an answer such as "First, turn off the power and remove the safety device." The input is the acquired information, and the output is the generated answer text.

[0808] Step 7:

[0809] The server sends the generated response to the terminal.

[0810] Specifically, the generated answer text is sent from the server to the terminal as an HTTP response. The input is the generated answer text, and the output is the response to the terminal.

[0811] Step 8:

[0812] The answer received by the terminal is passed to a speech synthesis engine and converted into voice data.

[0813] Specifically, the device passes the received text response to the Pyttsx3 library and converts it into audio data. The input is the text response and the output is audio data.

[0814] Step 9:

[0815] The terminal plays the generated audio data to the user through a speaker or earphone.

[0816] Specifically, a voice message saying, "First, turn off the power and remove the safety device" is played back. The input is voice data, and the output is voice playback.

[0817] This series of steps creates a system in which the user can ask questions by voice and receive answers by voice.

[0818] 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.

[0819] The system of the present invention enables new and junior employees to easily solve work-related questions and problems in a remote work environment. The system includes a means for a user to input a question by voice, a means for converting the voice question into text, a means for transmitting the text data to a server, a means for analyzing the text question and extracting keywords, a means for retrieving information from a database based on the keywords, a means for generating an answer using the retrieved information, a means for transmitting the generated answer to a terminal, a means for converting the text answer into voice, and a means for playing back the voice answer.

[0820] Furthermore, the present invention provides a more personalized response by combining an emotion engine that recognizes emotions from the user's voice. This emotion engine can recognize the user's emotional state and incorporate that information into the answer generation process.

[0821] First, the user speaks a work-related question into the device's microphone. For example, "Senior, where is the format for this report?" This spoken question is converted into text data by the device, and the emotion engine then analyzes the user's emotional state. The emotion engine analyzes the tone, pace, emphasis, etc. of the voice, and recognizes the user's emotional state, such as "I'm wondering" or "I'm confused."

[0822] The device then sends the converted text data and the recognized emotion data to the server. The server then passes the received text data to a natural language processing engine, which analyzes the question and extracts important keywords and phrases. For example, from the question "Where can I find the format for this report?", keywords such as "report," "format," and "where" are extracted.

[0823] The server searches the company database based on these keywords to retrieve relevant information, for example, "The report format can be found in the document template section of the company portal."

[0824] The server then uses the generative AI model to generate an appropriate response, taking into account the emotional data from the emotion engine. For example, if the server detects that the user is confused, the response will be more polite and detailed. The generated response is sent to the device in text format.

[0825] The device passes the received text response to a speech synthesis engine, which converts it into voice data. This voice data is then played through the device's speaker or earphones. For example, a polite voice might say, "The report format can be found in the document template section of the internal portal."

[0826] In this way, users can receive answers to their questions via voice, as if they were being answered directly by a senior employee. Furthermore, the emotion engine recognizes the user's emotional state and responds accordingly, providing more appropriate and personalized support. For example, if a user asks a work-related question, "Senior, where can I find the format for this report?", the system will go through the steps of voice conversion, emotion recognition, information acquisition, answer generation, and voice synthesis to provide the user with an appropriate answer via voice. This is particularly beneficial for new and junior employees in remote work environments, improving work efficiency and contributing to lower employee turnover.

[0827] The processing flow will be explained below.

[0828] Step 1:

[0829] The user speaks into the device's microphone to ask a work-related question. For example, "Senior, where can I find the format for this report?"

[0830] Step 2:

[0831] The device uses a speech recognition engine to convert the user's voice question into text. The voice data is analyzed and the corresponding text is generated. For example, "Where can I find the format for this report?"

[0832] Step 3:

[0833] While the device converts the voice question into text, it also uses an emotion engine to analyze the voice data and recognize the user's emotional state, for example by extracting emotions such as "doubtful" or "confused" from the tone and pace of the voice.

[0834] Step 4:

[0835] The device sends the converted text data and the recognized emotion data to the server in the form of an HTTP request. This request also includes the user's authentication token. Example: Send a text question, emotion data, and authentication token to the server.

[0836] Step 5:

[0837] The server passes the received text data to a natural language processing engine, which analyzes the question. The engine extracts keywords and important phrases from the text. For example, it extracts keywords such as "report," "format," and "where."

[0838] Step 6:

[0839] The server searches the company database based on the analysis results to retrieve relevant information. It generates a search query and queries the database. For example, it executes the query "SELECT FROM documentation_templates WHERE category="report format"" to retrieve relevant information.

[0840] Step 7:

[0841] The generative AI model creates a response based on the information obtained by the server. It adjusts the tone and level of detail of the response taking into account emotional data from the emotion engine. For example, if the user is perceived as "confused," the response will be more detailed and polite.

[0842] Step 8:

[0843] The server generates a text response and sends it to the device as an HTTP response. The server returns the text response to the device. Example: "The report format can be found in the document template section of the internal portal."

[0844] Step 9:

[0845] The text response received by the device is passed to the speech synthesis engine, which converts the text into speech data. The synthesized speech file is saved in temporary memory. For example, it is converted to "The report format can be found in the document template section of the internal portal."

[0846] Step 10:

[0847] The device outputs the audio data to a playback device (speaker or earphones). By listening to the answer, the user feels as if a senior employee had answered them directly. Example: "The report format can be found in the document template section of the internal portal."

[0848] Through this series of steps, users can receive prompt and personalized responses even in a remote work environment. In particular, the introduction of an emotion engine allows for appropriate responses based on the user's emotional state, which is expected to improve the work efficiency of new and junior employees and reduce their psychological burden.

[0849] Example 2

[0850] 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."

[0851] In recent years, with the rapid spread of remote work environments, new and junior employees are seeking ways to quickly and efficiently resolve work-related questions and problems. Conventional systems lack support for resolving work-related questions, particularly the lack of a method that combines voice input and emotion analysis. Furthermore, there is a lack of systems that can provide personalized responses that take into account the user's emotional state. This reduces the work efficiency of new and junior employees and leads to increased turnover, so a new system is needed to solve this issue.

[0852] 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.

[0853] In this invention, the server includes a means for analyzing a user's emotions, a means for reflecting the analyzed emotion data in the answer generation process, and a means for generating an appropriate answer using a generative AI model, thereby enabling the server to quickly provide appropriate and personalized answers to questions entered by the user via voice.

[0854] "User" refers to a person who uses the system to solve a business question or problem.

[0855] "Means for inputting questions by voice" refers to a device including a microphone and an interface that a user uses to input voice.

[0856] "Means for converting voice queries into text" refers to speech recognition software or algorithms used to convert user-entered voice data into text data.

[0857] "Means for transmitting text data to a server" refers to a communication means for sending text data to a server via a network.

[0858] "Means for analyzing text questions and extracting keywords" refers to natural language processing techniques used to analyze received text data and identify important words and phrases.

[0859] "Means for retrieving information from a database based on keywords" refers to means for searching a database using extracted keywords and retrieving related information.

[0860] "Means for generating an answer using acquired information" refers to the technology or algorithm used to create an appropriate answer based on the acquired information.

[0861] The "means for transmitting the generated answer to the terminal" refers to a communication means for sending the generated answer to the terminal via a network.

[0862] "Means for converting text responses to speech" refers to a speech synthesis engine or algorithm for converting the generated text responses into speech data.

[0863] The "means for playing back an audio response" refers to a device such as a speaker or earphone for playing back audio data and providing a response to the user.

[0864] "Means for analyzing user emotions" refers to an emotion engine or algorithm for analyzing the user's emotional state from their voice or text input.

[0865] "Means for incorporating analyzed emotional data into the answer generation process" refers to technologies and algorithms for adjusting answers to take into account the user's emotional state.

[0866] "Generative AI model" refers to an artificial intelligence model for generating text using natural language generation technology.

