system

Abstract

We provide the system. [Solution] A means of acquiring voice input, A means of converting acquired audio into text data, A means of analyzing converted text data in real time and extracting multifaceted insights by considering cultural, professional, and generational backgrounds, A means of presenting extracted insights to the user, A method for automatically generating meeting minutes that summarize the key points of a meeting, A system that includes this.

Description

Technical Field

[0004]

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

Background Art

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

Prior Art Documents

Patent Documents

[0003] ​​​​​​​​​​​​​​​​​​​​This invention provides means for acquiring voice input and converting the acquired voice into text data, enabling real-time, multifaceted analysis. Furthermore, it includes means for analyzing the converted text data and presenting insights that take into account different backgrounds to the user in real time, supporting decision-making during meetings. By adding a means for automatically generating meeting minutes that summarize key points, this system can improve meeting efficiency and promote insights from diverse perspectives.

[0006] "Voice input" is the process of acquiring voice information as digital data.

[0007] "Text data" refers to a data format that converts voice input into written text.

[0008] "Real-time analysis" is a process that processes data instantaneously and makes the results immediately available for use.

[0009] "Cultural background" refers to the way of thinking and values ​​shaped by the culture to which the speaker belongs.

[0010] "Expert background" refers to the perspective based on the speaker's knowledge and experience related to a specific field of expertise.

[0011] "Generational background" refers to differences in values ​​and perspectives based on the generation to which the speaker belongs.

[0012] "Insight" refers to the insights and deep understanding gained from data and information.

[0013] Meeting minutes are documents that summarize the content of a meeting and record the main discussions and conclusions.

[0014] "Cloud storage" is a system that allows data to be managed and stored remotely using data storage services provided over the internet. [Brief explanation of the drawing]

[0015] [Figure 1] It is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

Embodiments for Carrying Out the Invention

[0016] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0017] First, the terms used in the following description will be explained.

[0018] In the following embodiments, a tagged processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0019] In the following embodiments, a tagged RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0020] In the following embodiments, a tagged storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0021] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0023] [First Embodiment]

[0024] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0025] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0026] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0027] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0028] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0031] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0033] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0034] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0035] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0036] This invention is an embodiment of a system that analyzes the content of meeting discussions in real time and provides multifaceted insights by considering cultural, professional, and generational backgrounds. To achieve this, the following program structure is employed.

[0037] 1. Acquisition of voice input

[0038] The device uses a high-quality microphone to capture participants' speech during the meeting as audio data. This allows for clear and low-noise audio.

[0039] 2. Speech-to-text conversion

[0040] Once audio is acquired, the server receives the audio data and uses a speech recognition engine to convert it into text data. The converted text is then used in the next parsing step.

[0041] 3. A multifaceted analysis of the statements made.

[0042] The server uses natural language processing techniques to analyze the background information contained in the text data. This includes participants' cultural backgrounds, expertise, and generational differences. This analysis generates multifaceted insights.

[0043] 4. Providing real-time feedback

[0044] Based on the analysis results, the server extracts different perspectives and key insights in real time during the meeting and presents them to the user via the terminal. This allows for faster and more accurate decision-making during the meeting.

[0045] 5. Automatic generation of meeting minutes

[0046] After the meeting ends, the server automatically generates meeting minutes summarizing all the points made and their analysis. These minutes can be referenced later, significantly improving meeting efficiency.

[0047] As a concrete example, consider a progress meeting for an international project. Each participant has a different cultural background and professional perspective. In this setting, the terminal captures the spoken audio, and the server analyzes the background information. This deepens the understanding of the participants' differing opinions from their respective fields of expertise, enabling appropriate project coordination. Finally, the server summarizes the key points of the meeting and generates concise and easy-to-understand minutes, providing convenience for later review.

[0048] The following describes the processing flow.

[0049] Step 1:

[0050] The device captures the speech of meeting participants in real time via the microphone and saves it as audio data.

[0051] Step 2:

[0052] The device compresses the acquired audio data and sends it to the server via the network.

[0053] Step 3:

[0054] The server processes the received audio data through a speech recognition engine to convert the audio into text data. Preprocessing, such as noise filtering, is also performed to improve the accuracy of the conversion results.

[0055] Step 4:

[0056] The server passes the converted text data to an analysis module, which uses natural language processing techniques to comprehensively analyze the content of the speech. This includes analysis of cultural, professional, and generational backgrounds.

[0057] Step 5:

[0058] The server organizes the analysis results, extracting key insights and different perspectives. This information is then compiled as real-time feedback.

[0059] Step 6:

[0060] The server extracts feedback information and sends it to the terminal, which then presents it to the user via screen display or audio.

[0061] Step 7:

[0062] After the meeting ends, the server automatically generates meeting minutes summarizing the key points based on all the analysis results. These minutes are saved to cloud storage, making them accessible to users later.

[0063] (Example 1)

[0064] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0065] In today's diverse meetings involving participants from various backgrounds, it is essential to immediately and accurately understand what is said and quickly gain multifaceted insights based on that understanding. However, when these processes are carried out manually, they are time-consuming and lead to delays in decision-making. Furthermore, insights that do not consider cultural, professional, and generational differences are insufficient. In addition, creating meeting minutes after the meeting is another time-consuming and laborious task.

[0066] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0067] In this invention, the server includes a device for acquiring voice input, a device for converting the acquired voice into text information, and a device for analyzing the converted text information in real time and extracting multifaceted insights considering cultural, professional, and generational backgrounds. This makes it possible to quickly and comprehensively analyze the content of speech during meetings and provide users with immediate insights. Furthermore, the automatic generation of meeting minutes summarizing the key points of the meeting allows for efficient review afterward.

[0068] "Voice input" is the process of acquiring audio signals generated during meetings or conversations.

[0069] A "device" is a piece of equipment or system designed to perform a specific function.

[0070] "Text information" refers to information obtained by converting audio data into written characters.

[0071] "Real-time analysis" refers to a process where data is processed and analyzed immediately as soon as it is generated.

[0072] "Extracting insights from multiple perspectives" is the process of considering different viewpoints and backgrounds to derive useful information and understanding.

[0073] "Cultural, professional, and generational background" refers to the set of characteristics and influences related to an individual's culture, professional knowledge, and age.

[0074] "Insight" refers to the insights and understanding gained from specific data or information.

[0075] A "user" is a person or group that uses a system or product.

[0076] "Key points" are elements or content within information or events that deserve particular attention or have high priority.

[0077] "Meeting minutes" are documents that record the content and decisions discussed and made at meetings or discussions.

[0078] A "communication protocol for rapid processing" refers to a set of rules and procedures for sending and receiving data smoothly and efficiently.

[0079] "Audio processing technology for noise reduction" refers to technology used to reduce undesirable noise and interference in audio signals.

[0080] This invention is a system that improves the effectiveness of meetings involving multiple participants. The system acquires audio input from meetings, converts it into text information, and then analyzes it in real time to provide users with multifaceted insights that take cultural, professional, and generational backgrounds into consideration. The following hardware and software are used to realize this configuration.

[0081] The terminal uses a high-quality microphone to capture speech during the meeting as audio. Noise cancellation technology minimizes background noise, sending clear audio data to the server. The server processes the received audio data through a speech recognition engine (e.g., a general speech recognition service) to convert the speech into text. Furthermore, the server uses natural language processing technology, specifically generative AI models such as BERT and GPT, to analyze this text information and generate multifaceted insights. This makes it possible to extract valuable insights in real time, taking into account the cultural and professional backgrounds of the participants.

[0082] As a concrete example, consider an international project meeting. This meeting includes multiple participants with different cultural backgrounds and professional perspectives. By having a terminal capture each participant's speech as audio and a server quickly process that audio, users can understand diverse opinions in real time and make appropriate decisions regarding the project's progress. This is expected to lead to smoother progress and improved results.

[0083] This system allows users to input prompts using a generative AI model to enable more accurate and effective meeting decision-making. For example, a prompt such as "Automatically generate minutes for this meeting, highlighting insights based on cultural context and expert perspectives" can be used. This ensures that meeting content is efficiently recorded and analyzed, and the information is provided in a format suitable for future reference.

[0084] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0085] Step 1:

[0086] The device uses a high-quality microphone to capture audio during the meeting. It utilizes noise cancellation to reduce background noise and collect clear, high-precision audio data. The input here is the participants' speech, and the output is digital audio data.

[0087] Step 2:

[0088] The terminal sends the collected audio data to the server. Due to the high-speed communication protocol used, the audio data reaches the server with virtually no delay. The input is the audio data from the terminal, and the output is the audio data transferred to the server.

[0089] Step 3:

[0090] The server processes the received audio data through a speech recognition engine and converts it into text data. Specifically, the speech recognition algorithm analyzes the sounds of words and generates the corresponding text. The input is audio data, and the output is text information.

[0091] Step 4:

[0092] The server analyzes the converted text data using a generation AI model. Specifically, it uses natural language processing techniques to extract the meaning of the text and generate insights that take into account cultural, professional, and generational contexts. The input is text data, and the output is analyzed insightful information.

[0093] Step 5:

[0094] Based on the analysis results, the server generates multifaceted perspectives and key insights for the user, which are then provided via the terminal. Users utilize this feedback to deepen discussions during meetings. The input is the analyzed insights, and the output is the feedback information presented to the user.

[0095] Step 6:

[0096] After the meeting ends, the server automatically generates meeting minutes summarizing the key points based on all the statements and analysis results. Specifically, it utilizes natural language generation technology to create a readable and well-organized document. The input is all the data generated during the meeting, and the output is the summarized meeting minutes.

[0097] (Application Example 1)

[0098] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0099] In autonomous vehicles, there is a need to facilitate communication among passengers, provide services based on their diverse backgrounds and interests, and improve the ride experience. However, currently, there is a lack of means to provide appropriate insights in real time based on what passengers say and to optimize the driving route and in-vehicle environment. A new system is needed to solve this problem.

[0100] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0101] In this invention, the server includes means for acquiring voice information, means for converting it into symbolic data, and means for extracting insights considering cultural background and expertise. This makes it possible to analyze passenger conversations in real time and optimize routes and services based on their interests.

[0102] "Voice information" refers to the recording of voices spoken by users, converted into data.

[0103] "Symbol sequence data" refers to text data obtained by converting audio information into characters and symbols.

[0104] "Insight" refers to a deep understanding and knowledge gained from analyzed data, and is information that is useful for decision-making.

[0105] A "user" is an individual or organization that uses the system to obtain information.

[0106] "Automatic generation" means that a system or program automatically creates data or documents without human intervention.

[0107] A "vehicle" is a vehicle used to transport people or goods on land.

[0108] A "proposal" is the act of presenting specific options or action plans.

[0109] This system aims to improve the ride experience by facilitating communication between passengers in autonomous vehicles. The server acquires voice information using a high-quality microphone. The acquired voice information is converted into symbolic data by the vehicle's computer. Google® Speech-to-Text API is used for this conversion. The server uses a natural language processing model running on Azure® or AWS® to analyze the converted symbolic data and extract insights that take into account the passengers' cultural backgrounds and expertise.

[0110] For example, if a passenger says they want to visit a nearby tourist attraction, the server uses the Google Maps API to search for the nearest attractions and suggests a suitable route in real time. This allows passengers to receive on-demand services tailored to their individual needs.

[0111] An example of a prompt message could be, "Analyze your conversation while driving and suggest recommendations based on places you want to visit and activities you're interested in." This system enables real-time feedback and highly adaptive service delivery, improving passenger satisfaction.

[0112] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0113] Step 1:

[0114] The terminal uses a high-quality microphone to acquire audio information from inside the vehicle. This results in clear audio data with minimal noise. This audio data is then sent to the server as input.

[0115] Step 2:

[0116] The server converts the acquired audio information into symbolic text data using the Google Speech-to-Text API. In this step, the audio data is automatically converted to text and used for the analysis process within the server. The converted symbolic text data is then output.

[0117] Step 3:

[0118] The server analyzes the symbol sequence data using a natural language processing model trained with TENSORFLOW®. During this analysis, a generative AI model is applied to identify passengers' intentions and interests, extracting multifaceted insights. The analysis results in the output of insight data.

[0119] Step 4:

[0120] The server generates appropriate suggestions for the user based on the extracted insight data. Specifically, it uses the Google Maps API to collect destination information and calculate the optimal route. This generates the suggestions that are presented to the user.

[0121] Step 5:

[0122] The user receives suggestions sent from the server via an in-vehicle display or voice prompt. At this stage, possible route information and activity suggestions are output and presented to the user as options.

[0123] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0124] This invention combines an emotion engine with a system that analyzes audio data acquired during meetings, thereby recognizing the user's emotions along with the content of their speech and providing real-time feedback. The following program structure is used to implement this invention.

