System

Abstract

A system is provided.SOLUTION: A system comprising: means for acquiring voice data; means for converting the acquired voice data into text data; means for generating minutes by analyzing the converted text data; means for generating a to-do list by analyzing the converted text data; means for generating a schedule by analyzing the converted text data; means for generating a draft of an e-mail by analyzing the converted text data; and means for transmitting the generated data to a user terminal.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Today's salespeople are busy, especially after customer visits, and there is a wide range of tasks to be done, which necessitates a significant reduction in time and effort. Typically, after a customer visit, numerous tasks occur, such as taking meeting minutes, creating a to-do list, scheduling the next meeting, and sending a thank-you email. However, these tasks are time-consuming and take away time that could otherwise be spent focusing on the next sales activity. The objective of this invention is to automate these post-customer visit processes, thereby reducing the burden on salespeople and providing a more efficient sales process. [Means for solving the problem]

[0005] The present invention is a system including a means for acquiring voice data, a means for converting the acquired voice data into text data, a means for analyzing the converted text data to generate minutes of a meeting, a means for analyzing the converted text data to generate a to-do list, a means for analyzing the converted text data to generate a schedule, a means for analyzing the converted text data to generate a draft of an email, and a means for transmitting the generated data to a user terminal. Specifically, the system converts voice recorded during a customer visit into high-quality Japanese text, and automatically generates necessary documents and data based on the text data using a generation AI. This significantly reduces the workload after a customer visit and improves the efficiency of sales activities.

[0006] "Audio data" refers to audio information recorded during a customer visit.

[0007] "Text data" is voice data that has been analyzed and converted into text information.

[0008] "Minutes" are documents that record the main points of business negotiations.

[0009] A "ToDo list" is a list of homework items and next tasks to be done in a business negotiation.

[0010] "Plan" is schedule information about the next meeting or important dates that are set based on the results of the business negotiations.

[0011] An "email draft" is a draft of an email to send as a thank you or follow-up message after a business meeting.

[0012] "User terminal" refers to electronic devices such as smartphones and tablets used by sales representatives.

[0013] A "server" is a computer system that receives voice data, converts it to text data, and processes the generative AI model.

[0014] The "means for transmitting" refers to a network communication function for transmitting the generated data to a user terminal.

[0015] "Means for converting" refers to the technology or method for converting audio data into high-quality text data.

[0016] "Means of analysis" refers to the method of using a generative AI model to analyze text data and generate minutes, to-do lists, schedules, email drafts, etc.

[0017] "Generative AI" is a technology that uses artificial intelligence models to generate desired output from text data. [Brief explanation of the drawings]

[0018] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10]1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0019] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0020] First, the terms used in the following description will be explained.

[0021] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0022] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0023] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0024] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0025] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0026] [First embodiment]

[0027] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0028] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0029] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0030] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0031] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0032] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0033] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0034] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0036] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0037] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0038] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0039] This invention is a system to reduce the workload of sales representatives after customer visits. Specifically, it converts voice data into high-quality text data, and then uses a generative AI model to automatically generate meeting minutes, to-do lists, schedules, and thank-you emails.

[0040] Explanation of program processing

[0041] Acquiring and uploading audio data

[0042] When visiting a customer, the user records the details of the business meeting using a device such as a smartphone or tablet. Once the recording is complete, the user uploads the recorded data from the device to the server. The server then receives the recorded data and prepares it for the next process.

[0043] Japanese text

[0044] The server converts the received recording data into high-quality Japanese text using a speech recognition engine. The converted text data is stored in the server and used in the next step.

[0045] Processing of generated AI

[0046] Once the text data is created, the server uses a generative AI model to generate the following outputs:

[0047] 1. Generate transcripts:

[0048] The server analyzes the text data, extracts the key points of the sales negotiation, and generates minutes in bullet-point format. The minutes summarize the important points of the sales negotiation, which are useful for sales representatives to review later.

[0049] 2. Create a to-do list:

[0050] Tasks and homework are extracted from the text data and the server creates a to-do list, which lists the tasks that need to be completed before the next business meeting.

[0051] 3. Create an appointment:

[0052] By parsing the date and time information in the text data, the server generates a Google Calendar of upcoming meetings and important events, including the date and time of your next client visit and important deadlines.