[0867] The system of the present invention allows new and junior employees to easily solve work-related questions and problems in a remote work environment. The system includes a means for users to input questions by voice, a means for converting the voice question into text, a means for transmitting the text data to a server, a means for analyzing the text question and extracting keywords, a means for retrieving information from a database based on the keywords, a means for generating an answer using the retrieved information, a means for transmitting the generated answer to a terminal, a means for converting the text answer into voice, and a means for playing back the voice answer. Furthermore, the present invention provides more personalized responses by combining a means for analyzing the user's emotions with a means for incorporating the analyzed emotional data into the answer generation process.

[0868] The user speaks a question about work into the device's microphone. The device used for this is a computer device such as a regular PC, smartphone, or tablet. For example, a user might ask, "Senior, where can I find the format for this report?" This spoken question is converted into text data using speech recognition software on the device (for example, Google Cloud Speech-to-Text).

[0869] The text data is then analyzed by an emotion engine (e.g., IBM Watson Tone Analyzer) to determine the user's emotional state. The emotion engine analyzes the tone, pace, and emphasis of the voice to recognize the user's emotional state (e.g., "I'm wondering," "I'm confused," etc.). The device then sends the converted text data and the recognized emotion data to the server.

[0870] The server passes the received text data to a natural language processing engine (e.g., spaCy) and analyzes the question. Important keywords and phrases are extracted, such as "report," "format," and "where." The server then searches a database (e.g., a MySQL database) based on these keywords to retrieve relevant information. For example, it retrieves the information, "The report format can be found in the document templates section of the internal portal."

[0871] The server then generates an appropriate response using a generative AI model (e.g., OpenAI's GPT-3) while taking into account the emotional data from the emotion engine. For example, if the server recognizes that the user is "confused," the response will be more polite and detailed. The generated response is sent to the device in text format.

[0872] The device passes the received text response to a speech synthesis engine (e.g., Amazon Polly) and converts it into voice data. This voice data is played back through the device's speaker or earphones. For example, it may say in a polite tone, "The report format can be found in the document template section of the internal portal."

[0873] A unique feature of this system is that it analyzes the user's emotions and uses that information to adjust the answer generation process, allowing users to receive personalized support as if they were being answered directly by a senior employee.

[0874] As a concrete example, if a user asks a work-related question such as "Senior, where can I find the format for this report?", the system will go through the steps of voice conversion, emotion recognition, information acquisition, answer generation, and voice synthesis to provide the user with an appropriate answer via voice.An example of a prompt sentence for the generative AI model is shown below.

[0875] Example prompt sentence:

[0876] User question: "Senior, where can I find the format for this report?"

[0877] Extracted keywords: "report", "format", "where"

[0878] Emotion recognized: "I'm confused"

[0879] Database lookup result: "The report format can be found in the Document Templates section of the Intranet Portal."

[0880] Generate an appropriate response considering the recognized emotion.

[0881] These prompts can be fed into a generative AI model to generate appropriate answers.

[0882] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0883] Step 1:

[0884] The user speaks into the microphone of the terminal to input a question about the business.

[0885] As a specific operation, when a user says, "Senior, where is the format for this report?", this speech is input.

[0886] Step 2:

[0887] The device uses voice recognition software such as Google Cloud Speech-to-Text to convert the user's voice data into text data.

[0888] Input: Voice data "Senior, where can I find the format for this report?"

[0889] Output: Text data "Senior, where can I find the format for this report?"

[0890] Specifically, speech recognition software analyzes the speech waveform and converts it into corresponding text.

[0891] Step 3:

[0892] The device then sends the converted text data to an emotion engine such as IBM Watson Tone Analyzer to analyze the user's emotional state.

[0893] Input: Text data "Senior, where can I find the format for this report?"

[0894] Output: Emotion data "Confused"

[0895] Specifically, the emotion engine analyzes the tone, pace, and emphasis of the voice to recognize the user's emotions.

[0896] Step 4:

[0897] The terminal transmits the converted text data and the recognized emotion data to the server.

[0898] Input: Text data "Senior, where is the format for this report?", Emotion data "Confused"

[0899] Output: Data sent to the server

[0900] As a specific operation, data is transmitted from the terminal to the server via the network.

[0901] Step 5:

[0902] The server passes the received text data to a natural language processing engine (e.g., spaCy) to analyze the question content and extract keywords.

[0903] Input: Text data "Senior, where can I find the format for this report?"

[0904] Output: Keywords "report", "format", "where"

[0905] Specifically, a natural language processing engine analyzes the text and identifies important words and phrases.

[0906] Step 6:

[0907] Based on the extracted keywords, the server searches internal databases such as MySQL databases to obtain relevant information.

[0908] Input: Keywords "report", "format", "where"

[0909] Output: Retrieved information: "The report format can be found in the document template section of the intranet."

[0910] Specifically, a database query is performed to retrieve relevant information.

[0911] Step 7:

[0912] The server takes into account the acquired information and emotional data and generates an appropriate answer using a generative AI model such as OpenAI's GPT-3.

[0913] Input: Information retrieved from the database: "The report format can be found in the document template section of the internal portal.", Emotion data: "Confused"

[0914] Output: Generated answer "The report format can be found in the document templates section of the internal portal. Please contact us if you require more information."

[0915] Specifically, the generative AI model receives a prompt and generates a natural-sounding response based on the context.

[0916] Step 8:

[0917] The generated text response is sent to the device, which converts it into voice data using a speech synthesis engine such as Amazon Polly.

[0918] Input: Text response "The report format can be found in the document templates section of the intranet. Please contact us if you require more information."

[0919] Output: Audio data

[0920] Specifically, the speech synthesis engine converts the text into speech and generates speech data.

[0921] Step 9:

[0922] The terminal plays back the generated voice data and provides the answer to the user.

[0923] Input: Audio data

[0924] Output: The user hears the audio response

[0925] Specifically, the audio data is played through the terminal's speaker or earphones.

[0926] (Application example 2)

[0927] 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."

[0928] It is difficult for workers on factory production lines to receive prompt and appropriate support when they have questions or problems related to their work. New or junior employees in particular need real-time assistance when they have questions or are confused about their work. Traditional methods require direct interaction with superiors or senior employees, which can lead to lower productivity. Furthermore, responses that do not take into account the worker's emotional state run the risk of increasing stress. Therefore, personalized support for workers is needed to improve work efficiency within factories.

[0929] 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.

[0930] In this invention, the server includes means for a user to input a question by voice, means for converting the voice question into text, means for transmitting the text data to the server, means for analyzing the text question and extracting keywords, means for retrieving information from a database based on the keywords, means for generating an answer using the retrieved information, means for transmitting the generated answer to a terminal, means for converting the text answer into voice, means for playing back the voice answer, and means for analyzing emotions. This allows factory workers to solve their questions and problems in real time and receive appropriate support that takes their emotional state into consideration.

[0931] "Users" refer to factory workers who use the system.

[0932] "Means for inputting questions by voice" refers to a device or method that allows a worker to input questions by voice using a microphone or the like.

[0933] "Means for converting voice queries into text" refers to a device or method that uses voice recognition technology to convert an input voice query into text data.

[0934] The "means for transmitting text data to a server" refers to a device or method for transmitting the converted text data to a server via a network.

[0935] "Means for analyzing text questions and extracting keywords" refers to a device or method that uses natural language processing techniques to extract important keywords from text questions.

[0936] The term "means for retrieving information from a database based on keywords" refers to a device or method that uses the extracted keywords to search for and retrieve related information from a database.

[0937] "Means for generating an answer using acquired information" refers to a device or method that generates an appropriate answer based on information acquired from a database.

[0938] The "means for transmitting the generated answer to the terminal" refers to a device or method for transmitting the generated answer to the user's terminal via a network.

[0939] "Means for converting text responses to speech" refers to a device or method that uses speech synthesis technology to convert generated text responses into speech data.

[0940] "Means for playing back audio responses" refers to a device or method for playing back converted audio data through a speaker or earphones.

[0941] "Emotion analysis means" refers to a device or method that analyzes the emotional state of a worker from voice or text and personalizes assistance based on that information.