[0125] 1. Voice input acquisition and emotion recognition

[0126] The device acquires participants' statements during the meeting and uses an emotion engine to recognize the user's emotional state through facial recognition cameras and voice intonation analysis. This emotion data is then associated with each statement.

[0127] 2. Transmission of voice and emotion data

[0128] The device transmits the acquired voice and emotion data to the server in real time. The server then prepares to process this data in an integrated manner.

[0129] 3. Analysis and conversion of audio data

[0130] The server converts the audio data into text data using a speech recognition engine. Furthermore, it utilizes natural language processing technology to analyze the content and background information of the speech, and also incorporates emotion recognition data into the analysis.

[0131] 4. Providing insights and emotion-oriented feedback

[0132] The server extracts insights from multiple perspectives based on analysis results and emotional data. The device then presents the user with real-time feedback, adjusted for user acceptance and empathy, taking into account the emotional recognition results.

[0133] 5. Automated generation of meeting minutes and integration of sentiment data

[0134] After the meeting ends, the server summarizes the key points based on the content of the speeches and sentiment data, and automatically generates meeting minutes. These minutes also incorporate sentiment data, allowing for an understanding of the emotional context of the speeches. The generated data is saved to cloud storage in a format that can be accessed later.

[0135] As a concrete example, considering a development meeting involving an international team, it becomes easier to understand the level of employee motivation and concerns regarding the project's progress during the meeting. This allows the project manager to take necessary measures during the meeting and effectively promote communication among stakeholders. Furthermore, the generation of meeting minutes, which include changes in emotions, can be used for future decision-making.

[0136] The following describes the processing flow.

[0137] Step 1:

[0138] The device uses microphones and cameras installed in the meeting room to capture participants' speech and facial expressions in real time. Both audio and video data are temporarily stored in local storage.

[0139] Step 2:

[0140] The device compresses the stored audio data, performs noise filtering, and then sends it to the server via the network. Simultaneously, it sends facial expression information obtained from video data to the server as a feature vector.

[0141] Step 3:

[0142] The server analyzes the received audio data using a speech recognition engine and converts it into text data. During this process, an acoustic model is applied to improve the accuracy of speech recognition.

[0143] Step 4:

[0144] The server analyzes text data using natural language processing technology to comprehensively analyze the meaning of the utterances. This analysis includes extracting field-specific keywords and understanding the context.

[0145] Step 5:

[0146] The server simultaneously uses an emotion engine based on the received facial expression data to recognize the emotional state of the participants. Emotional states are classified into categories such as "joy," "surprise," "sadness," and "anger."

[0147] Step 6:

[0148] The server integrates emotion recognition data with the spoken content and generates context-appropriate, emotion-oriented feedback. The generated feedback is adjusted to a tone that is considerate of the recipient of the statement.

[0149] Step 7:

[0150] The device displays this feedback to the user in real time. This clarifies the intent and nuances during the meeting and improves communication among participants.

[0151] Step 8:

[0152] After the meeting ends, the server automatically generates meeting minutes, summarizing all spoken and emotional data to highlight key points. These minutes also record emotional changes during speaking, which can be used for later analysis and decision-making.

[0153] Step 9:

[0154] The generated meeting minutes and related data are saved to cloud storage by the server, allowing users to access them securely later.

[0155] (Example 2)

[0156] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0157] In meetings, it is essential to understand not only the content of participants' statements but also their emotional state at the time of speaking, and to provide appropriate feedback immediately. However, existing systems are limited to speech-to-text conversion and simple content analysis, and do not adequately grasp emotional states or provide insights based on them. As a result, the quality of meetings does not improve, and it is a factor that diminishes participants' motivation.

[0158] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0159] In this invention, the server includes means for acquiring voice input and recognizing emotional states, means for converting the acquired voice into text data, and means for analyzing the converted text data and emotional data in real time and extracting multifaceted insights while considering background information. This enables the provision of feedback that reflects the emotional states of participants during a meeting and the automatic generation of meeting minutes that include emotional information.

[0160] "Voice input" refers to the digital acquisition of participants' statements during a meeting.

[0161] "Emotional state" refers to information that describes the type and intensity of emotions a participant expresses when speaking.

[0162] "Text data" refers to character information converted from voice input, and is the subject of analysis.

[0163] "Real-time" refers to the temporal characteristic of a system that processes and analyzes information instantly and presents results without delay.

[0164] "Insight" refers to information that facilitates the understanding and insights gained from analyzed data.

[0165] "Feedback" refers to information and suggestions provided to the user based on the analysis results.

[0166] "Meeting minutes" are documents that record the content of discussions and important points during a meeting.

[0167] A "digital storage medium" is a means of electronically storing and managing data.

[0168] This invention is a system that analyzes voice input and emotional state during a meeting in real time and provides immediate feedback. The user first acquires voice input through a terminal. The terminal is equipped with a high-performance microphone and a facial recognition camera, which simultaneously records the user's speech and emotional state. The voice input is converted into text data using speech recognition software. In this process, a general-purpose speech processing API is used for the speech recognition engine.

[0169] The device uses emotion recognition software to recognize emotional states. The voice and emotion data analyzed by the terminal are transmitted to a server via the network. The server analyzes the acquired text data using natural language processing technology and applies a generative AI model as needed. This extracts integrated insights from the analysis of the spoken content and the emotion data.

[0170] The server generates feedback for the user based on the analysis results. This feedback is displayed on the terminal and can be received by the user in real time. For example, in a development meeting of an international team, it is possible to instantly grasp the participants' feelings about the progress of the project and take necessary measures during the meeting.

[0171] The generated data is integrated into meeting minutes, which are automatically generated on the server at the end of the meeting. These minutes include key points based on what was said and the corresponding emotional states, and are stored in cloud storage. This allows users to access them later and use them to help with future decision-making.

[0172] An example of a prompt would be, "Analyze the user's comments and emotional state during this meeting, and incorporate this into the feedback." Through this prompt, users can easily obtain guidance for improving the efficiency of the meeting.

[0173] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0174] Step 1:

[0175] The terminal captures audio during the meeting using a high-performance microphone. At the same time, a facial recognition camera is also used to record the participants' facial expressions, and the emotional state of the speaker is inferred through voice intonation analysis software. The input here is raw audio data and camera footage, and the output is audio data and emotional state data. Specifically, the system analyzes the audio waveform and detects changes in voice tone and pitch, allowing the emotion engine to tag emotions such as "joy" and "sadness."

[0176] Step 2:

[0177] The terminal transmits the acquired voice data and emotional state data to the server via a secure network. The input is the voice data and emotional state data obtained in the previous step, and the output is a data packet whose integrity has been verified. Specifically, the data is encrypted and a checksum is included during transmission to ensure data integrity.

[0178] Step 3:

[0179] The server uses a speech recognition engine to convert audio data into text data. The input is audio data, and the output is text data. Specifically, it performs dictionary lookups and phoneme analysis to convert speech to text, and performs correction processing to prevent misrecognition, especially for specialized terminology.

[0180] Step 4:

[0181] The server uses natural language processing techniques to comprehensively analyze text and sentiment data and extract insights. The input is text data and associated sentiment data, while the output is analyzed insights and suggestible feedback. Specifically, it performs keyword extraction from utterances, contextual understanding, and trend analysis of sentiment changes.

[0182] Step 5:

[0183] The server uses an AI model to generate feedback based on the analysis results and presents it to the user via the terminal. The input is the analyzed insights, and the output is a feedback message. Specifically, prompts are sent to the AI ​​model, which generates feedback in natural language. For example, it might display, "This statement received a lot of support."

[0184] Step 6:

[0185] The server automatically generates meeting minutes and extracts key points using the data accumulated after the meeting. Input consists of the entire text data and sentiment data, while output is a summarized meeting minute. Specific operations include ranking important statements and creating graphs showing the intensity of emotions.

[0186] Step 7:

[0187] The server saves the generated meeting minutes to secure cloud storage, making them accessible later. The input is the automatically generated meeting minutes, and the output is the saved data. Specific actions include adding classification tags before saving and setting specific access permissions within the storage.

[0188] (Application Example 2)

[0189] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0190] In modern meetings and discussions, it is difficult to accurately grasp not only the content of participants' statements but also the emotions and intentions behind them. This can lead to misunderstandings, a lack of communication, and decreased communication efficiency. In particular, in workplaces such as factories, where rapid decision-making and appropriate feedback are crucial to work efficiency, real-time information provision, including emotional recognition, is required.

[0191] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0192] In this invention, the server includes means for acquiring voice input, means for converting the acquired voice into text data, means for analyzing the converted text data in real time and extracting multifaceted insights considering cultural, professional, and generational backgrounds, means for presenting the extracted insights to the user, means for automatically generating meeting minutes that summarize the important points of the meeting, means for acquiring audio and video and recognizing the emotional state of participants in real time, and means for providing feedback to the user based on emotional data. As a result, information integrating the content of speech and emotions is provided, enabling efficient and accurate communication.

[0193] "Means for acquiring voice input" refers to a general term for devices and software used to collect voice data from people's conversations in meetings, work environments, and other settings.

[0194] "Means of converting to text data" refers to speech recognition technology and related software used to analyze acquired audio data and convert it into text information.

[0195] "A means of analyzing and extracting insights in real time" refers to a system that instantly analyzes converted text data and extracts the information necessary to gain a multifaceted understanding.

[0196] "Means of presenting insights to users" refers to interfaces and display devices that clearly show and communicate information derived from analysis results to users.

[0197] "Methods for automatically generating meeting minutes" refers to technologies or algorithms that automatically create documents summarizing the content of discussions at a meeting and highlighting the key points.

[0198] A "means for acquiring audio and video and recognizing emotional states" is a system that uses audio and visual data to identify emotions by analyzing the tone of voice and facial expressions of the speaker.

[0199] "Means of providing feedback based on emotional data" refers to applications or devices that utilize recognized emotional information to provide appropriate advice or responses tailored to the user's situation.

[0200] This invention is a system that analyzes voice input in real time, comprehensively understands its content and emotions, and provides appropriate feedback to the user. Specifically, it is implemented using the following hardware and software.

[0201] The server first acquires audio and video data from devices connected to microphones and cameras. This acquired data is converted into text data by speech recognition engines such as Google Cloud Speech-to-Text or Amazon Transcribe. Next, an emotion recognition library such as Affectiva SDK is used to analyze the user's facial expressions from the acquired video and extract emotion data.

[0202] The server uses this speech-text data and sentiment data to apply natural language processing technologies such as Google Cloud Natural Language to analyze the context of the speech. This process yields multifaceted insights that take into account cultural, professional, and generational backgrounds. Furthermore, by fusing sentiment data, the insights are delivered in a way that is relevant to the user.

[0203] The terminal provides feedback to the user based on analysis results sent from the server. This feedback is delivered in real time via the display and speakers, facilitating user understanding. For example, during a factory meeting, analyzing the statements and emotions of workers at the time can enable revisions to instructions and more persuasive suggestions.

[0204] As a concrete example, in a factory production meeting, this system could be used to immediately propose improvement measures if stress or anxiety is detected along with the content of each worker's comments. An example of a prompt message in this case would be: "Based on Mr. / Ms. A's facial expression and comments, it is clear that they are experiencing significant anxiety regarding process B. We will support your suggestion for process improvement."

[0205] This enables a system where the server provides insights that integrate the content and emotions of what is said, helping users make efficient and accurate decisions.

[0206] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0207] Step 1:

[0208] The terminal activates its microphone and camera at the start of the meeting, and continuously acquires the participants' audio and video. The input here is real-time audio and video data, which forms the basis for analysis in the next processing step. The output is raw audio and video data sent to the server.

[0209] Step 2:

[0210] The server receives audio data and converts it into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text). The input for this step is audio data, which is converted from an acoustic signal to text. The output is text data suitable for analysis.

[0211] Step 3:

[0212] The server analyzes video data using emotion recognition libraries such as the Affectiva SDK, extracting emotional data from the participants' facial expressions. The input is video data, where emotions are identified from facial movements and changes in facial muscles. The output is data indicating the user's emotional state.

[0213] Step 4:

[0214] The server uses natural language processing technology (e.g., Google Cloud Natural Language) to analyze text data and extract insights from the content and context of the statements. The input is the text data generated in step 2, and a language model is used to analyze the meaning and intent of the content. The output is structured information as insights.

[0215] Step 5:

[0216] The input for the server to fuse insights and sentiment data and generate feedback for the user is the sentiment data and insights obtained in steps 3 and 4. This feedback is tailored to help the user understand and is then output. Specifically, when in operation, a prompt generation AI model creates prompt sentences that are displayed or spoken in real time at the appropriate time.