[0053] 4. Draft thank you email:

[0054] The server creates a draft of a thank-you email based on the details of the sales meeting, allowing the sales representative to send a thank-you email quickly after the meeting.

[0055] feedback

[0056] The generated minutes, to-do lists, schedules, and draft thank-you emails are sent from the server to the user's device. The user receives a notification and can check and edit each piece of data. This significantly reduces the time spent on post-processing after returning to the office, allowing them to focus on their next sales activity.

[0057] Specific examples

[0058] Customer Visit Scenario

[0059] A user visits a customer and conducts a business negotiation. During the negotiation, the contents of the negotiation are recorded on the device. After the negotiation is over, the user presses the "Stop" button to stop recording, and then presses the "Upload" button to send the recorded data to the server.

[0060] Specific operations of server-side processing

[0061] 1. The server receives the recording data and saves it in storage.

[0062] 2. The server uses a speech recognition engine to convert the recorded data into high-quality Japanese text, and stores the converted text in storage.

[0063] 3. The server uses the generative AI model to generate the following data:

[0064] Minutes: "Key points of the meeting were..."

[0065] To-do list: "Homework items for next time..."

[0066] Google Calendar events: "Your next meeting is scheduled for August 10th at 10am."

[0067] A sample thank you email: "Thank you for the business meeting the other day..."

[0068] 4. The generated data is collected and sent to the user's device. The user receives a notification, checks and edits the generated data, and uses it as needed.

[0069] As described above, the system according to the present invention has the effect of largely automating the post-processing after a customer visit, thereby reducing the burden on sales staff.

[0070] The processing flow will be explained below.

[0071] Step 1:

[0072] When a user visits a customer, they use a device (smartphone or tablet) to record the details of the business meeting. They launch the device's voice recording app and press the "Record" button to start recording. When the business meeting is over, they press the "Stop" button to stop recording.

[0073] Step 2:

[0074] The user uploads the recorded data to the server from the voice recording app on their device. After pressing the "Upload" button on the app and confirming the connection to the server, the recorded data is sent to the server.

[0075] Step 3:

[0076] The server receives the uploaded recording, stores it in storage, and adds it to the conversion queue.

[0077] Step 4:

[0078] The server calls the speech recognition engine to convert the recorded data into high-quality Japanese text. The speech recognition engine analyzes the recorded data and generates Japanese text data. The converted text data is saved in storage.

[0079] Step 5:

[0080] The server inputs the stored Japanese text data into the generative AI model for analysis, analyzes the target text data, and generates various output data.

[0081] Step 6:

[0082] The server analyzes the text data, extracts the main points of the business negotiations, and generates minutes in bullet-point format. The generated minutes data is then saved in storage.

[0083] Step 7:

[0084] The server extracts action items (homework items) from the text data and generates a ToDo list. The generated ToDo list data is saved in storage.

[0085] Step 8:

[0086] The server extracts date and time information from the text data and adds the event to Google Calendar. It also calls the Google Calendar API to register upcoming meetings and important events.

[0087] Step 9:

[0088] The server creates a draft of the thank-you email based on the details of the business negotiation. It analyzes the text data and generates a draft email incorporating appropriate greetings and words of thanks. The generated draft email is then saved in storage.

[0089] Step 10:

[0090] The server sends the generated minutes, to-do list, Google Calendar schedule, and a draft of the thank-you email to the user's device. It also sends a notification to the user's device and provides a download link for the generated data.

[0091] Step 11:

[0092] The user receives a notification and clicks the provided link to download the generated data to their device. The user can provide feedback as needed, check and edit the data, and use it in their next sales activity.

[0093] Through the above processing steps, this system can improve the efficiency of the work that sales representatives perform after visiting customers and reduce their workload.

[0094] Example 1

[0095] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0096] The post-processing work that sales representatives perform after customer visits requires time and effort, hindering efficient sales activities. Furthermore, manual text generation and information organization have limited accuracy, leading to the risk of important information being overlooked or typographical errors. There is a need for a system that can solve these problems, reduce the workload of sales representatives, and enable fast and accurate information processing.