[0942] The system of the present invention helps factory production line workers to quickly resolve work-related questions and problems remotely. The system's main components combine functions such as voice input, text conversion, emotion analysis, data transmission, information acquisition, answer generation, and voice playback.

[0943] In order to implement the present invention, the following hardware and software are used.

[0944] 1. Voice input

[0945] Hardware: Microphone

[0946] Software: Google Cloud Speech-to-Text API

[0947] Description: A user speaks into a microphone to ask a question about a task. For example, a question could be, "What is the maintenance procedure for this machine?"

[0948] 2. Text Conversion

[0949] Hardware: Devices (smartphones, tablets, PCs, etc.)

[0950] Software: Google Cloud Speech-to-Text API

[0951] Description: The input speech is converted to text data. The converted text is "What is the maintenance procedure for this machine?"

[0952] 3. Emotion analysis

[0953] Hardware: Built-in microphone

[0954] Software: Microsoft Azure Emotion API

[0955] Description: Analyzes a user's emotional state from speech and text. For example, recognizes the emotion "confused" from a user's voice.

[0956] 4. Data Transmission

[0957] Hardware: Terminal, Wi-Fi module

[0958] Software: HTTPS communication

[0959] Description: Sends the converted text data and analyzed emotion data to the server.

[0960] 5. Information acquisition

[0961] Hardware: Server

[0962] Software: Natural language processing engine, database search algorithm

[0963] Description: The server extracts keywords from the received text data and searches the database for related information based on these keywords. For example, it retrieves related procedure manuals based on the keyword "machine maintenance procedures."

[0964] 6. Answer generation

[0965] Hardware: Server

[0966] Software: Generative AI model (OpenAI GPT-4)

[0967] Description: Generate an appropriate answer based on the acquired information and emotion data. The generated answer will be in text format, such as "Details of the machine maintenance procedure can be found on page 5 of the procedure manual."

[0968] 7. Audio playback

[0969] Hardware: Device, speakers or earphones

[0970] Software: Google Cloud Text-to-Speech API

[0971] Description: The generated text response is converted to audio data and played back. The user hears, "The machine maintenance procedure is detailed on page 5 of the procedure manual."

[0972] As a result, workers feel as if they are being directly answered by a senior employee, allowing them to solve problems quickly and accurately.Furthermore, emotion analysis makes it possible to provide personalized responses based on the user's emotions, which is expected to improve work efficiency and reduce stress.

[0973] For example, if a user asks, "What is the maintenance procedure for this machine?", the system goes through the steps of voice conversion, emotion analysis, information acquisition, answer generation, and voice synthesis, and provides a spoken answer such as, "Details of the machine's maintenance procedure are on page 5 of the instruction manual."

[0974] An example of a prompt sentence is as follows:

[0975] Question: "What is the maintenance procedure for this machine?" Emotion: "I'm confused."

[0976] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0977] Step 1:

[0978] Users speak into a microphone to ask questions about their work. This voice becomes input to the system. For example, they might ask, "What is the maintenance procedure for this machine?"

[0979] Step 2:

[0980] The device receives voice input and converts it to text using the Google Cloud Speech-to-Text API. The converted text becomes the input for the next step in the system. The output is the text, "What are the maintenance procedures for this machine?"

[0981] Step 3:

[0982] The device then passes the converted text data to the Microsoft Azure Emotion API for emotion analysis. The analysis identifies the emotion the user expressed while asking the question. For example, emotional information such as "confused" is output. This emotional data is then used as input for the next step.

[0983] Step 4:

[0984] The device sends text data and emotion data to the server. The sent data is analyzed by the server's natural language processing engine, and important keywords are extracted. For example, from the question "What is the maintenance procedure for this machine?" keywords such as "machine," "maintenance," and "procedure" are extracted.

[0985] Step 5:

[0986] The server retrieves related information from the database based on the extracted keywords. For example, it searches using the keywords "machine," "maintenance," and "procedure" to retrieve procedure manuals and related technical documents. This retrieved information becomes the input for the next step.

[0987] Step 6:

[0988] The server generates an appropriate answer using a generative AI model (e.g., OpenAI GPT-4) based on the acquired information and emotion data. The generated answer is output as text data. For example, a response such as "Details of the machine maintenance procedure are on page 5 of the instruction manual" may be generated.

[0989] Step 7:

[0990] The server sends the generated text response to the device. The device passes it to the Google Cloud Text-to-Speech API and converts it into voice data. The converter outputs the voice data, "Details of the machine maintenance procedure can be found on page 5 of the procedure manual."

[0991] Step 8:

[0992] The terminal plays the generated voice data through a speaker or earphone, allowing the user to receive a voice response in real time. Specifically, the speaker will say, "Details of the machine maintenance procedure can be found on page 5 of the procedure manual."

[0993] 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.

[0994] 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.

[0995] 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.

[0996] [Fourth embodiment]

[0997] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0998] 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.

[0999] 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).

[1000] 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.

[1001] 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.

[1002] 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).

[1003] 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.

[1004] 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.

[1005] 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.

[1006] 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.

[1007] 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.

[1008] 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.

[1009] 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."

[1010] The system of the present invention enables new and junior employees to easily solve work-related questions and problems in a remote work environment. The system includes a means for a user to input a question by voice, a means for converting the voice question into text, a means for transmitting the text data to a server, a means for analyzing the text question and extracting keywords, a means for retrieving information from a database based on the keywords, a means for generating an answer using the retrieved information, a means for transmitting the generated answer to a terminal, a means for converting the text answer into voice, and a means for playing back the voice answer.

[1011] First, the user speaks a question about work into the device's microphone. This spoken question is converted into text data by the device's voice recognition engine. For example, if a user asks, "Senior, where can I find the format for this report?", the voice is converted into the text "Where can I find the format for this report?"

[1012] The device then sends the converted text data to the server as an HTTP request, which also includes user authentication information and is transmitted over a secure communication channel.

[1013] The server passes the received text data to a natural language processing engine, which analyzes the question. Specifically, it extracts important keywords and phrases from the question. In this example, keywords such as "report," "format," and "where" are extracted. The server uses these keywords to search the company database to obtain relevant information. For example, it obtains information such as "The report format can be found in the document template section of the company portal."

[1014] Based on the acquired information, the generative AI model on the server generates an appropriate answer. This answer is then formatted using natural language generation technology into a format that is easy for the user to understand. The generated answer is then sent to the device in text format.

[1015] The device passes the received text response to a speech synthesis engine, which converts it into voice data. Finally, the converted voice data is played back through the device's speaker or earphones. This allows the user to hear the answer to their question in voice, as if it were being answered directly by a senior employee.

[1016] For example, if a user asks a work-related question, "Senior, where can I find the format for this report?", the process is as follows: The device converts this speech into text and sends it to the server. The server searches the database based on the analysis and generates the answer, "The report format can be found in the document template section of the in-house portal." The generated answer is converted into speech and played back to the user via the device.

[1017] This system utilizes natural language processing and generation technologies to provide timely answers in a format that is easy for users to understand. It also allows new and junior employees to receive support as if they were being supported by a senior employee, even in a remote work environment, improving the work efficiency of new and junior employees and contributing to a lower turnover rate.

[1018] The processing flow will be explained below.

[1019] Step 1:

[1020] The user speaks into the device's microphone to ask a work-related question. For example, "Senior, where can I find the format for this report?"

[1021] Step 2:

[1022] The device uses a speech recognition engine to convert the user's spoken question into text. The speech data is analyzed and the corresponding text is generated. For example, the speech is converted to "Where can I find the format for this report?"

[1023] Step 3:

[1024] The device sends the converted text data to the server in the form of an HTTP request, which also includes the user's authentication token. For example, a text question and the authentication token are sent.

[1025] Step 4:

[1026] The server passes the received text data to a natural language processing engine, which analyzes the question. The engine extracts keywords and important phrases from the text. For example, it extracts keywords such as "report," "format," and "where."