[0217] Step 6:

[0218] The terminal receives feedback from the server and presents the content to the user using a display or speaker. The input here is the feedback information, which is the output of step 5. Specifically, during a factory meeting, a message such as "We support your process improvement suggestions" is conveyed visually or audibly. The output is an interaction provided in a way that facilitates user understanding.

[0219] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0220] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0221] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0222] [Second Embodiment]

[0223] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0224] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0225] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0226] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0227] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0228] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0229] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0230] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0231] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0232] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0233] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0234] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0235] This invention is an embodiment of a system that analyzes the content of meeting discussions in real time and provides multifaceted insights by considering cultural, professional, and generational backgrounds. To achieve this, the following program structure is employed.

[0236] 1. Acquisition of voice input

[0237] The device uses a high-quality microphone to capture participants' speech during the meeting as audio data. This allows for clear and low-noise audio.

[0238] 2. Speech-to-text conversion

[0239] Once audio is acquired, the server receives the audio data and uses a speech recognition engine to convert it into text data. The converted text is then used in the next parsing step.

[0240] 3. A multifaceted analysis of the statements made.

[0241] The server uses natural language processing techniques to analyze the background information contained in the text data. This includes participants' cultural backgrounds, expertise, and generational differences. This analysis generates multifaceted insights.

[0242] 4. Providing real-time feedback

[0243] Based on the analysis results, the server extracts different perspectives and key insights in real time during the meeting and presents them to the user via the terminal. This allows for faster and more accurate decision-making during the meeting.

[0244] 5. Automatic generation of meeting minutes

[0245] After the meeting ends, the server automatically generates meeting minutes summarizing all the points made and their analysis. These minutes can be referenced later, significantly improving meeting efficiency.

[0246] As a concrete example, consider a progress meeting for an international project. Each participant has a different cultural background and professional perspective. In this setting, the terminal captures the spoken audio, and the server analyzes the background information. This deepens the understanding of the participants' differing opinions from their respective fields of expertise, enabling appropriate project coordination. Finally, the server summarizes the key points of the meeting and generates concise and easy-to-understand minutes, providing convenience for later review.

[0247] The following describes the processing flow.

[0248] Step 1:

[0249] The device captures the speech of meeting participants in real time via the microphone and saves it as audio data.

[0250] Step 2:

[0251] The device compresses the acquired audio data and sends it to the server via the network.

[0252] Step 3:

[0253] The server processes the received audio data through a speech recognition engine to convert the audio into text data. Preprocessing, such as noise filtering, is also performed to improve the accuracy of the conversion results.

[0254] Step 4:

[0255] The server passes the converted text data to an analysis module, which uses natural language processing techniques to comprehensively analyze the content of the speech. This includes analysis of cultural, professional, and generational backgrounds.

[0256] Step 5:

[0257] The server organizes the analysis results, extracting key insights and different perspectives. This information is then compiled as real-time feedback.

[0258] Step 6:

[0259] The server extracts feedback information and sends it to the terminal, which then presents it to the user via screen display or audio.

[0260] Step 7:

[0261] After the meeting ends, the server automatically generates meeting minutes summarizing the key points based on all the analysis results. These minutes are saved to cloud storage, making them accessible to users later.

[0262] (Example 1)

[0263] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0264] In today's diverse meetings involving participants from various backgrounds, it is essential to immediately and accurately understand what is said and quickly gain multifaceted insights based on that understanding. However, when these processes are carried out manually, they are time-consuming and lead to delays in decision-making. Furthermore, insights that do not consider cultural, professional, and generational differences are insufficient. In addition, creating meeting minutes after the meeting is another time-consuming and laborious task.

[0265] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0266] In this invention, the server includes a device for acquiring voice input, a device for converting the acquired voice into text information, and a device for analyzing the converted text information in real time and extracting multifaceted insights considering cultural, professional, and generational backgrounds. This makes it possible to quickly and comprehensively analyze the content of speech during meetings and provide users with immediate insights. Furthermore, the automatic generation of meeting minutes summarizing the key points of the meeting allows for efficient review afterward.

[0267] "Voice input" is the process of acquiring audio signals generated during meetings or conversations.

[0268] A "device" is a piece of equipment or system designed to perform a specific function.

[0269] "Text information" refers to information obtained by converting audio data into written characters.

[0270] "Real-time analysis" refers to a process where data is processed and analyzed immediately as soon as it is generated.

[0271] "Extracting insights from multiple perspectives" is the process of considering different viewpoints and backgrounds to derive useful information and understanding.

[0272] "Cultural, professional, and generational background" refers to the set of characteristics and influences related to an individual's culture, professional knowledge, and age.

[0273] "Insight" refers to the insights and understanding gained from specific data or information.

[0274] A "user" is a person or group that uses a system or product.

[0275] "Key points" are elements or content within information or events that deserve particular attention or have high priority.

[0276] "Meeting minutes" are documents that record the content and decisions discussed and made at meetings or discussions.

[0277] A "communication protocol for rapid processing" refers to a set of rules and procedures for sending and receiving data smoothly and efficiently.

[0278] "Audio processing technology for noise reduction" refers to technology used to reduce undesirable noise and interference in audio signals.

[0279] This invention is a system that improves the effectiveness of meetings involving multiple participants. The system acquires audio input from meetings, converts it into text information, and then analyzes it in real time to provide users with multifaceted insights that take cultural, professional, and generational backgrounds into consideration. The following hardware and software are used to realize this configuration.

[0280] The terminal uses a high-quality microphone to acquire the speech during the meeting as audio. Through noise cancellation technology, the terminal minimizes noise and transmits clear audio data to the server. The server applies the received audio data to an audio recognition engine (e.g., a common audio recognition service) to convert the audio into text information. Furthermore, the server utilizes natural language processing technology, specifically generative AI models such as BERT or GPT, to analyze this text information and generate comprehensive insights. This enables the extraction of useful insights in real-time, taking into account the cultural and professional background information of the participants.

[0281] As a specific example, consider an international project meeting. In this meeting, there are multiple participants with different cultural backgrounds and professional perspectives. By the terminal acquiring the speech of each participant as audio and the server quickly processing that audio, users can understand multi-faceted opinions in real-time and make appropriate decisions regarding the progress of the project. It is expected that the progress will be smoother and the results will be improved.

[0282] In this system, in order for users to make more accurate and effective meeting decisions, it is possible to input prompt texts utilizing generative AI models. For example, a prompt text such as "Automatically generate the minutes of this meeting and emphasize insights based on cultural background and professional perspectives" can be used. This enables the efficient recording and analysis of the meeting content, and information is provided in a format suitable for later reference.

[0283] The flow of the specific process in Example 1 will be described using FIG. 11.

[0284] Step 1:

[0285] The terminal uses a high-quality microphone to acquire the audio during the meeting. The terminal utilizes a noise cancellation function to reduce noise and collect clear and high-precision audio data. The input here is the speech of the participants, and the output is digital audio data.

[0286] Step 2:

[0287] The terminal sends the collected voice data to the server. Due to the high-speed communication protocol used, the voice data reaches the server almost without delay. The input is the voice data from the terminal, and the output is the voice data transferred onto the server.

[0288] Step 3:

[0289] The server applies the received voice data to a voice recognition engine and converts it into text data. The specific operation here is to analyze the sounds of words by a voice recognition algorithm to generate corresponding text. The input is the voice data, and the output is character information.

[0290] Step 4:

[0291] The server analyzes the converted text data using a generative AI model. Specifically, it utilizes natural language processing technology to extract the meaning of the text and generate insights considering cultural, professional, and generational backgrounds. The input is the text data, and the output is the analyzed insight information.

[0292] Step 5:

[0293] Based on the analysis results, the server generates multifaceted perspectives and important insights for the user and provides them through the terminal. The user utilizes this feedback to deepen the discussions during the meeting. The input is the analyzed insight information, and the output is the feedback information presented to the user.

[0294] Step 6:

[0295] After the meeting ends, the server automatically generates minutes summarizing the important points of the meeting based on all the speeches and analysis results. As a specific operation, it utilizes natural language generation technology to create a document that is easy to read and well-organized. The input is all the data generated during the meeting, and the output is the summary minutes.

[0296] (Application Example 1)

[0297] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0298] In autonomous vehicles, there is a need to facilitate communication among passengers, provide services based on their diverse backgrounds and interests, and improve the ride experience. However, currently, there is a lack of means to provide appropriate insights in real time based on what passengers say and to optimize the driving route and in-vehicle environment. A new system is needed to solve this problem.

[0299] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0300] In this invention, the server includes means for acquiring voice information, means for converting it into symbolic data, and means for extracting insights considering cultural background and expertise. This makes it possible to analyze passenger conversations in real time and optimize routes and services based on their interests.

[0301] "Voice information" refers to the recording of voices spoken by users, converted into data.

[0302] "Symbol sequence data" refers to text data obtained by converting audio information into characters and symbols.

[0303] "Insight" refers to a deep understanding and knowledge gained from analyzed data, and is information that is useful for decision-making.

[0304] A "user" is an individual or organization that uses the system to obtain information.

[0305] "Automatic generation" means that a system or program automatically creates data or documents without human intervention.

[0306] A "vehicle" is a conveyance for carrying people or goods on land.

[0307] A "proposal" is an act of presenting specific options or action plans.

[0308] This system aims to facilitate communication among passengers in an autonomous vehicle and improve the riding experience. The server acquires voice information using a high-quality microphone. The acquired voice information is converted into symbol sequence data by a computer in the vehicle. Google Speech-to-Text API is used for this conversion. The server analyzes the converted symbol sequence data using a natural language processing model operating on Azure or AWS and extracts insights considering the cultural background and professional knowledge of the passengers.

[0309] For example, when a passenger is talking about "wanting to visit a nearby tourist attraction", the server searches for the nearest tourist spots using Google Maps API and proposes an appropriate route in real time. This enables passengers to receive on-demand services according to their individual needs.

[0310] As an example of a prompt sentence, it can be described as "Analyze conversations during driving and propose recommendations based on places to visit or activities of interest". This system enables real-time feedback and highly adaptable service provision, improving passenger satisfaction.

[0311] The flow of specific processing in Application Example 1 will be described using FIG. 12.

[0312] Step 1:

[0313] The terminal acquires voice information inside the vehicle using a high-quality microphone. This obtains clear voice data with minimized noise. This voice data is transmitted as an input to the server.

[0314] Step 2:

[0315] The server converts the acquired audio information into symbolic text data using the Google Speech-to-Text API. In this step, the audio data is automatically converted to text and used for the analysis process within the server. The converted symbolic text data is then output.

[0316] Step 3:

[0317] The server uses a natural language processing model trained with TensorFlow to analyze the symbolic data. During this analysis, a generative AI model is applied to identify passengers' intentions and interests, extracting multifaceted insights. The analysis results in the output of insightful data.

[0318] Step 4:

[0319] The server generates appropriate suggestions for the user based on the extracted insight data. Specifically, it uses the Google Maps API to collect destination information and calculate the optimal route. This generates the suggestions that are presented to the user.

[0320] Step 5:

[0321] The user receives suggestions sent from the server via an in-vehicle display or voice prompt. At this stage, possible route information and activity suggestions are output and presented to the user as options.

[0322] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0323] This invention combines an emotion engine with a system that analyzes audio data acquired during meetings, thereby recognizing the user's emotions along with the content of their speech and providing real-time feedback. The following program structure is used to implement this invention.

[0324] 1. Voice input acquisition and emotion recognition

[0325] The device acquires participants' statements during the meeting and uses an emotion engine to recognize the user's emotional state through facial recognition cameras and voice intonation analysis. This emotion data is then associated with each statement.

[0326] 2. Transmission of voice and emotion data

[0327] The device transmits the acquired voice and emotion data to the server in real time. The server then prepares to process this data in an integrated manner.

[0328] 3. Analysis and conversion of audio data

[0329] The server converts the audio data into text data using a speech recognition engine. Furthermore, it utilizes natural language processing technology to analyze the content and background information of the speech, and also incorporates emotion recognition data into the analysis.

[0330] 4. Providing insights and emotion-oriented feedback

[0331] The server extracts insights from multiple perspectives based on analysis results and emotional data. The device then presents the user with real-time feedback, adjusted for user acceptance and empathy, taking into account the emotional recognition results.

[0332] 5. Automated generation of meeting minutes and integration of sentiment data

[0333] After the meeting ends, the server summarizes the key points based on the content of the speeches and sentiment data, and automatically generates meeting minutes. These minutes also incorporate sentiment data, allowing for an understanding of the emotional context of the speeches. The generated data is saved to cloud storage in a format that can be accessed later.

[0334] As a concrete example, considering a development meeting involving an international team, it becomes easier to understand the level of employee motivation and concerns regarding the project's progress during the meeting. This allows the project manager to take necessary measures during the meeting and effectively promote communication among stakeholders. Furthermore, the generation of meeting minutes, which include changes in emotions, can be used for future decision-making.