[0097] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0098] In this invention, the server includes means for acquiring voice data, means for converting the acquired voice data into text data, means for analyzing the converted text data to generate minutes, means for analyzing the converted text data to generate a to-do list, means for analyzing the converted text data to generate a schedule, means for analyzing the converted text data to generate a draft of an email, means for transmitting the generated data to a user terminal, means for converting the recorded data into Japanese text using a voice recognition engine, and means for generating various information from the text data using a generative AI model. This allows a user to simply record the contents of a business meeting and upload the data to the server, which automatically generates drafts of minutes, to-do lists, schedules, and thank-you emails, enabling efficient and accurate post-event processing.

[0099] "Voice data" refers to data that has been saved in digital format containing the content of conversations and dialogues recorded during customer visits, business negotiations, etc.

[0100] "Means of capture" refers to the combination of devices and software used by a user to record audio data and upload it to a server.

[0101] "Text data" refers to data obtained by converting acquired voice data into text format.

[0102] "Means for converting" refers collectively to a speech recognition engine and related software used to convert voice data into text data.

[0103] "Means of analysis" is a general term for the algorithms and generative AI models used to extract and organize key information from the converted text data.

[0104] Minutes are documents that summarize the main points of business negotiations, meetings, etc., and are used to confirm the contents later.

[0105] A "ToDo list" is a list of tasks or homework that need to be completed by the next time.

[0106] "Schedule" refers to schedule information such as the next date and time of a business meeting or meeting, or important deadlines.

[0107] An "email draft" is a draft of a greeting or thank you email to be used after a business meeting.

[0108] "Generative Means" refers collectively to generative AI models and related algorithms used to create meeting minutes, to-do lists, schedules, and email drafts from text data.

[0109] "Transmission means" is a general term for communication means and protocols for transmitting generated data to a user terminal.

[0110] This invention is a system that uses advanced voice data to automate the post-processing process after sales representatives visit customers. Specifically, it converts voice data into Japanese text data, and then uses a generative AI model based on that text data to automatically generate meeting minutes, to-do lists, schedules, and thank-you emails. This system is primarily comprised of components that play the roles of a server, terminal, and user.

[0111] Hardware and Software Overview

[0112] Terminal

[0113] Users use devices such as smartphones and tablets, which have an application with a recording function installed, allowing them to record the contents of business negotiations.

[0114] server

[0115] The server is the main device that receives, stores, analyzes, and distributes the generated audio data. The following software is installed on the server:

[0116] Speech Recognition Engine: Speech recognition software such as the Google Cloud Speech-to-Text API.

[0117] Generative AI models: Advanced generative AI software such as GPT-3.

[0118] User

[0119] The user is a sales representative who records voice data during a sales meeting and uploads the data to the server after the recording is complete. Post-processing is performed based on the generated information.

[0120] Data processing and calculation flow

[0121] 1. Acquiring and uploading audio data

[0122] When a user visits a customer, the user uses the device to record the details of the business meeting. After completing the recording, the user uploads the audio data from the device to the server. The server receives the recorded data via an HTTP request and stores it in storage.

[0123] 2. Japanese text

[0124] The server passes the received recorded data to a speech recognition engine and converts it into high-quality Japanese text data, which is then saved back into storage.

[0125] 3. Text Data Analysis and Generation

[0126] The server inputs the stored text data into a generative AI model to generate the following information:

[0127] Minutes: Summarize the main points of the business meeting in bullet points.

[0128] To-Do List: Make a list of tasks and homework that need to be done by the next time.

[0129] Appointments: Generate appointments in Google Calendar format based on the dates and times mentioned in the deal.

[0130] Draft a thank you email: Create a draft of a thank you email based on the details of the business meeting.

[0131] 4. Data distribution and feedback

[0132] The server sends the generated minutes, to-do lists, schedules, and thank-you email drafts to the user's device. The user receives a notification, checks and edits the generated data, and uses it as needed.

[0133] Examples and prompts

[0134] Specific examples

[0135] A user visits a customer and conducts a business meeting. The contents of the business meeting are recorded on the device, and after the business meeting is completed, the recording is stopped and the recorded data is uploaded to the server.

[0136] The server receives the recording, converts it into Japanese text using a speech recognition engine, and then uses a generative AI model to generate drafts of meeting minutes, to-do lists, schedules, and thank-you emails, which are then sent to the user's device.