[1027] Step 5:

[1028] The server searches the company database based on the analysis results to retrieve relevant information. It generates a search query and queries the database. For example, it executes the query "SELECT FROM documentation_templates WHERE category="report format"" and retrieves the results.

[1029] Step 6:

[1030] The generative AI model creates an answer based on the information obtained by the server. The answer is formatted using natural language generation technology in a way that is easy for the user to understand. For example, "The report format can be found in the document template section of the internal portal."

[1031] Step 7:

[1032] The server generates a text response and sends it to the terminal as an HTTP response. For example, the response text "The report format can be found in the document template section of the internal portal" is sent.

[1033] Step 8:

[1034] The text response received by the device is passed to the speech synthesis engine, which converts the text into speech data. The synthesized speech file is saved in temporary memory. Example: "The report format can be found in the document template section of the in-house portal" is converted into speech data.

[1035] Step 9:

[1036] The device outputs the audio data to a playback device (speaker or earphones). By listening to the answer, the user feels as if a senior employee had answered them directly. Example: Play the audio "The report format can be found in the document template section of the internal portal."

[1037] By following these steps, users can quickly get answers to their business-related questions even in a remote work environment.

[1038] Example 1

[1039] 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."

[1040] In a remote work environment, new and junior employees lack the means to easily resolve work-related questions and problems. This can lead to reduced work efficiency and delayed employee growth. There are also concerns that the difficulty of communication in a remote environment could lead to increased employee turnover.

[1041] 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.

[1042] In this invention, the server includes means for allowing a user to input a question by voice, means for converting the voice question into text, means for transmitting the text data to an information processing device, means for analyzing the text question and extracting keywords, means for acquiring information from an information storage device based on the keywords, means for generating an answer using the acquired information, means for transmitting the generated answer to a display device, means for converting the text answer into voice, means for playing back the voice answer, means for generating an answer in a format that is easy for the user to understand using natural language generation technology, and means for ensuring secure communication using HTTPS. This enables smooth communication even in a remote work environment, allowing new and junior employees to quickly solve problems, which is expected to improve work efficiency and reduce turnover.

[1043] A "user" is a person who uses the system to input questions by voice and receive answers.

[1044] A "means for inputting a question by voice" is a microphone or other voice input device that allows a user to input a voice question into the system.

[1045] A "means for converting voice queries into text" is a device or software that includes speech recognition technology or algorithms that analyzes input voice data and converts it into text data.

[1046] The "means for transmitting text data to the information processing device" refers to a communication function or protocol for transferring text data to a server via a network.

[1047] An "information processing device" is a computer system or server that analyzes text data and generates answers.

[1048] "Means for analyzing text questions and extracting keywords" refers to natural language processing techniques and algorithms that identify and extract important keywords from received text data.

[1049] "Means for retrieving information from an information storage device based on keywords" refers to a technique for using extracted keywords to search for and retrieve related information from databases, knowledge bases, and other information storage devices.

[1050] An "information storage device" is a storage device for storing databases and other digital information.

[1051] The "means for generating a response using acquired information" refers to a device or software that generates a response sentence using natural language generation technology based on the acquired information.

[1052] The "means for transmitting the generated answer to the display device" refers to a communication function or protocol for transferring the generated answer to the user's terminal via a network.

[1053] The "display device" refers to a display, speaker, or earphone for displaying or playing back the answer to the user.

[1054] "Means for converting text responses to speech" refers to speech synthesis techniques and algorithms that convert text responses into speech data.

[1055] The "means for playing back audio responses" refers to a device or system for playing back audio data through a speaker or earphones.

[1056] "Natural language generation technology" refers to the technology and algorithms used to analyze text data and generate natural-sounding sentences that are easy for humans to understand.

[1057] "Means of ensuring secure communications through HTTPS" refers to protocols and technologies that ensure the confidentiality and integrity of data by encrypting and sending and receiving communication data.

[1058] A "prompt" is text containing specific instructions or questions that a generative AI model uses to generate appropriate answers.

[1059] The system of the present invention allows new and junior employees to easily resolve work-related questions and problems in a remote work environment. This system is realized using the following hardware and software.

[1060] The user speaks a question about work into the device's microphone. For example, they might ask, "Senior, where can I find the format for this report?" This spoken question is converted into text data using the device's built-in speech recognition engine (e.g., Google Speech-to-Text API). The converted text data is then sent from the device to the server via secure communication using HTTPS.

[1061] The server passes the received text data to a natural language processing engine (e.g., SpaCy or BERT) and analyzes the question. Specifically, it tokenizes the text and extracts keywords such as "report," "format," and "where." Next, the server searches an information storage device (database or knowledge base) based on the extracted keywords to obtain related information. For example, it can obtain information such as "The report format can be found in the document template section of the internal portal."

[1062] Based on the acquired information, an appropriate answer is generated using a generative AI model on the server (e.g., OpenAI's GPT-3). At this time, the prompt text is "Please generate an answer that is easy for the user to understand based on the information, 'The report format can be found in the document template section of the internal portal.'" The generative AI model generates a natural and easy-to-understand answer based on this prompt text.

[1063] The generated answer is sent to the device in text format. The device then passes the received text answer to a speech synthesis engine (e.g., Amazon Polly or Google Text-to-Speech) and converts it into audio data. Finally, the device plays the converted audio data through a speaker or earphones. This process allows the user to hear the answer to their question in audio, as if it were being answered directly by a senior employee.

[1064] This system's unique feature is its use of natural language processing and natural language generation technologies, allowing it to quickly provide answers in a format that is easy for users to understand. It also uses the HTTPS protocol, which ensures secure communications, ensuring high communication security. By using this system, new and junior employees can communicate smoothly even in a remote work environment, which is expected to improve work efficiency and reduce turnover.

[1065] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1066] Step 1:

[1067] The user speaks a question into the microphone of the device. Specifically, the user speaks, "Senior, where is the format for this report?" The input of this step is the user's voice data, and the output is also voice data.

[1068] Step 2:

[1069] The device passes the voice data acquired from the microphone to a voice recognition engine. Specifically, a voice recognition engine (e.g., Google Speech-to-Text API) is used to convert the voice data into text data. The input of this step is the voice data, and the output is the text data, "Where can I find the format for this report?"

[1070] Step 3:

[1071] The terminal sends the converted text data to the server as an HTTPS request. Specifically, it generates a request including the text data and user authentication information and sends it to the server via a communication channel encrypted by SSL / TLS. The input of this step is the text data and authentication information, and the output is an HTTPS request to the server.

[1072] Step 4:

[1073] The server passes the received text data to a natural language processing engine. Specifically, it uses a natural language processing engine (e.g., SpaCy or BERT) to analyze the text and extract important keywords. For example, it extracts keywords such as "report," "format," and "where." The input of this step is the received text data, and the output is the extracted keywords.

[1074] Step 5:

[1075] The server searches the information storage device based on the extracted keywords. Specifically, it generates an SQL query to search a database (e.g., MySQL or PostgreSQL) and retrieves related information. For example, it obtains information such as "The report format is in the document template section of the in-house portal." The input of this step is the extracted keywords, and the output is the retrieved information.

[1076] Step 6:

[1077] The server uses a generative AI model to generate an answer based on the acquired information. Specifically, the prompt "Generate an answer that is easy for the user to understand based on the information 'The report format can be found in the document template section of the internal portal'" is input into a generative AI model (e.g., OpenAI's GPT-3) to generate a natural and easy-to-understand answer. The input for this step is the acquired information and the prompt, and the output is the generated answer in text format.

[1078] Step 7:

[1079] The server sends the generated text answer to the device using HTTPS. Specifically, the text answer is sent over a communication channel that is again encrypted with SSL / TLS. The input to this step is the generated text answer, and the output is an HTTPS response to the device.

[1080] Step 8:

[1081] The device passes the received text response to a speech synthesis engine (e.g., Amazon Polly or Google Text-to-Speech) to convert the text to speech data. The input of this step is the received text response, and the output is speech data.