[0335] The following describes the processing flow.

[0336] Step 1:

[0337] The device uses microphones and cameras installed in the meeting room to capture participants' speech and facial expressions in real time. Both audio and video data are temporarily stored in local storage.

[0338] Step 2:

[0339] The device compresses the stored audio data, performs noise filtering, and then sends it to the server via the network. Simultaneously, it sends facial expression information obtained from video data to the server as a feature vector.

[0340] Step 3:

[0341] The server analyzes the received audio data using a speech recognition engine and converts it into text data. During this process, an acoustic model is applied to improve the accuracy of speech recognition.

[0342] Step 4:

[0343] The server analyzes text data using natural language processing technology to comprehensively analyze the meaning of the utterances. This analysis includes extracting field-specific keywords and understanding the context.

[0344] Step 5:

[0345] The server simultaneously uses an emotion engine based on the received facial expression data to recognize the emotional state of the participants. Emotional states are classified into categories such as "joy," "surprise," "sadness," and "anger."

[0346] Step 6:

[0347] The server integrates emotion recognition data with the spoken content and generates context-appropriate, emotion-oriented feedback. The generated feedback is adjusted to a tone that is considerate of the recipient of the statement.

[0348] Step 7:

[0349] The device displays this feedback to the user in real time. This clarifies the intent and nuances during the meeting and improves communication among participants.

[0350] Step 8:

[0351] After the meeting ends, the server automatically generates meeting minutes, summarizing all spoken and emotional data to highlight key points. These minutes also record emotional changes during speaking, which can be used for later analysis and decision-making.

[0352] Step 9:

[0353] The generated meeting minutes and related data are saved to cloud storage by the server, allowing users to access them securely later.

[0354] (Example 2)

[0355] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0356] In meetings, it is essential to understand not only the content of participants' statements but also their emotional state at the time of speaking, and to provide appropriate feedback immediately. However, existing systems are limited to speech-to-text conversion and simple content analysis, and do not adequately grasp emotional states or provide insights based on them. As a result, the quality of meetings does not improve, and it is a factor that diminishes participants' motivation.

[0357] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0358] In this invention, the server includes means for acquiring voice input and recognizing emotional states, means for converting the acquired voice into text data, and means for analyzing the converted text data and emotional data in real time and extracting multifaceted insights while considering background information. This enables the provision of feedback that reflects the emotional states of participants during a meeting and the automatic generation of meeting minutes that include emotional information.

[0359] "Voice input" refers to the digital acquisition of participants' statements during a meeting.

[0360] "Emotional state" refers to information that describes the type and intensity of emotions a participant expresses when speaking.

[0361] "Text data" refers to character information converted from voice input, and is the subject of analysis.

[0362] "Real-time" refers to the temporal characteristic of a system that processes and analyzes information instantly and presents results without delay.

[0363] "Insight" refers to information that facilitates the understanding and insights gained from analyzed data.

[0364] "Feedback" refers to information and suggestions provided to the user based on the analysis results.

[0365] "Meeting minutes" are documents that record the content of discussions and important points during a meeting.

[0366] A "digital storage medium" is a means of electronically storing and managing data.

[0367] This invention is a system that analyzes voice input and emotional state during a meeting in real time and provides immediate feedback. The user first acquires voice input through a terminal. The terminal is equipped with a high-performance microphone and a facial recognition camera, which simultaneously records the user's speech and emotional state. The voice input is converted into text data using speech recognition software. In this process, a general-purpose speech processing API is used for the speech recognition engine.

[0368] The device uses emotion recognition software to recognize emotional states. The voice and emotion data analyzed by the terminal are transmitted to a server via the network. The server analyzes the acquired text data using natural language processing technology and applies a generative AI model as needed. This extracts integrated insights from the analysis of the spoken content and the emotion data.

[0369] The server generates feedback for the user based on the analysis results. This feedback is displayed on the terminal and can be received by the user in real time. For example, in a development meeting of an international team, it is possible to instantly grasp the participants' feelings about the progress of the project and take necessary measures during the meeting.

[0370] The generated data is integrated into meeting minutes, which are automatically generated on the server at the end of the meeting. These minutes include key points based on what was said and the corresponding emotional states, and are stored in cloud storage. This allows users to access them later and use them to help with future decision-making.

[0371] An example of a prompt would be, "Analyze the user's comments and emotional state during this meeting, and incorporate this into the feedback." Through this prompt, users can easily obtain guidance for improving the efficiency of the meeting.

[0372] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0373] Step 1:

[0374] The terminal captures audio during the meeting using a high-performance microphone. At the same time, a facial recognition camera is also used to record the participants' facial expressions, and the emotional state of the speaker is inferred through voice intonation analysis software. The input here is raw audio data and camera footage, and the output is audio data and emotional state data. Specifically, the system analyzes the audio waveform and detects changes in voice tone and pitch, allowing the emotion engine to tag emotions such as "joy" and "sadness."

[0375] Step 2:

[0376] The terminal transmits the acquired voice data and emotional state data to the server via a secure network. The input is the voice data and emotional state data obtained in the previous step, and the output is a data packet whose integrity has been verified. Specifically, the data is encrypted and a checksum is included during transmission to ensure data integrity.

[0377] Step 3:

[0378] The server uses a speech recognition engine to convert audio data into text data. The input is audio data, and the output is text data. Specifically, it performs dictionary lookups and phoneme analysis to convert speech to text, and performs correction processing to prevent misrecognition, especially for specialized terminology.

[0379] Step 4:

[0380] The server uses natural language processing techniques to comprehensively analyze text and sentiment data and extract insights. The input is text data and associated sentiment data, while the output is analyzed insights and suggestible feedback. Specifically, it performs keyword extraction from utterances, contextual understanding, and trend analysis of sentiment changes.

[0381] Step 5:

[0382] The server uses an AI model to generate feedback based on the analysis results and presents it to the user via the terminal. The input is the analyzed insights, and the output is a feedback message. Specifically, prompts are sent to the AI ​​model, which generates feedback in natural language. For example, it might display, "This statement received a lot of support."

[0383] Step 6:

[0384] The server automatically generates meeting minutes and extracts key points using the data accumulated after the meeting. Input consists of the entire text data and sentiment data, while output is a summarized meeting minute. Specific operations include ranking important statements and creating graphs showing the intensity of emotions.

[0385] Step 7:

[0386] The server saves the generated meeting minutes to secure cloud storage, making them accessible later. The input is the automatically generated meeting minutes, and the output is the saved data. Specific actions include adding classification tags before saving and setting specific access permissions within the storage.

[0387] (Application Example 2)

[0388] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0389] In modern meetings and discussions, it is difficult to accurately grasp not only the content of participants' statements but also the emotions and intentions behind them. This can lead to misunderstandings, a lack of communication, and decreased communication efficiency. In particular, in workplaces such as factories, where rapid decision-making and appropriate feedback are crucial to work efficiency, real-time information provision, including emotional recognition, is required.

[0390] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0391] In this invention, the server includes means for acquiring voice input, means for converting the acquired voice into text data, means for analyzing the converted text data in real time and extracting multifaceted insights considering cultural, professional, and generational backgrounds, means for presenting the extracted insights to the user, means for automatically generating meeting minutes that summarize the important points of the meeting, means for acquiring audio and video and recognizing the emotional state of participants in real time, and means for providing feedback to the user based on emotional data. As a result, information integrating the content of speech and emotions is provided, enabling efficient and accurate communication.

[0392] "Means for acquiring voice input" refers to a general term for devices and software used to collect voice data from people's conversations in meetings, work environments, and other settings.

[0393] "Means of converting to text data" refers to speech recognition technology and related software used to analyze acquired audio data and convert it into text information.

[0394] "A means of analyzing and extracting insights in real time" refers to a system that instantly analyzes converted text data and extracts the information necessary to gain a multifaceted understanding.

[0395] "Means of presenting insights to users" refers to interfaces and display devices that clearly show and communicate information derived from analysis results to users.

[0396] "Methods for automatically generating meeting minutes" refers to technologies or algorithms that automatically create documents summarizing the content of discussions at a meeting and highlighting the key points.

[0397] A "means for acquiring audio and video and recognizing emotional states" is a system that uses audio and visual data to identify emotions by analyzing the tone of voice and facial expressions of the speaker.

[0398] "Means of providing feedback based on emotional data" refers to applications or devices that utilize recognized emotional information to provide appropriate advice or responses tailored to the user's situation.

[0399] This invention is a system that analyzes voice input in real time, comprehensively understands its content and emotions, and provides appropriate feedback to the user. Specifically, it is implemented using the following hardware and software.

[0400] The server first acquires audio and video data from devices connected to microphones and cameras. This acquired data is converted into text data by speech recognition engines such as Google Cloud Speech-to-Text or Amazon Transcribe. Next, an emotion recognition library such as Affectiva SDK is used to analyze the user's facial expressions from the acquired video and extract emotion data.

[0401] The server uses this speech-text data and sentiment data to apply natural language processing technologies such as Google Cloud Natural Language to analyze the context of the speech. This process yields multifaceted insights that take into account cultural, professional, and generational backgrounds. Furthermore, by fusing sentiment data, the insights are delivered in a way that is relevant to the user.

[0402] The terminal provides feedback to the user based on analysis results sent from the server. This feedback is delivered in real time via the display and speakers, facilitating user understanding. For example, during a factory meeting, analyzing the statements and emotions of workers at the time can enable revisions to instructions and more persuasive suggestions.

[0403] As a concrete example, in a factory production meeting, this system could be used to immediately propose improvement measures if stress or anxiety is detected along with the content of each worker's comments. An example of a prompt message in this case would be: "Based on Mr. / Ms. A's facial expression and comments, it is clear that they are experiencing significant anxiety regarding process B. We will support your suggestion for process improvement."

[0404] This enables a system where the server provides insights that integrate the content and emotions of what is said, helping users make efficient and accurate decisions.

[0405] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0406] Step 1:

[0407] The terminal activates its microphone and camera at the start of the meeting, and continuously acquires the participants' audio and video. The input here is real-time audio and video data, which forms the basis for analysis in the next processing step. The output is raw audio and video data sent to the server.

[0408] Step 2:

[0409] The server receives audio data and converts it into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text). The input for this step is audio data, which is converted from an acoustic signal to text. The output is text data suitable for analysis.

[0410] Step 3:

[0411] The server analyzes video data using emotion recognition libraries such as the Affectiva SDK, extracting emotional data from the participants' facial expressions. The input is video data, where emotions are identified from facial movements and changes in facial muscles. The output is data indicating the user's emotional state.

[0412] Step 4:

[0413] The server uses natural language processing technology (e.g., Google Cloud Natural Language) to analyze text data and extract insights from the content and context of the statements. The input is the text data generated in step 2, and a language model is used to analyze the meaning and intent of the content. The output is structured information as insights.

[0414] Step 5:

[0415] The input for the server to fuse insights and sentiment data and generate feedback for the user is the sentiment data and insights obtained in steps 3 and 4. This feedback is tailored to help the user understand and is then output. Specifically, when in operation, a prompt generation AI model creates prompt sentences that are displayed or spoken in real time at the appropriate time.

[0416] Step 6:

[0417] The terminal receives feedback from the server and presents the content to the user using a display or speaker. The input here is the feedback information, which is the output of step 5. Specifically, during a factory meeting, a message such as "We support your process improvement suggestions" is conveyed visually or audibly. The output is an interaction provided in a way that facilitates user understanding.

[0418] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0419] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0420] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0421] [Third Embodiment]

[0422] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0423] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0424] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0425] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0426] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0427] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0428] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0429] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0430] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0431] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0432] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0433] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0434] This invention is an embodiment of a system that analyzes the content of meeting discussions in real time and provides multifaceted insights by considering cultural, professional, and generational backgrounds. To achieve this, the following program structure is employed.

[0435] 1. Acquisition of voice input

[0436] The device uses a high-quality microphone to capture participants' speech during the meeting as audio data. This allows for clear and low-noise audio.

[0437] 2. Speech-to-text conversion

[0438] Once audio is acquired, the server receives the audio data and uses a speech recognition engine to convert it into text data. The converted text is then used in the next parsing step.

[0439] 3. A multifaceted analysis of the statements made.

[0440] The server uses natural language processing techniques to analyze the background information contained in the text data. This includes participants' cultural backgrounds, expertise, and generational differences. This analysis generates multifaceted insights.

[0441] 4. Providing real-time feedback

[0442] Based on the analysis results, the server extracts different perspectives and key insights in real time during the meeting and presents them to the user via the terminal. This allows for faster and more accurate decision-making during the meeting.

[0443] 5. Automatic generation of meeting minutes

[0444] After the meeting ends, the server automatically generates meeting minutes summarizing all the points made and their analysis. These minutes can be referenced later, significantly improving meeting efficiency.