[0137] Prompt Sentence Examples

[0138] By inputting the following prompt sentence into the generative AI model, we can get the corresponding output:

[0139] Prompt for generating transcripts:

[0140] "Please extract the key points from the following sales pitch in bullet points format:

[0141] 1. Customer needs

[0142] 2. Proposal details

[0143] 3. Questions and Answers

[0144] 4. Next assignment

[0145] Prompt for generating a ToDo list:

[0146] From the text below, please list the tasks you need to complete by next time:

[0147] Summary of business negotiations

[0148] Confirm next step

[0149] Preparing for follow-up

[0150] Prompt for generating an appointment:

[0151] "Extract important dates and times from the text below and generate a Google Calendar event."

[0152] Prompt for generating a draft thank you email:

[0153] "Please draft a thank you email based on the following business details:

[0154] Thank you message

[0155] Review the main points of the deal

[0156] Proposal for next steps

[0157] As a result, this system significantly improves the work efficiency of sales staff and reduces the time and effort required for post-sales processing.

[0158] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0159] Step 1:

[0160] Acquiring and uploading audio data

[0161] The user uses a device such as a smartphone or tablet to record the contents of a business meeting. They open a recording application and press the record button to record the contents of the business meeting. The input data is the audio spoken during the business meeting. After the business meeting is over, the user presses the "Stop" button to end recording, and then presses the "Upload" button to send the recorded data to the server. The server receives the HTTP request and saves the recorded data in storage. The output data is an audio file saved in storage.

[0162] Step 2:

[0163] Japanese text

[0164] The server passes the saved recording data to a speech recognition engine. The input data is an audio file. The server uses a speech recognition engine such as the Google Cloud Speech-to-Text API to convert this audio data into high-quality Japanese text data. Specifically, the speech recognition engine analyzes the audio waveform and converts it into text information. The converted text data is then saved back into storage within the server. The output data is Japanese text data.

[0165] Step 3:

[0166] Text data analysis and generation

[0167] The server inputs text data into the generative AI model to generate specific information. The input data is Japanese text data. The server uses the following prompt to have the generative AI model (e.g., GPT-3) analyze the data:

[0168] Prompt for generating transcripts:

[0169] "Please extract the key points from the following sales pitch in bullet points format:

[0170] 1. Customer needs

[0171] 2. Proposal details

[0172] 3. Questions and Answers

[0173] 4. Next assignment

[0174] Prompt for generating a ToDo list:

[0175] From the text below, please list the tasks you need to complete by next time:

[0176] Summary of business negotiations

[0177] Confirm next step

[0178] Preparing for follow-up

[0179] Prompt for generating an appointment:

[0180] "Extract important dates and times from the text below and generate a Google Calendar event."

[0181] Prompt for generating a draft thank you email:

[0182] "Please draft a thank you email based on the following business details:

[0183] Thank you message

[0184] Review the main points of the deal

[0185] Proposal for next steps

[0186] The server uses these prompts to have the generative AI model analyze the text data and generate information for each purpose. The generated information becomes the basis for meeting minutes, to-do lists, schedules, and thank-you emails, all of which are stored in the server's storage. The output data is the various types of generated information.

[0187] Step 4:

[0188] Data distribution and feedback

[0189] The server sends the generated data (minutes, to-do list, schedule, thank-you email draft) to the user's device. The input data is the various types of generated information. The server uses an HTTP response to send the generated data to the user's device. The user receives a notification on their device and checks and edits the generated data. Specifically, the user clicks the notification to open the application, checks the generated minutes, to-do list, schedule, and thank-you email, and edits them as necessary. The output data is the various types of information after it has been checked and edited by the user.

[0190] By following these steps, users can simply upload their voice data to the server, and minutes, to-do lists, schedules, and thank-you email drafts will be automatically generated and provided to them. This series of processes will greatly improve the efficiency of post-sales processing work for sales representatives.

[0191] (Application example 1)

[0192] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0193] During maintenance work in a factory, recording the details of the work and any problems that arise, and then creating work reports and to-do lists accurately and efficiently based on that information, places a heavy burden on workers. This creates the problem of time being spent creating reports and checking schedules, which reduces the efficiency of the work. Furthermore, mistakes are likely to occur in work reports and schedule settings, which can have an impact on factory operations. To solve these problems, an automated system is needed.