[1082] Step 9:

[1083] The terminal plays the audio data through a speaker or earphone. Specifically, using the terminal's audio output device, the user hears the audio response as if the senior employee were speaking directly to them. The input for this step is the audio data, and the output is the audio response provided to the user.

[1084] By following these steps, users can get quick and accurate answers even in a remote work environment.

[1085] (Application example 1)

[1086] 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."

[1087] Factory floors require a means for operators to quickly and accurately give instructions and ask questions to robots. However, existing systems mainly use text input, and few support voice input. This prevents operators from responding quickly, resulting in reduced production efficiency. Furthermore, even if systems exist that support voice input, they lack natural language processing and generation technologies, making it difficult to provide accurate answers and instructions. There is a need to provide a system that can solve these issues and enable operators to communicate smoothly with robots.

[1088] 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.

[1089] In this invention, the server includes means for a user to input a question by voice, means for converting the voice question into text, means for transmitting the text data to the server, means for analyzing the text question and extracting keywords, means for retrieving information from a database based on the keywords, means for generating an answer using the retrieved information, means for transmitting the generated answer to a terminal, means for converting the text answer into voice, means for playing back the voice answer, and means for a user to give instructions or questions to a robot in a factory by voice, and for the robot to analyze the voice and return an appropriate answer or instruction by voice. This makes it possible for an operator to give instructions or questions to the robot by voice, thereby quickly and accurately acquiring information or issuing operating instructions.

[1090] "Voice input" is a means of capturing and processing the user's voice into an electronic device.

[1091] "Text conversion" is the process of converting information captured through speech into written data.

[1092] "Server transmission" refers to the act of transmitting the converted character data to the central processing unit.

[1093] "Text analysis" is a technology that extracts important keywords from transmitted text data and analyzes its meaning.

[1094] "Database acquisition" refers to searching and acquiring related information from a database based on the analysis results.

[1095] "Answer generation" is the process of creating an appropriate answer to a user's question based on the acquired information.

[1096] "Terminal transmission" is the act of transmitting the generated answer to a terminal accessible by the user.

[1097] "Speech conversion" is a technology that converts answers from text data into voice data.

[1098] "Audio playback" refers to the act of playing audio data through a speaker or the like and providing it to the user.

[1099] A "factory floor" is a facility where various production activities take place and where employees and robots work.

[1100] A "robot" is a machine that performs programmed actions and automates tasks.

[1101] This invention is a system that supports effective communication between operators and robots in factories. This system is composed of means for speech recognition, text conversion, text analysis, database search, answer generation, speech synthesis, and speech playback.

[1102] First, the user, an operator, gives instructions or asks questions to the robot by voice. To achieve this, the robot or its connected device is equipped with a microphone. For example, when the operator says, "Robot, how do I maintain this machine?", the voice data is captured.

[1103] Next, the terminal converts the acquired voice data into text data using a voice recognition engine (for example, a SpeechRecognition library), and sends the converted text data to a central processing unit (server) as an HTTP request.

[1104] Once the text data arrives at the server, it is analyzed by a natural language processing engine (for example, Hugging Face's Transformers library) to extract important keywords. In this example, keywords such as "machine," "maintenance," and "method" are extracted.

[1105] Next, the server uses these keywords to search for and retrieve relevant information from a database. The database stores various manuals and work instructions for factory machines. For example, the server might retrieve information such as, "To maintain a machine, first turn off the power and remove the safety devices."

[1106] Based on the acquired information, the server uses a generative AI model to generate an appropriate answer. This generative AI model incorporates technology to generate answers in a format that is easy for the user to understand. The generated answer is sent to the device in text format.

[1107] Finally, the terminal passes the received text response to a speech synthesis engine (for example, the Pyttsx3 library) and converts it into voice data. This voice data is then played back to the operator through a speaker, allowing the user to hear the answer to their question aloud, as if the robot were answering them directly.

[1108] For example, if an operator asks, "Robot, what's the next step?", the system will respond, "The next step is to install part A." Or, if the operator asks, "Robot, how do you maintain this machine?", the system will respond, "The way to maintain this machine is to first turn off the power and remove the safety devices."

[1109] An example of a prompt sentence to input to the generative AI model is:

[1110] "Robot, how do you maintain this machine?"

[1111] "Robot, what's the next step?"

[1112] Let's say.

[1113] In this way, a system can be realized that improves communication between operators and robots in factories and increases work efficiency.

[1114] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1115] Step 1:

[1116] The user enters a question by voice.

[1117] Specifically, the operator speaks into the microphone and asks a question, for example, "Robot, how do you maintain this machine?" The input here is voice data.

[1118] Step 2:

[1119] The terminal sends the voice data to a voice recognition engine and converts it into text data.

[1120] Specifically, the voice data acquired by the device's microphone is converted into text data using the SpeechRecognition library, resulting in the text data "Robot, how do you maintain this machine?"

[1121] Step 3:

[1122] The terminal transmits the text data to the server.

[1123] Specifically, the converted text data is sent to the server as an HTTP request. This request also includes user authentication information. The input is text data, and the output is a request to the server.

[1124] Step 4:

[1125] The server passes the text data to a natural language processing engine to extract keywords.

[1126] Specifically, the server analyzes the received text data using Hugging Face's Transformers library and extracts the keywords "machine," "maintenance," and "method." The input is text data, and the output is keyword data.

[1127] Step 5:

[1128] The server searches the database based on the keywords to retrieve relevant information.

[1129] Specifically, the server uses the extracted keywords to search a database containing manuals and work instructions for factory machines. The information obtained is, "To maintain the machine, first turn off the power and remove the safety devices." The input is keyword data, and the output is the acquired information.

[1130] Step 6:

[1131] Based on the information acquired by the server, an appropriate answer is generated using a generative AI model.

[1132] Specifically, the server uses the acquired information to send a prompt to a generative AI model (e.g., GPT-3) to generate an answer such as "First, turn off the power and remove the safety device." The input is the acquired information, and the output is the generated answer text.

[1133] Step 7:

[1134] The server sends the generated response to the terminal.

[1135] Specifically, the generated answer text is sent from the server to the terminal as an HTTP response. The input is the generated answer text, and the output is the response to the terminal.

[1136] Step 8:

[1137] The answer received by the terminal is passed to a speech synthesis engine and converted into voice data.

[1138] Specifically, the device passes the received text response to the Pyttsx3 library and converts it into audio data. The input is the text response and the output is audio data.

[1139] Step 9:

[1140] The terminal plays the generated audio data to the user through a speaker or earphone.

[1141] Specifically, a voice message saying, "First, turn off the power and remove the safety device" is played back. The input is voice data, and the output is voice playback.

[1142] This series of steps creates a system in which the user can ask questions by voice and receive answers by voice.

[1143] 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.

[1144] The system of the present invention enables new and junior employees to easily solve work-related questions and problems in a remote work environment. The system includes a means for a user to input a question by voice, a means for converting the voice question into text, a means for transmitting the text data to a server, a means for analyzing the text question and extracting keywords, a means for retrieving information from a database based on the keywords, a means for generating an answer using the retrieved information, a means for transmitting the generated answer to a terminal, a means for converting the text answer into voice, and a means for playing back the voice answer.

[1145] Furthermore, the present invention provides a more personalized response by combining an emotion engine that recognizes emotions from the user's voice. This emotion engine can recognize the user's emotional state and incorporate that information into the answer generation process.

[1146] First, the user speaks a work-related question into the device's microphone. For example, "Senior, where is the format for this report?" This spoken question is converted into text data by the device, and the emotion engine then analyzes the user's emotional state. The emotion engine analyzes the tone, pace, emphasis, etc. of the voice, and recognizes the user's emotional state, such as "I'm wondering" or "I'm confused."

[1147] The device then sends the converted text data and the recognized emotion data to the server. The server then passes the received text data to a natural language processing engine, which analyzes the question and extracts important keywords and phrases. For example, from the question "Where can I find the format for this report?", keywords such as "report," "format," and "where" are extracted.