[0445] As a concrete example, consider a progress meeting for an international project. Each participant has a different cultural background and professional perspective. In this setting, the terminal captures the spoken audio, and the server analyzes the background information. This deepens the understanding of the participants' differing opinions from their respective fields of expertise, enabling appropriate project coordination. Finally, the server summarizes the key points of the meeting and generates concise and easy-to-understand minutes, providing convenience for later review.

[0446] The following describes the processing flow.

[0447] Step 1:

[0448] The device captures the speech of meeting participants in real time via the microphone and saves it as audio data.

[0449] Step 2:

[0450] The device compresses the acquired audio data and sends it to the server via the network.

[0451] Step 3:

[0452] The server processes the received audio data through a speech recognition engine to convert the audio into text data. Preprocessing, such as noise filtering, is also performed to improve the accuracy of the conversion results.

[0453] Step 4:

[0454] The server passes the converted text data to an analysis module, which uses natural language processing techniques to comprehensively analyze the content of the speech. This includes analysis of cultural, professional, and generational backgrounds.

[0455] Step 5:

[0456] The server organizes the analysis results, extracting key insights and different perspectives. This information is then compiled as real-time feedback.

[0457] Step 6:

[0458] The server extracts feedback information and sends it to the terminal, which then presents it to the user via screen display or audio.

[0459] Step 7:

[0460] After the meeting ends, the server automatically generates meeting minutes summarizing the key points based on all the analysis results. These minutes are saved to cloud storage, making them accessible to users later.

[0461] (Example 1)

[0462] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0463] In today's diverse meetings involving participants from various backgrounds, it is essential to immediately and accurately understand what is said and quickly gain multifaceted insights based on that understanding. However, when these processes are carried out manually, they are time-consuming and lead to delays in decision-making. Furthermore, insights that do not consider cultural, professional, and generational differences are insufficient. In addition, creating meeting minutes after the meeting is another time-consuming and laborious task.

[0464] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0465] In this invention, the server includes a device for acquiring voice input, a device for converting the acquired voice into text information, and a device for analyzing the converted text information in real time and extracting multifaceted insights considering cultural, professional, and generational backgrounds. This makes it possible to quickly and comprehensively analyze the content of speech during meetings and provide users with immediate insights. Furthermore, the automatic generation of meeting minutes summarizing the key points of the meeting allows for efficient review afterward.

[0466] "Voice input" is the process of acquiring audio signals generated during meetings or conversations.

[0467] A "device" is a piece of equipment or system designed to perform a specific function.

[0468] "Text information" refers to information obtained by converting audio data into written characters.

[0469] "Real-time analysis" refers to a process where data is processed and analyzed immediately as soon as it is generated.

[0470] "Extracting insights from multiple perspectives" is the process of considering different viewpoints and backgrounds to derive useful information and understanding.

[0471] "Cultural, professional, and generational background" refers to the set of characteristics and influences related to an individual's culture, professional knowledge, and age.

[0472] "Insight" refers to the insights and understanding gained from specific data or information.

[0473] A "user" is a person or group that uses a system or product.

[0474] "Key points" are elements or content within information or events that deserve particular attention or have high priority.

[0475] "Meeting minutes" are documents that record the content and decisions discussed and made at meetings or discussions.

[0476] A "communication protocol for rapid processing" refers to a set of rules and procedures for sending and receiving data smoothly and efficiently.

[0477] "Audio processing technology for noise reduction" refers to technology used to reduce undesirable noise and interference in audio signals.

[0478] This invention is a system that improves the effectiveness of meetings involving multiple participants. The system acquires audio input from meetings, converts it into text information, and then analyzes it in real time to provide users with multifaceted insights that take cultural, professional, and generational backgrounds into consideration. The following hardware and software are used to realize this configuration.

[0479] The terminal uses a high-quality microphone to capture speech during the meeting as audio. Noise cancellation technology minimizes background noise, sending clear audio data to the server. The server processes the received audio data through a speech recognition engine (e.g., a general speech recognition service) to convert the speech into text. Furthermore, the server uses natural language processing technology, specifically generative AI models such as BERT and GPT, to analyze this text information and generate multifaceted insights. This makes it possible to extract valuable insights in real time, taking into account the cultural and professional backgrounds of the participants.

[0480] As a concrete example, consider an international project meeting. This meeting includes multiple participants with different cultural backgrounds and professional perspectives. By having a terminal capture each participant's speech as audio and a server quickly process that audio, users can understand diverse opinions in real time and make appropriate decisions regarding the project's progress. This is expected to lead to smoother progress and improved results.

[0481] This system allows users to input prompts using a generative AI model to enable more accurate and effective meeting decision-making. For example, a prompt such as "Automatically generate minutes for this meeting, highlighting insights based on cultural context and expert perspectives" can be used. This ensures that meeting content is efficiently recorded and analyzed, and the information is provided in a format suitable for future reference.

[0482] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0483] Step 1:

[0484] The device uses a high-quality microphone to capture audio during the meeting. It utilizes noise cancellation to reduce background noise and collect clear, high-precision audio data. The input here is the participants' speech, and the output is digital audio data.

[0485] Step 2:

[0486] The terminal sends the collected audio data to the server. Due to the high-speed communication protocol used, the audio data reaches the server with virtually no delay. The input is the audio data from the terminal, and the output is the audio data transferred to the server.

[0487] Step 3:

[0488] The server processes the received audio data through a speech recognition engine and converts it into text data. Specifically, the speech recognition algorithm analyzes the sounds of words and generates the corresponding text. The input is audio data, and the output is text information.

[0489] Step 4:

[0490] The server analyzes the converted text data using a generation AI model. Specifically, it uses natural language processing techniques to extract the meaning of the text and generate insights that take into account cultural, professional, and generational contexts. The input is text data, and the output is analyzed insightful information.

[0491] Step 5:

[0492] Based on the analysis results, the server generates multifaceted perspectives and key insights for the user, which are then provided via the terminal. Users utilize this feedback to deepen discussions during meetings. The input is the analyzed insights, and the output is the feedback information presented to the user.

[0493] Step 6:

[0494] After the meeting ends, the server automatically generates meeting minutes summarizing the key points based on all the statements and analysis results. Specifically, it utilizes natural language generation technology to create a readable and well-organized document. The input is all the data generated during the meeting, and the output is the summarized meeting minutes.

[0495] (Application Example 1)

[0496] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0497] In autonomous vehicles, there is a need to facilitate communication among passengers, provide services based on their diverse backgrounds and interests, and improve the ride experience. However, currently, there is a lack of means to provide appropriate insights in real time based on what passengers say and to optimize the driving route and in-vehicle environment. A new system is needed to solve this problem.

[0498] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0499] In this invention, the server includes means for acquiring voice information, means for converting it into symbolic data, and means for extracting insights considering cultural background and expertise. This makes it possible to analyze passenger conversations in real time and optimize routes and services based on their interests.

[0500] "Voice information" refers to the recording of voices spoken by users, converted into data.

[0501] "Symbol sequence data" refers to text data obtained by converting audio information into characters and symbols.

[0502] "Insight" refers to a deep understanding and knowledge gained from analyzed data, and is information that is useful for decision-making.

[0503] A "user" is an individual or organization that uses the system to obtain information.

[0504] "Automatic generation" means that a system or program automatically creates data or documents without human intervention.

[0505] A "vehicle" is a vehicle used to transport people or goods on land.

[0506] A "proposal" is the act of presenting specific options or action plans.

[0507] This system aims to improve the ride experience by facilitating communication between passengers in autonomous vehicles. The server acquires voice information using a high-quality microphone. The acquired voice information is converted into symbolic data by the vehicle's computer. The Google Speech-to-Text API is used for this conversion. The server uses a natural language processing model running on Azure or AWS to analyze the converted symbolic data and extract insights that take into account the passengers' cultural backgrounds and expertise.

[0508] For example, if a passenger says they want to visit a nearby tourist attraction, the server uses the Google Maps API to search for the nearest attractions and suggests a suitable route in real time. This allows passengers to receive on-demand services tailored to their individual needs.

[0509] An example of a prompt message could be, "Analyze your conversation while driving and suggest recommendations based on places you want to visit and activities you're interested in." This system enables real-time feedback and highly adaptive service delivery, improving passenger satisfaction.

[0510] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0511] Step 1:

[0512] The terminal uses a high-quality microphone to acquire audio information from inside the vehicle. This results in clear audio data with minimal noise. This audio data is then sent to the server as input.

[0513] Step 2:

[0514] The server converts the acquired audio information into symbolic text data using the Google Speech-to-Text API. In this step, the audio data is automatically converted to text and used for the analysis process within the server. The converted symbolic text data is then output.

[0515] Step 3:

[0516] The server uses a natural language processing model trained with TensorFlow to analyze the symbolic data. During this analysis, a generative AI model is applied to identify passengers' intentions and interests, extracting multifaceted insights. The analysis results in the output of insightful data.

[0517] Step 4:

[0518] The server generates appropriate suggestions for the user based on the extracted insight data. Specifically, it uses the Google Maps API to collect destination information and calculate the optimal route. This generates the suggestions that are presented to the user.

[0519] Step 5:

[0520] The user receives suggestions sent from the server via an in-vehicle display or voice prompt. At this stage, possible route information and activity suggestions are output and presented to the user as options.

[0521] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0522] This invention combines an emotion engine with a system that analyzes audio data acquired during meetings, thereby recognizing the user's emotions along with the content of their speech and providing real-time feedback. The following program structure is used to implement this invention.

[0523] 1. Voice input acquisition and emotion recognition

[0524] The device acquires participants' statements during the meeting and uses an emotion engine to recognize the user's emotional state through facial recognition cameras and voice intonation analysis. This emotion data is then associated with each statement.

[0525] 2. Transmission of voice and emotion data

[0526] The device transmits the acquired voice and emotion data to the server in real time. The server then prepares to process this data in an integrated manner.

[0527] 3. Analysis and conversion of audio data

[0528] The server converts the audio data into text data using a speech recognition engine. Furthermore, it utilizes natural language processing technology to analyze the content and background information of the speech, and also incorporates emotion recognition data into the analysis.

[0529] 4. Providing insights and emotion-oriented feedback

[0530] The server extracts insights from multiple perspectives based on analysis results and emotional data. The device then presents the user with real-time feedback, adjusted for user acceptance and empathy, taking into account the emotional recognition results.

[0531] 5. Automated generation of meeting minutes and integration of sentiment data

[0532] After the meeting ends, the server summarizes the key points based on the content of the speeches and sentiment data, and automatically generates meeting minutes. These minutes also incorporate sentiment data, allowing for an understanding of the emotional context of the speeches. The generated data is saved to cloud storage in a format that can be accessed later.

[0533] As a concrete example, considering a development meeting involving an international team, it becomes easier to understand the level of employee motivation and concerns regarding the project's progress during the meeting. This allows the project manager to take necessary measures during the meeting and effectively promote communication among stakeholders. Furthermore, the generation of meeting minutes, which include changes in emotions, can be used for future decision-making.

[0534] The following describes the processing flow.

[0535] Step 1:

[0536] The device uses microphones and cameras installed in the meeting room to capture participants' speech and facial expressions in real time. Both audio and video data are temporarily stored in local storage.

[0537] Step 2:

[0538] The device compresses the stored audio data, performs noise filtering, and then sends it to the server via the network. Simultaneously, it sends facial expression information obtained from video data to the server as a feature vector.

[0539] Step 3:

[0540] The server analyzes the received audio data using a speech recognition engine and converts it into text data. During this process, an acoustic model is applied to improve the accuracy of speech recognition.

[0541] Step 4:

[0542] The server analyzes text data using natural language processing technology to comprehensively analyze the meaning of the utterances. This analysis includes extracting field-specific keywords and understanding the context.

[0543] Step 5:

[0544] The server simultaneously uses an emotion engine based on the received facial expression data to recognize the emotional state of the participants. Emotional states are classified into categories such as "joy," "surprise," "sadness," and "anger."

[0545] Step 6:

[0546] The server integrates emotion recognition data with the spoken content and generates context-appropriate, emotion-oriented feedback. The generated feedback is adjusted to a tone that is considerate of the recipient of the statement.

[0547] Step 7:

[0548] The device displays this feedback to the user in real time. This clarifies the intent and nuances during the meeting and improves communication among participants.

[0549] Step 8:

[0550] After the meeting ends, the server automatically generates meeting minutes, summarizing all spoken and emotional data to highlight key points. These minutes also record emotional changes during speaking, which can be used for later analysis and decision-making.

[0551] Step 9:

[0552] The generated meeting minutes and related data are saved to cloud storage by the server, allowing users to access them securely later.