[0194] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0195] In this invention, the server includes means for acquiring voice data, means for converting the acquired voice data into text data, means for analyzing the converted text data to generate minutes, means for analyzing the converted text data to generate a to-do list, means for analyzing the converted text data to generate a schedule, means for analyzing the converted text data to generate a draft of an e-mail, means for sending the generated data to a user terminal, means for outputting the generated data as an in-factory work report, means for outputting the generated data as a work schedule in calendar format, and means for outputting the generated data as a work report e-mail. This automates in-factory maintenance work reporting and schedule setting, not only reducing the burden on workers but also improving work efficiency and reducing reporting and scheduling errors.

[0196] "Voice data" refers to audio information that records events, problems, and work conditions that occur during factory work.

[0197] "Text data" refers to data obtained by converting voice data into character information.

[0198] "Analysis" is the process of analyzing the converted text data and generating reports, lists, schedules, etc. based on it.

[0199] Minutes are documents that organize and summarize the work content, important points, and issues.

[0200] A "ToDo list" is a list in bulleted form that lists the next tasks to be done and items that need to be improved.

[0201] "Schedule" is a calendar listing the next work schedule and maintenance plan.

[0202] An "email draft" is a draft of an email containing a work report or contact information.

[0203] A "user terminal" is an electronic device such as a smartphone or tablet used by a worker.

[0204] The "Factory Work Report" is a report summarizing the details of the maintenance work carried out within the factory.

[0205] "Work schedule" is schedule information including the next maintenance work and check items.

[0206] A "work report email" is an email containing the report content to be sent to the relevant parties after the maintenance work has been completed.

[0207] "Cloud storage" is a service for storing data over the Internet.

[0208] This invention is a system for improving the efficiency of maintenance work in factories. Specifically, it converts voice data during work into text data, and then uses a generative AI model based on that text data to automatically generate drafts of work reports, to-do lists, work schedules, and thank-you emails. This system is implemented as follows.

[0209] First, the user, a worker, uses a device such as a smartphone or tablet to record the details of the maintenance work in the factory by voice. The hardware used for this is a smartphone or tablet. The recorded data is managed using a cloud server.

[0210] Once the audio data is acquired, the user uploads the recording to the cloud server via their device, and the cloud server stores the received recording data in its storage.

[0211] The server then uses a cloud-based voice recognition engine (e.g., Google Cloud Speech-to-Text) to convert the recorded data into high-quality text data, which is then stored in cloud storage.

[0212] The generated text data is used with a generative AI model (e.g., OpenAI GPT-4) to produce the following output:

[0213] 1. Generation of work reports: Analyze text data, extract key points of maintenance work, and generate reports.

[0214] 2. Generate a to-do list: Extract next tasks and areas for improvement from text data and create a to-do list.

[0215] 3. Generate work schedule: Analyze the date and time information in the text data and generate the next maintenance schedule in calendar format.

[0216] 4. Generate a thank you email: Draft a thank you email based on the maintenance work.

[0217] This generated data is sent from the server to the user's device. The user receives a notification, can check and edit the generated data, and use it as needed. This significantly reduces the time required for post-processing after maintenance work.

[0218] Examples of illustrative prompts include:

[0219] Generate meeting minutes, to-do lists, appointments, and thank you emails based on:

[0220] "The following problem was discovered during today's maintenance work. There is a small leak in the hydraulic system and the valve needs to be replaced. The next inspection date is October 20th. All other inspection items are normal."

[0221] The system according to the present invention has the effect of significantly reducing the burden on workers and improving work efficiency, and also reduces errors in reporting and schedule setting.

[0222] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0223] Step 1:

[0224] Users can use their smartphones or tablets to record their maintenance work by voice, and the recorded data is saved on the device.

[0225] Input: User voice input

[0226] Output: Recording data saved on the device

[0227] Specific operation: The user presses the start recording button, talks about the maintenance work and problems, and then presses the stop button to end the recording.

[0228] Step 2:

[0229] Users upload recorded data from their devices to the cloud server, which receives the data and stores it in its storage.

[0230] Input: Recording data stored on the device

[0231] Output: Recording data stored on a cloud server

[0232] Specific operation: The user presses the upload button to send the recorded data from the device to the cloud server. The server receives the recorded data and stores it in a database.