[1148] The server searches the company database based on these keywords to retrieve relevant information, for example, "The report format can be found in the document template section of the company portal."

[1149] The server then uses the generative AI model to generate an appropriate response, taking into account the emotional data from the emotion engine. For example, if the server detects that the user is confused, the response will be more polite and detailed. The generated response is sent to the device in text format.

[1150] The device passes the received text response to a speech synthesis engine, which converts it into voice data. This voice data is then played through the device's speaker or earphones. For example, a polite voice might say, "The report format can be found in the document template section of the internal portal."

[1151] In this way, users can receive answers to their questions via voice, as if they were being answered directly by a senior employee. Furthermore, the emotion engine recognizes the user's emotional state and responds accordingly, providing more appropriate and personalized support. For example, if a user asks a work-related question, "Senior, where can I find the format for this report?", the system will go through the steps of voice conversion, emotion recognition, information acquisition, answer generation, and voice synthesis to provide the user with an appropriate answer via voice. This is particularly beneficial for new and junior employees in remote work environments, improving work efficiency and contributing to lower employee turnover.

[1152] The processing flow will be explained below.

[1153] Step 1:

[1154] The user speaks into the device's microphone to ask a work-related question. For example, "Senior, where can I find the format for this report?"

[1155] Step 2:

[1156] The device uses a speech recognition engine to convert the user's voice question into text. The voice data is analyzed and the corresponding text is generated. For example, "Where can I find the format for this report?"

[1157] Step 3:

[1158] While the device converts the voice question into text, it also uses an emotion engine to analyze the voice data and recognize the user's emotional state, for example by extracting emotions such as "doubtful" or "confused" from the tone and pace of the voice.

[1159] Step 4:

[1160] The device sends the converted text data and the recognized emotion data to the server in the form of an HTTP request. This request also includes the user's authentication token. Example: Send a text question, emotion data, and authentication token to the server.

[1161] Step 5:

[1162] The server passes the received text data to a natural language processing engine, which analyzes the question. The engine extracts keywords and important phrases from the text. For example, it extracts keywords such as "report," "format," and "where."

[1163] Step 6:

[1164] The server searches the company database based on the analysis results to retrieve relevant information. It generates a search query and queries the database. For example, it executes the query "SELECT FROM documentation_templates WHERE category="report format"" to retrieve relevant information.

[1165] Step 7:

[1166] The generative AI model creates a response based on the information obtained by the server. It adjusts the tone and level of detail of the response taking into account emotional data from the emotion engine. For example, if the user is perceived as "confused," the response will be more detailed and polite.

[1167] Step 8:

[1168] The server generates a text response and sends it to the device as an HTTP response. The server returns the text response to the device. Example: "The report format can be found in the document template section of the internal portal."

[1169] Step 9:

[1170] The text response received by the device is passed to the speech synthesis engine, which converts the text into speech data. The synthesized speech file is saved in temporary memory. For example, it is converted to "The report format can be found in the document template section of the internal portal."

[1171] Step 10:

[1172] The device outputs the audio data to a playback device (speaker or earphones). By listening to the answer, the user feels as if a senior employee had answered them directly. Example: "The report format can be found in the document template section of the internal portal."

[1173] Through this series of steps, users can receive prompt and personalized responses even in a remote work environment. In particular, the introduction of an emotion engine allows for appropriate responses based on the user's emotional state, which is expected to improve the work efficiency of new and junior employees and reduce their psychological burden.

[1174] Example 2

[1175] 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."

[1176] In recent years, with the rapid spread of remote work environments, new and junior employees are seeking ways to quickly and efficiently resolve work-related questions and problems. Conventional systems lack support for resolving work-related questions, particularly the lack of a method that combines voice input and emotion analysis. Furthermore, there is a lack of systems that can provide personalized responses that take into account the user's emotional state. This reduces the work efficiency of new and junior employees and leads to increased turnover, so a new system is needed to solve this issue.

[1177] 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.

[1178] In this invention, the server includes a means for analyzing a user's emotions, a means for reflecting the analyzed emotion data in the answer generation process, and a means for generating an appropriate answer using a generative AI model, thereby enabling the server to quickly provide appropriate and personalized answers to questions entered by the user via voice.

[1179] "User" refers to a person who uses the system to solve a business question or problem.

[1180] "Means for inputting questions by voice" refers to a device including a microphone and an interface that a user uses to input voice.

[1181] "Means for converting voice queries into text" refers to speech recognition software or algorithms used to convert user-entered voice data into text data.

[1182] "Means for transmitting text data to a server" refers to a communication means for sending text data to a server via a network.

[1183] "Means for analyzing text questions and extracting keywords" refers to natural language processing techniques used to analyze received text data and identify important words and phrases.

[1184] "Means for retrieving information from a database based on keywords" refers to means for searching a database using extracted keywords and retrieving related information.

[1185] "Means for generating an answer using acquired information" refers to the technology or algorithm used to create an appropriate answer based on the acquired information.

[1186] The "means for transmitting the generated answer to the terminal" refers to a communication means for sending the generated answer to the terminal via a network.

[1187] "Means for converting text responses to speech" refers to a speech synthesis engine or algorithm for converting the generated text responses into speech data.

[1188] The "means for playing back an audio response" refers to a device such as a speaker or earphone for playing back audio data and providing a response to the user.

[1189] "Means for analyzing user emotions" refers to an emotion engine or algorithm for analyzing the user's emotional state from their voice or text input.

[1190] "Means for incorporating analyzed emotional data into the answer generation process" refers to technologies and algorithms for adjusting answers to take into account the user's emotional state.

[1191] "Generative AI model" refers to an artificial intelligence model for generating text using natural language generation technology.

[1192] The system of the present invention allows new and junior employees to easily solve work-related questions and problems in a remote work environment. The system includes a means for users to input questions by voice, a means for converting the voice question into text, a means for transmitting the text data to a server, a means for analyzing the text question and extracting keywords, a means for retrieving information from a database based on the keywords, a means for generating an answer using the retrieved information, a means for transmitting the generated answer to a terminal, a means for converting the text answer into voice, and a means for playing back the voice answer. Furthermore, the present invention provides more personalized responses by combining a means for analyzing the user's emotions with a means for incorporating the analyzed emotional data into the answer generation process.

[1193] The user speaks a question about work into the device's microphone. The device used for this is a computer device such as a regular PC, smartphone, or tablet. For example, a user might ask, "Senior, where can I find the format for this report?" This spoken question is converted into text data using speech recognition software on the device (for example, Google Cloud Speech-to-Text).

[1194] The text data is then analyzed by an emotion engine (e.g., IBM Watson Tone Analyzer) to determine the user's emotional state. The emotion engine analyzes the tone, pace, and emphasis of the voice to recognize the user's emotional state (e.g., "I'm wondering," "I'm confused," etc.). The device then sends the converted text data and the recognized emotion data to the server.

[1195] The server passes the received text data to a natural language processing engine (e.g., spaCy) and analyzes the question. Important keywords and phrases are extracted, such as "report," "format," and "where." The server then searches a database (e.g., a MySQL database) based on these keywords to retrieve relevant information. For example, it retrieves the information, "The report format can be found in the document templates section of the internal portal."

[1196] The server then generates an appropriate response using a generative AI model (e.g., OpenAI's GPT-3) while taking into account the emotional data from the emotion engine. For example, if the server recognizes that the user is "confused," the response will be more polite and detailed. The generated response is sent to the device in text format.

[1197] The device passes the received text response to a speech synthesis engine (e.g., Amazon Polly) and converts it into voice data. This voice data is played back through the device's speaker or earphones. For example, it may say in a polite tone, "The report format can be found in the document template section of the internal portal."

[1198] A unique feature of this system is that it analyzes the user's emotions and uses that information to adjust the answer generation process, allowing users to receive personalized support as if they were being answered directly by a senior employee.