[0553] (Example 2)

[0554] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0555] In meetings, it is essential to understand not only the content of participants' statements but also their emotional state at the time of speaking, and to provide appropriate feedback immediately. However, existing systems are limited to speech-to-text conversion and simple content analysis, and do not adequately grasp emotional states or provide insights based on them. As a result, the quality of meetings does not improve, and it is a factor that diminishes participants' motivation.

[0556] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0557] In this invention, the server includes means for acquiring voice input and recognizing emotional states, means for converting the acquired voice into text data, and means for analyzing the converted text data and emotional data in real time and extracting multifaceted insights while considering background information. This enables the provision of feedback that reflects the emotional states of participants during a meeting and the automatic generation of meeting minutes that include emotional information.

[0558] "Voice input" refers to the digital acquisition of participants' statements during a meeting.

[0559] "Emotional state" refers to information that describes the type and intensity of emotions a participant expresses when speaking.

[0560] "Text data" refers to character information converted from voice input, and is the subject of analysis.

[0561] "Real-time" refers to the temporal characteristic of a system that processes and analyzes information instantly and presents results without delay.

[0562] "Insight" refers to information that facilitates the understanding and insights gained from analyzed data.

[0563] "Feedback" refers to information and suggestions provided to the user based on the analysis results.

[0564] "Meeting minutes" are documents that record the content of discussions and important points during a meeting.

[0565] A "digital storage medium" is a means of electronically storing and managing data.

[0566] This invention is a system that analyzes voice input and emotional state during a meeting in real time and provides immediate feedback. The user first acquires voice input through a terminal. The terminal is equipped with a high-performance microphone and a facial recognition camera, which simultaneously records the user's speech and emotional state. The voice input is converted into text data using speech recognition software. In this process, a general-purpose speech processing API is used for the speech recognition engine.

[0567] The device uses emotion recognition software to recognize emotional states. The voice and emotion data analyzed by the terminal are transmitted to a server via the network. The server analyzes the acquired text data using natural language processing technology and applies a generative AI model as needed. This extracts integrated insights from the analysis of the spoken content and the emotion data.

[0568] The server generates feedback for the user based on the analysis results. This feedback is displayed on the terminal and can be received by the user in real time. For example, in a development meeting of an international team, it is possible to instantly grasp the participants' feelings about the progress of the project and take necessary measures during the meeting.

[0569] The generated data is integrated into meeting minutes, which are automatically generated on the server at the end of the meeting. These minutes include key points based on what was said and the corresponding emotional states, and are stored in cloud storage. This allows users to access them later and use them to help with future decision-making.

[0570] An example of a prompt would be, "Analyze the user's comments and emotional state during this meeting, and incorporate this into the feedback." Through this prompt, users can easily obtain guidance for improving the efficiency of the meeting.

[0571] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0572] Step 1:

[0573] The terminal captures audio during the meeting using a high-performance microphone. At the same time, a facial recognition camera is also used to record the participants' facial expressions, and the emotional state of the speaker is inferred through voice intonation analysis software. The input here is raw audio data and camera footage, and the output is audio data and emotional state data. Specifically, the system analyzes the audio waveform and detects changes in voice tone and pitch, allowing the emotion engine to tag emotions such as "joy" and "sadness."

[0574] Step 2:

[0575] The terminal transmits the acquired voice data and emotional state data to the server via a secure network. The input is the voice data and emotional state data obtained in the previous step, and the output is a data packet whose integrity has been verified. Specifically, the data is encrypted and a checksum is included during transmission to ensure data integrity.

[0576] Step 3:

[0577] The server uses a speech recognition engine to convert audio data into text data. The input is audio data, and the output is text data. Specifically, it performs dictionary lookups and phoneme analysis to convert speech to text, and performs correction processing to prevent misrecognition, especially for specialized terminology.

[0578] Step 4:

[0579] The server uses natural language processing techniques to comprehensively analyze text and sentiment data and extract insights. The input is text data and associated sentiment data, while the output is analyzed insights and suggestible feedback. Specifically, it performs keyword extraction from utterances, contextual understanding, and trend analysis of sentiment changes.

[0580] Step 5:

[0581] The server uses an AI model to generate feedback based on the analysis results and presents it to the user via the terminal. The input is the analyzed insights, and the output is a feedback message. Specifically, prompts are sent to the AI ​​model, which generates feedback in natural language. For example, it might display, "This statement received a lot of support."

[0582] Step 6:

[0583] The server automatically generates meeting minutes and extracts key points using the data accumulated after the meeting. Input consists of the entire text data and sentiment data, while output is a summarized meeting minute. Specific operations include ranking important statements and creating graphs showing the intensity of emotions.

[0584] Step 7:

[0585] The server saves the generated meeting minutes to secure cloud storage, making them accessible later. The input is the automatically generated meeting minutes, and the output is the saved data. Specific actions include adding classification tags before saving and setting specific access permissions within the storage.

[0586] (Application Example 2)

[0587] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0588] In modern meetings and discussions, it is difficult to accurately grasp not only the content of participants' statements but also the emotions and intentions behind them. This can lead to misunderstandings, a lack of communication, and decreased communication efficiency. In particular, in workplaces such as factories, where rapid decision-making and appropriate feedback are crucial to work efficiency, real-time information provision, including emotional recognition, is required.

[0589] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0590] In this invention, the server includes means for acquiring voice input, means for converting the acquired voice into text data, means for analyzing the converted text data in real time and extracting multifaceted insights considering cultural, professional, and generational backgrounds, means for presenting the extracted insights to the user, means for automatically generating meeting minutes that summarize the important points of the meeting, means for acquiring audio and video and recognizing the emotional state of participants in real time, and means for providing feedback to the user based on emotional data. As a result, information integrating the content of speech and emotions is provided, enabling efficient and accurate communication.

[0591] "Means for acquiring voice input" refers to a general term for devices and software used to collect voice data from people's conversations in meetings, work environments, and other settings.

[0592] "Means of converting to text data" refers to speech recognition technology and related software used to analyze acquired audio data and convert it into text information.

[0593] "A means of analyzing and extracting insights in real time" refers to a system that instantly analyzes converted text data and extracts the information necessary to gain a multifaceted understanding.

[0594] "Means of presenting insights to users" refers to interfaces and display devices that clearly show and communicate information derived from analysis results to users.

[0595] "Methods for automatically generating meeting minutes" refers to technologies or algorithms that automatically create documents summarizing the content of discussions at a meeting and highlighting the key points.

[0596] A "means for acquiring audio and video and recognizing emotional states" is a system that uses audio and visual data to identify emotions by analyzing the tone of voice and facial expressions of the speaker.

[0597] "Means of providing feedback based on emotional data" refers to applications or devices that utilize recognized emotional information to provide appropriate advice or responses tailored to the user's situation.

[0598] This invention is a system that analyzes voice input in real time, comprehensively understands its content and emotions, and provides appropriate feedback to the user. Specifically, it is implemented using the following hardware and software.

[0599] The server first acquires audio and video data from devices connected to microphones and cameras. This acquired data is converted into text data by speech recognition engines such as Google Cloud Speech-to-Text or Amazon Transcribe. Next, an emotion recognition library such as Affectiva SDK is used to analyze the user's facial expressions from the acquired video and extract emotion data.

[0600] The server uses this speech-text data and sentiment data to apply natural language processing technologies such as Google Cloud Natural Language to analyze the context of the speech. This process yields multifaceted insights that take into account cultural, professional, and generational backgrounds. Furthermore, by fusing sentiment data, the insights are delivered in a way that is relevant to the user.

[0601] The terminal provides feedback to the user based on analysis results sent from the server. This feedback is delivered in real time via the display and speakers, facilitating user understanding. For example, during a factory meeting, analyzing the statements and emotions of workers at the time can enable revisions to instructions and more persuasive suggestions.

[0602] As a concrete example, in a factory production meeting, this system could be used to immediately propose improvement measures if stress or anxiety is detected along with the content of each worker's comments. An example of a prompt message in this case would be: "Based on Mr. / Ms. A's facial expression and comments, it is clear that they are experiencing significant anxiety regarding process B. We will support your suggestion for process improvement."

[0603] This enables a system where the server provides insights that integrate the content and emotions of what is said, helping users make efficient and accurate decisions.

[0604] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0605] Step 1:

[0606] The terminal activates its microphone and camera at the start of the meeting, and continuously acquires the participants' audio and video. The input here is real-time audio and video data, which forms the basis for analysis in the next processing step. The output is raw audio and video data sent to the server.

[0607] Step 2:

[0608] The server receives audio data and converts it into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text). The input for this step is audio data, which is converted from an acoustic signal to text. The output is text data suitable for analysis.

[0609] Step 3:

[0610] The server analyzes video data using emotion recognition libraries such as the Affectiva SDK, extracting emotional data from the participants' facial expressions. The input is video data, where emotions are identified from facial movements and changes in facial muscles. The output is data indicating the user's emotional state.

[0611] Step 4:

[0612] The server uses natural language processing technology (e.g., Google Cloud Natural Language) to analyze text data and extract insights from the content and context of the statements. The input is the text data generated in step 2, and a language model is used to analyze the meaning and intent of the content. The output is structured information as insights.

[0613] Step 5:

[0614] The input for the server to fuse insights and sentiment data and generate feedback for the user is the sentiment data and insights obtained in steps 3 and 4. This feedback is tailored to help the user understand and is then output. Specifically, when in operation, a prompt generation AI model creates prompt sentences that are displayed or spoken in real time at the appropriate time.

[0615] Step 6:

[0616] The terminal receives feedback from the server and presents the content to the user using a display or speaker. The input here is the feedback information, which is the output of step 5. Specifically, during a factory meeting, a message such as "We support your process improvement suggestions" is conveyed visually or audibly. The output is an interaction provided in a way that facilitates user understanding.

[0617] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0618] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0619] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0620] [Fourth Embodiment]

[0621] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0622] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0623] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0624] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0625] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0626] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0627] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0628] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0629] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0630] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0631] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0632] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0633] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0634] This invention is an embodiment of a system that analyzes the content of meeting discussions in real time and provides multifaceted insights by considering cultural, professional, and generational backgrounds. To achieve this, the following program structure is employed.

[0635] 1. Acquisition of voice input

[0636] The device uses a high-quality microphone to capture participants' speech during the meeting as audio data. This allows for clear and low-noise audio.

[0637] 2. Speech-to-text conversion

[0638] Once audio is acquired, the server receives the audio data and uses a speech recognition engine to convert it into text data. The converted text is then used in the next parsing step.

[0639] 3. A multifaceted analysis of the statements made.

[0640] The server uses natural language processing techniques to analyze the background information contained in the text data. This includes participants' cultural backgrounds, expertise, and generational differences. This analysis generates multifaceted insights.

[0641] 4. Providing real-time feedback

[0642] Based on the analysis results, the server extracts different perspectives and key insights in real time during the meeting and presents them to the user via the terminal. This allows for faster and more accurate decision-making during the meeting.

[0643] 5. Automatic generation of meeting minutes

[0644] After the meeting ends, the server automatically generates meeting minutes summarizing all the statements and their analysis results. These minutes can be referenced later, significantly improving meeting efficiency.

[0645] As a concrete example, consider a progress meeting for an international project. Each participant has a different cultural background and professional perspective. In this setting, the terminal captures the spoken audio, and the server analyzes the background information. This deepens the understanding of the participants' differing opinions from their respective fields of expertise, enabling appropriate project coordination. Finally, the server summarizes the key points of the meeting and generates concise and easy-to-understand minutes, providing convenience for later review.

[0646] The following describes the processing flow.

[0647] Step 1:

[0648] The device captures the speech of meeting participants in real time via the microphone and saves it as audio data.

[0649] Step 2:

[0650] The device compresses the acquired audio data and sends it to the server via the network.

[0651] Step 3:

[0652] The server processes the received audio data through a speech recognition engine to convert the audio into text data. Preprocessing, such as noise filtering, is also performed to improve the accuracy of the conversion results.

[0653] Step 4:

[0654] The server passes the converted text data to an analysis module, which uses natural language processing techniques to comprehensively analyze the content of the speech. This includes analysis of cultural, professional, and generational backgrounds.

[0655] Step 5:

[0656] The server organizes the analysis results, extracting key insights and different perspectives. This information is then compiled as real-time feedback.

[0657] Step 6:

[0658] The server extracts feedback information and sends it to the terminal, which then presents it to the user via screen display or audio.

[0659] Step 7:

[0660] After the meeting ends, the server automatically generates meeting minutes summarizing the key points based on all the analysis results. These minutes are saved to cloud storage, making them accessible to users later.

[0661] (Example 1)

[0662] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0663] In today's diverse meetings involving participants from various backgrounds, it is essential to immediately and accurately understand what is said and quickly gain multifaceted insights based on that understanding. However, when these processes are carried out manually, they are time-consuming and lead to delays in decision-making. Furthermore, insights that do not consider cultural, professional, and generational differences are insufficient. In addition, creating meeting minutes after the meeting is another time-consuming and laborious task.