[0233] Step 3:

[0234] The server converts the recorded data into text data using a speech recognition engine, and stores the converted text data in cloud storage.

[0235] Input: Recording data stored on a cloud server

[0236] Output: Text data stored in cloud storage

[0237] Specific operation: The server uses a speech recognition engine such as Google Cloud Speech-to-Text to analyze the recorded data and convert it into high-quality Japanese text. The converted text data is then stored in cloud storage.

[0238] Step 4:

[0239] The server uses a generative AI model to generate the following outputs based on the text data:

[0240] business report

[0241] To-Do List

[0242] Work Schedule

[0243] Work report email

[0244] Input: Text data stored in cloud storage

[0245] Output: Generated work reports, ToDo lists, work schedules, work report emails

[0246] Specific operation: The server inputs text data into a generative AI model such as OpenAI GPT-4, and generates prompt sentences for minutes, to-do lists, schedules, and thank-you emails.

[0247] Step 5:

[0248] The server sends the generated data to the user's terminal, where the user receives the data and checks and edits it.

[0249] Input: Generated work reports, ToDo lists, work schedules, work report emails

[0250] Output: Data sent to the user's terminal

[0251] Specific operation: The server sends the generated data to the user's terminal. The user receives a notification, checks it, and edits it if necessary.

[0252] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0253] This invention is an advanced sales support system incorporating an emotion engine that recognizes user emotions, and is intended to reduce the workload of sales representatives after customer visits. Specifically, it converts voice data into high-quality text data, analyzes that text data, and automatically generates drafts of meeting minutes, to-do lists, schedules, and thank-you emails. In addition, by adding emotion analysis using the emotion engine, it provides more flexible and personalized output.

[0254] Explanation of program processing

[0255] Acquiring and uploading audio data

[0256] When a user visits a customer, the user records the details of the business meeting using a device such as a smartphone or tablet. Once the recording is complete, the user uploads the recording data from the device to the server. During this process, the server receives the recording data and prepares it for processing.

[0257] Japanese text

[0258] The server converts the received recording data into high-quality Japanese text using a speech recognition engine. The converted text data is stored in the server and used in the next step.

[0259] Analysis by emotion engine

[0260] The server inputs the text data into the emotion engine to analyze the user's emotions. The emotion engine extracts emotional elements (happiness, sadness, surprise, etc.) from the text and reflects the results in the next generation AI process.

[0261] Processing of generated AI

[0262] When text data containing sentiment analysis results is input into the generative AI, the server generates the following outputs:

[0263] 1. Generate transcripts:

[0264] The server analyzes the text data and generates minutes that reflect the key points of the business meeting and the user's emotions, allowing the system to record not only the important points of the business meeting but also the emotional aspects of the discussion.

[0265] 2. Create a to-do list:

[0266] It extracts tasks and homework from text data and creates a to-do list based on the user's emotions, allowing the user to manage task priorities so that they do not feel stressed.

[0267] 3. Create an appointment:

[0268] Based on text data and sentiment analysis, it generates upcoming meetings and important appointments in Google Calendar format. By taking into account sentiment information, it helps users to comfortably complete their next appointment.

[0269] 4. Draft thank you email:

[0270] The server creates a draft thank-you email based on the details of the business and the results of sentiment analysis, allowing it to generate an email that incorporates emotionally appropriate greetings and words of thanks.

[0271] feedback

[0272] The generated minutes, to-do lists, schedules, and draft thank-you emails are sent from the server to the user's device. The user receives a notification and can check and edit each piece of data. This significantly reduces the time spent on post-processing after returning to the office, allowing them to focus on their next sales activity.

[0273] Specific examples

[0274] Customer Visit Scenario

[0275] A user visits a customer and conducts a business negotiation. During the negotiation, the contents of the negotiation are recorded on the device. After the negotiation is over, the user presses the "Stop" button to stop recording, and then presses the "Upload" button to send the recorded data to the server.

[0276] Specific operations of server-side processing

[0277] 1. The server receives the recording data and saves it in storage.

[0278] 2. The server uses a speech recognition engine to convert the recorded data into high-quality Japanese text, and stores the converted text in storage.