[1199] As a concrete example, if a user asks a work-related question such as "Senior, where can I find the format for this report?", the system will go through the steps of voice conversion, emotion recognition, information acquisition, answer generation, and voice synthesis to provide the user with an appropriate answer via voice.An example of a prompt sentence for the generative AI model is shown below.

[1200] Example prompt sentence:

[1201] User question: "Senior, where can I find the format for this report?"

[1202] Extracted keywords: "report", "format", "where"

[1203] Emotion recognized: "I'm confused"

[1204] Database lookup result: "The report format can be found in the Document Templates section of the Intranet Portal."

[1205] Generate an appropriate response considering the recognized emotion.

[1206] These prompts can be fed into a generative AI model to generate appropriate answers.

[1207] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1208] Step 1:

[1209] The user speaks into the microphone of the terminal to input a question about the business.

[1210] As a specific operation, when a user says, "Senior, where is the format for this report?", this speech is input.

[1211] Step 2:

[1212] The device uses voice recognition software such as Google Cloud Speech-to-Text to convert the user's voice data into text data.

[1213] Input: Voice data "Senior, where can I find the format for this report?"

[1214] Output: Text data "Senior, where can I find the format for this report?"

[1215] Specifically, speech recognition software analyzes the speech waveform and converts it into corresponding text.

[1216] Step 3:

[1217] The device then sends the converted text data to an emotion engine such as IBM Watson Tone Analyzer to analyze the user's emotional state.

[1218] Input: Text data "Senior, where can I find the format for this report?"

[1219] Output: Emotion data "Confused"

[1220] Specifically, the emotion engine analyzes the tone, pace, and emphasis of the voice to recognize the user's emotions.

[1221] Step 4:

[1222] The terminal transmits the converted text data and the recognized emotion data to the server.

[1223] Input: Text data "Senior, where is the format for this report?", Emotion data "Confused"

[1224] Output: Data sent to the server

[1225] As a specific operation, data is transmitted from the terminal to the server via the network.

[1226] Step 5:

[1227] The server passes the received text data to a natural language processing engine (e.g., spaCy) to analyze the question content and extract keywords.

[1228] Input: Text data "Senior, where can I find the format for this report?"

[1229] Output: Keywords "report", "format", "where"

[1230] Specifically, a natural language processing engine analyzes the text and identifies important words and phrases.

[1231] Step 6:

[1232] Based on the extracted keywords, the server searches internal databases such as MySQL databases to obtain relevant information.

[1233] Input: Keywords "report", "format", "where"

[1234] Output: Retrieved information: "The report format can be found in the document template section of the intranet."

[1235] Specifically, a database query is performed to retrieve relevant information.

[1236] Step 7:

[1237] The server takes into account the acquired information and emotional data and generates an appropriate answer using a generative AI model such as OpenAI's GPT-3.

[1238] Input: Information retrieved from the database: "The report format can be found in the document template section of the internal portal.", Emotion data: "Confused"

[1239] Output: Generated answer "The report format can be found in the document templates section of the internal portal. Please contact us if you require more information."

[1240] Specifically, the generative AI model receives a prompt and generates a natural-sounding response based on the context.

[1241] Step 8:

[1242] The generated text response is sent to the device, which converts it into voice data using a speech synthesis engine such as Amazon Polly.

[1243] Input: Text response "The report format can be found in the document templates section of the intranet. Please contact us if you require more information."

[1244] Output: Audio data

[1245] Specifically, the speech synthesis engine converts the text into speech and generates speech data.

[1246] Step 9:

[1247] The terminal plays back the generated voice data and provides the answer to the user.

[1248] Input: Audio data

[1249] Output: The user hears the audio response

[1250] Specifically, the audio data is played through the terminal's speaker or earphones.

[1251] (Application example 2)

[1252] 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."

[1253] It is difficult for workers on factory production lines to receive prompt and appropriate support when they have questions or problems related to their work. New or junior employees in particular need real-time assistance when they have questions or are confused about their work. Traditional methods require direct interaction with superiors or senior employees, which can lead to lower productivity. Furthermore, responses that do not take into account the worker's emotional state run the risk of increasing stress. Therefore, personalized support for workers is needed to improve work efficiency within factories.

[1254] 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.

[1255] In this invention, the server includes means for a user to input a question by voice, means for converting the voice question into text, means for transmitting the text data to the server, means for analyzing the text question and extracting keywords, means for retrieving information from a database based on the keywords, means for generating an answer using the retrieved information, means for transmitting the generated answer to a terminal, means for converting the text answer into voice, means for playing back the voice answer, and means for analyzing emotions. This allows factory workers to solve their questions and problems in real time and receive appropriate support that takes their emotional state into consideration.

[1256] "Users" refer to factory workers who use the system.

[1257] "Means for inputting questions by voice" refers to a device or method that allows a worker to input questions by voice using a microphone or the like.

[1258] "Means for converting voice queries into text" refers to a device or method that uses voice recognition technology to convert an input voice query into text data.

[1259] The "means for transmitting text data to a server" refers to a device or method for transmitting the converted text data to a server via a network.

[1260] "Means for analyzing text questions and extracting keywords" refers to a device or method that uses natural language processing techniques to extract important keywords from text questions.

[1261] The term "means for retrieving information from a database based on keywords" refers to a device or method that uses the extracted keywords to search for and retrieve related information from a database.

[1262] "Means for generating an answer using acquired information" refers to a device or method that generates an appropriate answer based on information acquired from a database.

[1263] The "means for transmitting the generated answer to the terminal" refers to a device or method for transmitting the generated answer to the user's terminal via a network.

[1264] "Means for converting text responses to speech" refers to a device or method that uses speech synthesis technology to convert generated text responses into speech data.

[1265] "Means for playing back audio responses" refers to a device or method for playing back converted audio data through a speaker or earphones.

[1266] "Emotion analysis means" refers to a device or method that analyzes the emotional state of a worker from voice or text and personalizes assistance based on that information.

[1267] The system of the present invention helps factory production line workers to quickly resolve work-related questions and problems remotely. The system's main components combine functions such as voice input, text conversion, emotion analysis, data transmission, information acquisition, answer generation, and voice playback.

[1268] In order to implement the present invention, the following hardware and software are used.

[1269] 1. Voice input

[1270] Hardware: Microphone

[1271] Software: Google Cloud Speech-to-Text API

[1272] Description: A user speaks into a microphone to ask a question about a task. For example, a question could be, "What is the maintenance procedure for this machine?"

[1273] 2. Text Conversion

[1274] Hardware: Devices (smartphones, tablets, PCs, etc.)

[1275] Software: Google Cloud Speech-to-Text API

[1276] Description: The input speech is converted to text data. The converted text is "What is the maintenance procedure for this machine?"

[1277] 3. Emotion analysis

[1278] Hardware: Built-in microphone

[1279] Software: Microsoft Azure Emotion API

[1280] Description: Analyzes a user's emotional state from speech and text. For example, recognizes the emotion "confused" from a user's voice.

[1281] 4. Data Transmission

[1282] Hardware: Terminal, Wi-Fi module

[1283] Software: HTTPS communication

[1284] Description: Sends the converted text data and analyzed emotion data to the server.

[1285] 5. Information acquisition

[1286] Hardware: Server

[1287] Software: Natural language processing engine, database search algorithm

[1288] Description: The server extracts keywords from the received text data and searches the database for related information based on these keywords. For example, it retrieves related procedure manuals based on the keyword "machine maintenance procedures."

[1289] 6. Answer generation

[1290] Hardware: Server

[1291] Software: Generative AI model (OpenAI GPT-4)

[1292] Description: Generate an appropriate answer based on the acquired information and emotion data. The generated answer will be in text format, such as "Details of the machine maintenance procedure can be found on page 5 of the procedure manual."