[0664] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0665] In this invention, the server includes a device for acquiring voice input, a device for converting the acquired voice into text information, and a device for analyzing the converted text information in real time and extracting multifaceted insights considering cultural, professional, and generational backgrounds. This makes it possible to quickly and comprehensively analyze the content of speech during meetings and provide users with immediate insights. Furthermore, the automatic generation of meeting minutes summarizing the key points of the meeting allows for efficient review afterward.

[0666] "Voice input" is the process of acquiring audio signals generated during meetings or conversations.

[0667] A "device" is a piece of equipment or system designed to perform a specific function.

[0668] "Text information" refers to information obtained by converting audio data into written characters.

[0669] "Real-time analysis" refers to a process where data is processed and analyzed immediately as soon as it is generated.

[0670] "Extracting insights from multiple perspectives" is the process of considering different viewpoints and backgrounds to derive useful information and understanding.

[0671] "Cultural, professional, and generational background" refers to the set of characteristics and influences related to an individual's culture, professional knowledge, and age.

[0672] "Insight" refers to the insights and understanding gained from specific data or information.

[0673] A "user" is a person or group that uses a system or product.

[0674] "Key points" are elements or content within information or events that deserve particular attention or have high priority.

[0675] "Meeting minutes" are documents that record the content and decisions discussed and made at meetings or discussions.

[0676] A "communication protocol for rapid processing" refers to a set of rules and procedures for sending and receiving data smoothly and efficiently.

[0677] "Audio processing technology for noise reduction" refers to technology used to reduce undesirable noise and interference in audio signals.

[0678] This invention is a system that improves the effectiveness of meetings involving multiple participants. The system acquires audio input from meetings, converts it into text information, and then analyzes it in real time to provide users with multifaceted insights that take cultural, professional, and generational backgrounds into consideration. The following hardware and software are used to realize this configuration.

[0679] The terminal uses a high-quality microphone to capture speech during the meeting as audio. Noise cancellation technology minimizes background noise, sending clear audio data to the server. The server processes the received audio data through a speech recognition engine (e.g., a general speech recognition service) to convert the speech into text. Furthermore, the server uses natural language processing technology, specifically generative AI models such as BERT and GPT, to analyze this text information and generate multifaceted insights. This makes it possible to extract valuable insights in real time, taking into account the cultural and professional backgrounds of the participants.

[0680] As a concrete example, consider an international project meeting. This meeting includes multiple participants with different cultural backgrounds and professional perspectives. By having a terminal capture each participant's speech as audio and a server quickly process that audio, users can understand diverse opinions in real time and make appropriate decisions regarding the project's progress. This is expected to lead to smoother progress and improved results.

[0681] This system allows users to input prompts using a generative AI model to enable more accurate and effective meeting decision-making. For example, a prompt such as "Automatically generate minutes for this meeting, highlighting insights based on cultural context and expert perspectives" can be used. This ensures that meeting content is efficiently recorded and analyzed, and the information is provided in a format suitable for future reference.

[0682] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0683] Step 1:

[0684] The device uses a high-quality microphone to capture audio during the meeting. It utilizes noise cancellation to reduce background noise and collect clear, high-precision audio data. The input here is the participants' speech, and the output is digital audio data.

[0685] Step 2:

[0686] The terminal sends the collected audio data to the server. Due to the high-speed communication protocol used, the audio data reaches the server with virtually no delay. The input is the audio data from the terminal, and the output is the audio data transferred to the server.

[0687] Step 3:

[0688] The server processes the received audio data through a speech recognition engine and converts it into text data. Specifically, the speech recognition algorithm analyzes the sounds of words and generates the corresponding text. The input is audio data, and the output is text information.

[0689] Step 4:

[0690] The server analyzes the converted text data using a generation AI model. Specifically, it uses natural language processing techniques to extract the meaning of the text and generate insights that take into account cultural, professional, and generational contexts. The input is text data, and the output is analyzed insightful information.

[0691] Step 5:

[0692] Based on the analysis results, the server generates multifaceted perspectives and key insights for the user, which are then provided via the terminal. Users utilize this feedback to deepen discussions during meetings. The input is the analyzed insights, and the output is the feedback information presented to the user.

[0693] Step 6:

[0694] After the meeting ends, the server automatically generates meeting minutes summarizing the key points based on all the statements and analysis results. Specifically, it utilizes natural language generation technology to create a readable and well-organized document. The input is all the data generated during the meeting, and the output is the summarized meeting minutes.

[0695] (Application Example 1)

[0696] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0697] In autonomous vehicles, there is a need to facilitate communication among passengers, provide services based on their diverse backgrounds and interests, and improve the ride experience. However, currently, there is a lack of means to provide appropriate insights in real time based on what passengers say and to optimize the driving route and in-vehicle environment. A new system is needed to solve this problem.

[0698] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0699] In this invention, the server includes means for acquiring voice information, means for converting it into symbolic data, and means for extracting insights considering cultural background and expertise. This makes it possible to analyze passenger conversations in real time and optimize routes and services based on their interests.

[0700] "Voice information" refers to the recording of voices spoken by users, converted into data.

[0701] "Symbol sequence data" refers to text data obtained by converting audio information into characters and symbols.

[0702] "Insight" refers to a deep understanding and knowledge gained from analyzed data, and is information that is useful for decision-making.

[0703] A "user" is an individual or organization that uses the system to obtain information.

[0704] "Automatic generation" means that a system or program automatically creates data or documents without human intervention.

[0705] A "vehicle" is a vehicle used to transport people or goods on land.

[0706] A "proposal" is the act of presenting specific options or action plans.

[0707] This system aims to improve the ride experience by facilitating communication between passengers in autonomous vehicles. The server acquires voice information using a high-quality microphone. The acquired voice information is converted into symbolic data by the vehicle's computer. The Google Speech-to-Text API is used for this conversion. The server uses a natural language processing model running on Azure or AWS to analyze the converted symbolic data and extract insights that take into account the passengers' cultural backgrounds and expertise.

[0708] For example, if a passenger says they want to visit a nearby tourist attraction, the server uses the Google Maps API to search for the nearest attractions and suggests a suitable route in real time. This allows passengers to receive on-demand services tailored to their individual needs.

[0709] An example of a prompt message could be, "Analyze your conversation while driving and suggest recommendations based on places you want to visit and activities you're interested in." This system enables real-time feedback and highly adaptive service delivery, improving passenger satisfaction.

[0710] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0711] Step 1:

[0712] The terminal uses a high-quality microphone to acquire audio information from inside the vehicle. This results in clear audio data with minimal noise. This audio data is then sent to the server as input.

[0713] Step 2:

[0714] The server converts the acquired audio information into symbolic text data using the Google Speech-to-Text API. In this step, the audio data is automatically converted to text and used for the analysis process within the server. The converted symbolic text data is then output.

[0715] Step 3:

[0716] The server uses a natural language processing model trained with TensorFlow to analyze the symbolic data. During this analysis, a generative AI model is applied to identify passengers' intentions and interests, extracting multifaceted insights. The analysis results in the output of insightful data.

[0717] Step 4:

[0718] The server generates appropriate suggestions for the user based on the extracted insight data. Specifically, it uses the Google Maps API to collect destination information and calculate the optimal route. This generates the suggestions presented to the user.

[0719] Step 5:

[0720] The user receives suggestions sent from the server via an in-vehicle display or voice prompt. At this stage, possible route information and activity suggestions are output and presented to the user as options.

[0721] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0722] This invention combines an emotion engine with a system that analyzes audio data acquired during meetings, thereby recognizing the user's emotions along with the content of their speech and providing real-time feedback. The following program structure is used to implement this invention.

[0723] 1. Voice input acquisition and emotion recognition

[0724] The device acquires participants' statements during the meeting and uses an emotion engine to recognize the user's emotional state through facial recognition cameras and voice intonation analysis. This emotion data is then associated with each statement.

[0725] 2. Transmission of voice and emotion data

[0726] The device transmits the acquired voice and emotion data to the server in real time. The server then prepares to process this data in an integrated manner.

[0727] 3. Analysis and conversion of audio data

[0728] The server converts the audio data into text data using a speech recognition engine. Furthermore, it utilizes natural language processing technology to analyze the content and background information of the speech, and also incorporates emotion recognition data into the analysis.

[0729] 4. Providing insights and emotion-oriented feedback

[0730] The server extracts insights from multiple perspectives based on analysis results and emotional data. The device then presents the user with real-time feedback, adjusted for user acceptance and empathy, taking into account the emotional recognition results.

[0731] 5. Automated generation of meeting minutes and integration of sentiment data

[0732] After the meeting ends, the server summarizes the key points based on the content of the speeches and sentiment data, and automatically generates meeting minutes. These minutes also incorporate sentiment data, allowing for an understanding of the emotional context of the speeches. The generated data is saved to cloud storage in a format that can be accessed later.

[0733] As a concrete example, considering a development meeting involving an international team, it becomes easier to understand the level of employee motivation and concerns regarding the project's progress during the meeting. This allows the project manager to take necessary measures during the meeting and effectively promote communication among stakeholders. Furthermore, the generation of meeting minutes, which include changes in emotions, can be used for future decision-making.

[0734] The following describes the processing flow.

[0735] Step 1:

[0736] The device uses microphones and cameras installed in the meeting room to capture participants' speech and facial expressions in real time. Both audio and video data are temporarily stored in local storage.

[0737] Step 2:

[0738] The device compresses the stored audio data, performs noise filtering, and then sends it to the server via the network. Simultaneously, it sends facial expression information obtained from the video data to the server as a feature vector.

[0739] Step 3:

[0740] The server analyzes the received audio data using a speech recognition engine and converts it into text data. During this process, an acoustic model is applied to improve the accuracy of speech recognition.

[0741] Step 4:

[0742] The server analyzes text data using natural language processing technology to comprehensively analyze the meaning of the utterances. This analysis includes extracting field-specific keywords and understanding the context.

[0743] Step 5:

[0744] The server simultaneously uses an emotion engine based on the received facial expression data to recognize the emotional state of the participants. Emotional states are classified into categories such as "joy," "surprise," "sadness," and "anger."

[0745] Step 6:

[0746] The server integrates emotion recognition data with the spoken content and generates context-appropriate, emotion-oriented feedback. The generated feedback is adjusted to a tone that is considerate of the recipient of the statement.

[0747] Step 7:

[0748] The device displays this feedback to the user in real time. This clarifies the intent and nuances during the meeting and improves communication among participants.

[0749] Step 8:

[0750] After the meeting ends, the server automatically generates meeting minutes, summarizing and organizing all spoken and emotional data to highlight key points. These minutes also record emotional changes during speaking, which can be used for later analysis and decision-making.

[0751] Step 9:

[0752] The generated meeting minutes and related data are saved to cloud storage by the server, allowing users to access them securely later.

[0753] (Example 2)

[0754] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0755] In meetings, it is essential to understand not only the content of participants' statements but also their emotional state at the time of speaking, and to provide appropriate feedback immediately. However, existing systems are limited to speech-to-text conversion and simple content analysis, and do not adequately grasp emotional states or provide insights based on them. As a result, the quality of meetings does not improve, and it is a factor that diminishes participants' motivation.

[0756] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0757] In this invention, the server includes means for acquiring voice input and recognizing emotional states, means for converting the acquired voice into text data, and means for analyzing the converted text data and emotional data in real time and extracting multifaceted insights while considering background information. This enables the provision of feedback that reflects the emotional states of participants during a meeting and the automatic generation of meeting minutes that include emotional information.

[0758] "Voice input" refers to the digital acquisition of participants' statements during a meeting.

[0759] "Emotional state" refers to information that describes the type and intensity of emotions a participant expresses when speaking.

[0760] "Text data" refers to character information converted from voice input, and is the subject of analysis.

[0761] "Real-time" refers to the temporal characteristic of a system that processes and analyzes information instantly and presents results without delay.

[0762] "Insight" refers to information that facilitates the understanding and insights gained from analyzed data.

[0763] "Feedback" refers to information and suggestions provided to the user based on the analysis results.

[0764] "Meeting minutes" are documents that record the content of discussions and important points during a meeting.

[0765] A "digital storage medium" is a means of electronically storing and managing data.

[0766] This invention is a system that analyzes voice input and emotional state during a meeting in real time and provides immediate feedback. The user first acquires voice input through a terminal. The terminal is equipped with a high-performance microphone and a facial recognition camera, which simultaneously records the user's speech and emotional state. The voice input is converted into text data using speech recognition software. In this process, a general-purpose speech processing API is used for the speech recognition engine.