[1293] 7. Audio playback

[1294] Hardware: Device, speakers or earphones

[1295] Software: Google Cloud Text-to-Speech API

[1296] Description: The generated text response is converted to audio data and played back. The user hears, "The machine maintenance procedure is detailed on page 5 of the procedure manual."

[1297] As a result, workers feel as if they are being directly answered by a senior employee, allowing them to solve problems quickly and accurately.Furthermore, emotion analysis makes it possible to provide personalized responses based on the user's emotions, which is expected to improve work efficiency and reduce stress.

[1298] For example, if a user asks, "What is the maintenance procedure for this machine?", the system goes through the steps of voice conversion, emotion analysis, information acquisition, answer generation, and voice synthesis, and provides a spoken answer such as, "Details of the machine's maintenance procedure are on page 5 of the instruction manual."

[1299] An example of a prompt sentence is as follows:

[1300] Question: "What is the maintenance procedure for this machine?" Emotion: "I'm confused."

[1301] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1302] Step 1:

[1303] Users speak into a microphone to ask questions about their work. This voice becomes input to the system. For example, they might ask, "What is the maintenance procedure for this machine?"

[1304] Step 2:

[1305] The device receives voice input and converts it to text using the Google Cloud Speech-to-Text API. The converted text becomes the input for the next step in the system. The output is the text, "What are the maintenance procedures for this machine?"

[1306] Step 3:

[1307] The device then passes the converted text data to the Microsoft Azure Emotion API for emotion analysis. The analysis identifies the emotion the user expressed while asking the question. For example, emotional information such as "confused" is output. This emotional data is then used as input for the next step.

[1308] Step 4:

[1309] The device sends text data and emotion data to the server. The sent data is analyzed by the server's natural language processing engine, and important keywords are extracted. For example, from the question "What is the maintenance procedure for this machine?" keywords such as "machine," "maintenance," and "procedure" are extracted.

[1310] Step 5:

[1311] The server retrieves related information from the database based on the extracted keywords. For example, it searches using the keywords "machine," "maintenance," and "procedure" to retrieve procedure manuals and related technical documents. This retrieved information becomes the input for the next step.

[1312] Step 6:

[1313] The server generates an appropriate answer using a generative AI model (e.g., OpenAI GPT-4) based on the acquired information and emotion data. The generated answer is output as text data. For example, a response such as "Details of the machine maintenance procedure are on page 5 of the instruction manual" may be generated.

[1314] Step 7:

[1315] The server sends the generated text response to the device. The device passes it to the Google Cloud Text-to-Speech API and converts it into voice data. The converter outputs the voice data, "Details of the machine maintenance procedure can be found on page 5 of the procedure manual."

[1316] Step 8:

[1317] The terminal plays the generated voice data through a speaker or earphone, allowing the user to receive a voice response in real time. Specifically, the speaker will say, "Details of the machine maintenance procedure can be found on page 5 of the procedure manual."

[1318] 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.

[1319] 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.

[1320] 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.

[1321] 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.

[1322] 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.

[1323] 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.

[1324] 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).

[1325] 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.

[1326] 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."

[1327] 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.

[1328] 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).

[1329] 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.

[1330] 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.

[1331] 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.

[1332] 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.

[1333] 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.

[1334] 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.

[1335] 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.

[1336] 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.

[1337] 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.

[1338] 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.

[1339] The following is further disclosed regarding the above embodiment.

[1340] (Claim 1)

[1341] means for a user to input a question by voice;

[1342] a means for converting the voice query into text;

[1343] means for transmitting text data to a server;

[1344] a means for analyzing the text question to extract keywords;

[1345] means for retrieving information from a database based on keywords;

[1346] a means for generating a response using the acquired information;

[1347] means for transmitting the generated answer to the terminal;

[1348] a means for converting the text response into speech;

[1349] The system includes a means for playing back an audio response.

[1350] (Claim 2)

[1351] 2. The system according to claim 1, wherein the means for converting voice queries into text uses natural language processing technology.

[1352] (Claim 3)

[1353] 2. The system according to claim 1, wherein the answer generating means uses natural language generation technology.

[1354] "Example 1"

[1355] (Claim 1)

[1356] means for a user to input a question by voice;

[1357] a means for converting the voice query into text;

[1358] means for transmitting text data to an information processing device;

[1359] a means for analyzing the text question to extract keywords;

[1360] means for retrieving information from an information storage device based on a keyword;

[1361] a means for generating a response using the acquired information;

[1362] means for transmitting the generated answer to a display device;

[1363] a means for converting the text response into speech;

[1364] means for playing the audio response;

[1365] A means for generating answers in a user-friendly format using natural language generation technology;

[1366] A system that includes a means to ensure secure communications via HTTPS.

[1367] (Claim 2)

[1368] 10. The system of claim 1, further comprising means for analyzing the spoken query using natural language processing techniques.

[1369] (Claim 3)

[1370] 10. The system of claim 1, wherein the answer generated is based on the prompt sentence using natural language generation techniques.

[1371] "Application Example 1"

[1372] (Claim 1)

[1373] means for a user to input a question by voice;

[1374] a means for converting the voice query into text;

[1375] means for transmitting text data to a server;

[1376] a means for analyzing the text question to extract keywords;

[1377] means for retrieving information from a database based on keywords;

[1378] a means for generating a response using the acquired information;

[1379] means for transmitting the generated answer to the terminal;

[1380] a means for converting the text response into speech;

[1381] means for playing the audio response;

[1382] A system that includes a means for a user to give instructions or questions to a robot in a factory by voice, and for the robot to analyze the voice and return appropriate answers or instructions by voice.

[1383] (Claim 2)

[1384] 2. The system according to claim 1, wherein the means for converting voice queries into text uses natural language processing technology.

[1385] (Claim 3)

[1386] 2. The system according to claim 1, wherein the answer generating means uses natural language generation technology.

[1387] "Example 2: Combining Emotion Engines"

[1388] (Claim 1)

[1389] means for a user to input a question by voice;

[1390] a means for converting the voice query into text;

[1391] means for transmitting text data to a server;

[1392] a means for analyzing the text question to extract keywords;

[1393] means for retrieving information from a database based on keywords;

[1394] a means for generating a response using the acquired information;

[1395] means for transmitting the generated answer to the terminal;

[1396] a means for converting the text response into speech;

[1397] means for playing the audio response;

[1398] means for analyzing user emotions;

[1399] A means for reflecting the analyzed emotion data in the answer generation process;

[1400] A system including:

[1401] (Claim 2)

[1402] 2. The system according to claim 1, wherein the means for converting voice queries into text uses natural language processing technology.

[1403] (Claim 3)

[1404] The system according to claim 1, characterized in that the answer generation means uses natural language generation technology and a generative AI model.

[1405] "Application example 2 when combining emotion engines"

[1406] (Claim 1)

[1407] means for a user to input a question by voice;

[1408] a means for converting the voice query into text;

[1409] means for transmitting text data to a server;

[1410] a means for analyzing the text question to extract keywords;

[1411] means for retrieving information from a database based on keywords;

[1412] a means for generating a response using the acquired information;

[1413] means for transmitting the generated answer to the terminal;

[1414] a means for converting the text response into speech;

[1415] means for playing back an audio response; and

[1416] A system including a means for sentiment analysis.

[1417] (Claim 2)

[1418] 2. The system according to claim 1, wherein the means for converting voice queries into text uses natural language processing technology.

[1419] (Claim 3)

[1420] 2. The system according to claim 1, wherein the answer generating means uses natural language generation technology. [Explanation of symbols]

[1421] 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 a user to input a question by voice; a means for converting the voice query into text; means for transmitting text data to a server; a means for analyzing the text question to extract keywords; means for retrieving information from a database based on keywords; a means for generating a response using the acquired information; means for transmitting the generated answer to the terminal; a means for converting the text response into speech; The system includes a means for playing back an audio response.

2. 2. The system according to claim 1, wherein the means for converting voice queries into text uses natural language processing techniques.

3. 2. The system of claim 1, wherein the answer generating means uses natural language generation techniques.

Images