[0767] The device uses emotion recognition software to recognize emotional states. The voice and emotion data analyzed by the terminal are transmitted to a server via the network. The server analyzes the acquired text data using natural language processing technology and applies a generative AI model as needed. This extracts integrated insights from the analysis of the spoken content and the emotion data.

[0768] The server generates feedback for the user based on the analysis results. This feedback is displayed on the terminal and can be received by the user in real time. For example, in a development meeting of an international team, it is possible to instantly grasp the participants' feelings about the progress of the project and take necessary measures during the meeting.

[0769] The generated data is integrated into meeting minutes, which are automatically generated on the server at the end of the meeting. These minutes include key points based on what was said and the corresponding emotional states, and are stored in cloud storage. This allows users to access them later and use them to help with future decision-making.

[0770] An example of a prompt message would be, "Analyze the user's comments and emotional state during this meeting, and incorporate this into the feedback." Through this prompt, users can easily obtain guidance for improving the efficiency of the meeting.

[0771] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0772] Step 1:

[0773] The terminal captures audio during the meeting using a high-performance microphone. At the same time, a facial recognition camera is also used to record the participants' facial expressions, and the emotional state of the speaker is inferred through voice intonation analysis software. The input here is raw audio data and camera footage, and the output is audio data and emotional state data. Specifically, the system analyzes the audio waveform and detects changes in voice tone and pitch, allowing the emotion engine to tag emotions such as "joy" and "sadness."

[0774] Step 2:

[0775] The terminal transmits the acquired voice data and emotional state data to the server via a secure network. The input is the voice data and emotional state data obtained in the previous step, and the output is a data packet whose integrity has been verified. Specifically, the data is encrypted and a checksum is included during transmission to ensure data integrity.

[0776] Step 3:

[0777] The server uses a speech recognition engine to convert audio data into text data. The input is audio data, and the output is text data. Specifically, it performs dictionary lookups and phoneme analysis to convert speech to text, and performs correction processing to prevent misrecognition, especially for specialized terminology.

[0778] Step 4:

[0779] The server uses natural language processing techniques to comprehensively analyze text and sentiment data and extract insights. The input is text data and associated sentiment data, while the output is analyzed insights and suggestible feedback. Specifically, it performs keyword extraction from utterances, contextual understanding, and trend analysis of sentiment changes.

[0780] Step 5:

[0781] The server uses an AI model to generate feedback based on the analysis results and presents it to the user via the terminal. The input is the analyzed insights, and the output is a feedback message. Specifically, prompts are sent to the AI ​​model, which generates feedback in natural language. For example, it might display, "This statement received a lot of support."

[0782] Step 6:

[0783] The server automatically generates meeting minutes and extracts key points using the data accumulated after the meeting. Input consists of the entire text data and sentiment data, while output is a summarized meeting minute. Specific operations include ranking important statements and creating graphs showing the intensity of emotions.

[0784] Step 7:

[0785] The server saves the generated meeting minutes to secure cloud storage, making them accessible later. The input is the automatically generated meeting minutes, and the output is the saved data. Specific actions include adding classification tags before saving and setting specific access permissions within the storage.

[0786] (Application Example 2)

[0787] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0788] In modern meetings and discussions, it is difficult to accurately grasp not only the content of participants' statements but also the emotions and intentions behind them. This can lead to misunderstandings, a lack of communication, and decreased communication efficiency. In particular, in workplaces such as factories, where rapid decision-making and appropriate feedback are crucial to work efficiency, real-time information provision, including emotional recognition, is required.

[0789] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0790] In this invention, the server includes means for acquiring voice input, means for converting the acquired voice into text data, means for analyzing the converted text data in real time and extracting multifaceted insights considering cultural, professional, and generational backgrounds, means for presenting the extracted insights to the user, means for automatically generating meeting minutes that summarize the important points of the meeting, means for acquiring audio and video and recognizing the emotional state of participants in real time, and means for providing feedback to the user based on emotional data. As a result, information integrating the content of speech and emotions is provided, enabling efficient and accurate communication.

[0791] "Means for acquiring voice input" refers to a general term for devices and software used to collect voice data from people's conversations in meetings, work environments, and other settings.

[0792] "Means of converting to text data" refers to speech recognition technology and related software used to analyze acquired audio data and convert it into text information.

[0793] "A means of analyzing and extracting insights in real time" refers to a system that instantly analyzes converted text data and extracts the information necessary to gain a multifaceted understanding.

[0794] "Means of presenting insights to users" refers to interfaces and display devices that clearly show and communicate information derived from analysis results to users.

[0795] "Methods for automatically generating meeting minutes" refers to technologies or algorithms that automatically create documents summarizing the content of discussions at a meeting and highlighting the key points.

[0796] A "means for acquiring audio and video and recognizing emotional states" is a system that uses audio and visual data to identify emotions by analyzing the tone of voice and facial expressions of the speaker.

[0797] "Means of providing feedback based on emotional data" refers to applications or devices that utilize recognized emotional information to provide appropriate advice or responses tailored to the user's situation.

[0798] This invention is a system that analyzes voice input in real time, comprehensively understands its content and emotions, and provides appropriate feedback to the user. Specifically, it is implemented using the following hardware and software.

[0799] The server first acquires audio and video data from devices connected to microphones and cameras. This acquired data is converted into text data by speech recognition engines such as Google Cloud Speech-to-Text or Amazon Transcribe. Next, an emotion recognition library such as Affectiva SDK is used to analyze the user's facial expressions from the acquired video and extract emotion data.

[0800] The server uses this speech-text data and sentiment data to apply natural language processing technologies such as Google Cloud Natural Language to analyze the context of the speech. This process yields multifaceted insights that take into account cultural, professional, and generational backgrounds. Furthermore, by fusing sentiment data, the insights are delivered in a way that is relevant to the user.

[0801] The terminal provides feedback to the user based on analysis results sent from the server. This feedback is delivered in real time via the display and speakers, facilitating user understanding. For example, during a factory meeting, analyzing the statements and emotions of workers at the time can enable revisions to instructions and more persuasive suggestions.

[0802] As a concrete example, in a factory production meeting, this system could be used to immediately propose improvement measures if stress or anxiety is detected along with the content of each worker's comments. An example of a prompt message in this case would be: "Based on Mr. / Ms. A's facial expression and comments, it is clear that they are experiencing significant anxiety regarding process B. We will support your suggestion for process improvement."

[0803] This enables a system where the server provides insights that integrate the content and emotions of what is said, helping users make efficient and accurate decisions.

[0804] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0805] Step 1:

[0806] The terminal activates its microphone and camera at the start of the meeting, and continuously acquires the participants' audio and video. The input here is real-time audio and video data, which forms the basis for analysis in the next processing step. The output is raw audio and video data sent to the server.

[0807] Step 2:

[0808] The server receives audio data and converts it into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text). The input for this step is audio data, which is converted from an acoustic signal to text. The output is text data suitable for analysis.

[0809] Step 3:

[0810] The server analyzes video data using emotion recognition libraries such as the Affectiva SDK, extracting emotional data from the participants' facial expressions. The input is video data, where emotions are identified from facial movements and changes in facial muscles. The output is data indicating the user's emotional state.

[0811] Step 4:

[0812] The server uses natural language processing technology (e.g., Google Cloud Natural Language) to analyze text data and extract insights from the content and context of the statements. The input is the text data generated in step 2, and a language model is used to analyze the meaning and intent of the content. The output is structured information as insights.

[0813] Step 5:

[0814] The input for the server to fuse insights and sentiment data and generate feedback for the user is the sentiment data and insights obtained in steps 3 and 4. This feedback is tailored to help the user understand and is then output. Specifically, when in operation, a prompt generation AI model creates prompt sentences that are displayed or spoken in real time at the appropriate time.

[0815] Step 6:

[0816] The terminal receives feedback from the server and presents the content to the user using a display or speaker. The input here is the feedback information, which is the output of step 5. Specifically, during a factory meeting, a message such as "We support your process improvement suggestions" is conveyed visually or audibly. The output is an interaction provided in a way that facilitates user understanding.

[0817] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0818] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0819] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0820] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0821] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0822] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0823] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0824] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0825] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0826] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0827] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0828] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0829] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0830] 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.

[0831] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0832] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0833] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0834] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0835] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0836] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0837] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[0838] The following is further disclosed regarding the embodiments described above.

[0839] (Claim 1)

[0840] A means of acquiring voice input,

[0841] A means of converting acquired audio into text data,

[0842] A means of analyzing converted text data in real time and extracting multifaceted insights by considering cultural, professional, and generational backgrounds,

[0843] A means of presenting extracted insights to the user,

[0844] A method for automatically generating meeting minutes that summarize the key points of a meeting,

[0845] A system that includes this.

[0846] (Claim 2)

[0847] The system according to claim 1, which uses natural language processing and machine learning models when analyzing speech input.

[0848] (Claim 3)

[0849] The system according to claim 1, which saves the generated meeting minutes on cloud storage and makes them accessible later.

[0850] "Example 1"

[0851] (Claim 1)

[0852] A device for acquiring voice input,

[0853] A device that converts acquired audio into text information,

[0854] A device that analyzes converted text information in real time and extracts multifaceted insights considering cultural, professional, and generational backgrounds,

[0855] A device that presents extracted insights to the user,

[0856] A device that automatically generates meeting minutes summarizing the key points of a meeting,

[0857] A device that implements a communication protocol for rapidly processing voice data,

[0858] A device that utilizes audio processing technology to reduce noise,

[0859] A device that provides information to support user decision-making based on generated insights,

[0860] A system that includes this.

[0861] (Claim 2)

[0862] The system according to claim 1, comprising a device for analyzing text information using natural language processing and machine learning models.

[0863] (Claim 3)

[0864] The system according to claim 1, comprising a device for saving the generated meeting minutes on a data storage area and making them accessible later.

[0865] "Application Example 1"

[0866] (Claim 1)

[0867] Means for acquiring audio information,

[0868] A means for converting acquired audio information into symbol sequence data,

[0869] A means of analyzing converted symbol sequence data in real time and extracting multifaceted insights by considering cultural background, expertise, and generational differences,

[0870] A means of presenting extracted insights to users,

[0871] A means of automatically generating a transcript that summarizes the key points of a conversation,

[0872] A means of analyzing conversations obtained inside a vehicle and making suggestions to optimize routes and environments based on passengers' interests,

[0873] A system that includes this.

[0874] (Claim 2)

[0875] The system according to claim 1, which uses natural language processing and machine learning models to analyze voice information, understand the passenger's intent, and generate optimal suggestions.

[0876] (Claim 3)

[0877] The system according to claim 1, which stores the generated record documents and proposed content on a remote storage system and makes them accessible at a later date.

[0878] "Example 2 of combining an emotion engine"

[0879] (Claim 1)

[0880] A means of acquiring voice input and recognizing emotional states,

[0881] A means of converting acquired audio into text data,

[0882] A method for analyzing converted text data and sentiment data in real time, and for extracting multifaceted insights while considering background information,

[0883] A means of presenting extracted insights and emotional feedback to the user,

[0884] A method for automatically generating meeting minutes that summarize key points and sentiment data from a meeting,

[0885] A means of saving the generated data to a digital storage medium in a form that can be accessed later,

[0886] A system that includes this.

[0887] (Claim 2)

[0888] The system according to claim 1, which uses natural language processing technology and a learning algorithm when analyzing voice input.

[0889] (Claim 3)

[0890] The system according to claim 1, which saves the generated meeting minutes on a digital storage medium and makes them accessible later.

[0891] "Application example 2 when combining with an emotional engine"

[0892] (Claim 1)

[0893] A means of acquiring voice input,

[0894] A means of converting acquired audio into text data,

[0895] A means of analyzing converted text data in real time and extracting multifaceted insights by considering cultural, professional, and generational backgrounds,

[0896] A means of presenting extracted insights to the user,

[0897] A method for automatically generating meeting minutes that summarize the key points of a meeting,

[0898] A means of acquiring audio and video and recognizing the emotional state of participants in real time,

[0899] A means of providing users with feedback based on emotional data,

[0900] A system that includes this.

[0901] (Claim 2)

[0902] The system according to claim 1, which uses natural language processing and machine learning models when analyzing speech input.

[0903] (Claim 3)

[0904] The system according to claim 1, which saves the generated meeting minutes on cloud storage and makes them accessible later. [Explanation of symbols]

[0905] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of acquiring voice input, A means of converting acquired audio into text data, A means of analyzing converted text data in real time and extracting multifaceted insights by considering cultural, professional, and generational backgrounds, A means of presenting extracted insights to the user, A method for automatically generating meeting minutes that summarize the key points of a meeting, A system that includes this.

2. The system according to claim 1, which uses natural language processing and machine learning models when analyzing voice input.

3. The system according to claim 1, which saves the generated meeting minutes on cloud storage and makes them accessible later.

Images