[0279] 3. The server inputs the text data into the emotion engine to analyze the user's emotions, and passes the emotion analysis results to the generative AI model.

[0280] 4. The server uses the generative AI model to generate the following data:

[0281] Minutes: Include emotional information, such as "The key points of the meeting were..."

[0282] To-do lists: Emotionally driven management, such as "Homework items for next time..."

[0283] Google Calendar Appointments: Emotionally-driven schedules, such as "Your next meeting is scheduled for August 10th at 10am."

[0284] Draft of a thank you email: A draft that reflects your feelings, such as "Thank you for the business meeting the other day..."

[0285] 5. The generated data is collected and sent to the user's device. The user receives a notification, checks and edits the generated data, and uses it as needed.

[0286] In this way, the system of the present invention recognizes and analyzes the user's emotions and generates output that reflects them, thereby providing more personalized sales support and reducing the workload of sales representatives.

[0287] The processing flow will be explained below.

[0288] Step 1:

[0289] When a user visits a customer, they use a device (smartphone or tablet) to record the details of the business meeting. They launch the device's voice recording app and press the "Record" button to start recording. When the business meeting is over, they press the "Stop" button to stop recording.

[0290] Step 2:

[0291] The user uploads the recorded data to the server from the voice recording app on their device. After pressing the "Upload" button on the app and confirming the connection to the server, the recorded data is sent to the server.

[0292] Step 3:

[0293] The server receives the uploaded recording, stores it in storage, and adds it to the conversion queue.

[0294] Step 4:

[0295] The server calls the speech recognition engine to convert the recorded data into high-quality Japanese text. The speech recognition engine analyzes the recorded data and generates Japanese text data. The converted text data is saved in storage.

[0296] Step 5:

[0297] The server inputs the stored Japanese text data into the emotion engine to analyze the user's emotions. The emotion engine analyzes the text data and extracts emotional elements (joy, sadness, surprise, etc.). The emotion analysis results are passed to the generative AI model.

[0298] Step 6:

[0299] The server generates meeting minutes using the generative AI model. Taking into account the results of sentiment analysis, it extracts key points from the text data and creates minutes in bullet point format. The minutes also reflect emotional comments and important tones. The generated minutes data is saved in storage.

[0300] Step 7:

[0301] The server generates a to-do list using the generative AI model. Based on the results of sentiment analysis, action items (homework items) are extracted from the text data. A to-do list is created that takes into account the user's emotional state, task priority, and workload. The generated to-do list data is saved in storage.

[0302] Step 8:

[0303] The server generates schedules using a generative AI model. Taking into account the results of sentiment analysis, it extracts date and time information from the text data and creates upcoming meetings and important appointments in Google Calendar format. Based on the emotional information, it adjusts the schedule so that the user can complete it comfortably. The schedule data is automatically registered in Google Calendar.

[0304] Step 9:

[0305] The server uses the generative AI model to create a draft of a thank-you email. Based on the results of the sentiment analysis, it creates a draft email that includes a greeting and expression of gratitude appropriate to the business negotiation. It uses an appropriate tone and expressions depending on the emotional information. The generated email draft data is saved in storage.

[0306] Step 10:

[0307] The server sends the generated minutes, to-do list, Google Calendar schedule, and a draft of the thank-you email to the user's device. The server also sends a notification to the user's device and provides a download link for the generated data.

[0308] Step 11:

[0309] The user receives a notification and clicks the provided link to download the generated data to their device. The user can then review and edit the data as needed and use it for their next sales activity.

[0310] Through the above processing steps, the system can recognize the user's emotions and generate output that reflects them, thereby providing more personalized sales support and reducing the workload of sales representatives.

[0311] Example 2

[0312] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0313] Modern sales activities require a wide range of post-event processing, such as creating reports after customer visits, organizing tasks, setting up next-time appointments, and writing follow-up emails. These tasks require time and effort, placing a significant burden on sales representatives. Furthermore, accurate recording of customer interactions and conveying information that incorporates emotions and nuances is required, but doing this manually is difficult. There is a need for the development of a system that can solve these issues, reduce the workload of sales representatives, and provide efficient, personalized sales support.

[0314] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for acquiring voice data, means for converting the acquired voice data into text data, means for analyzing the converted text data to recognize emotions, means for analyzing the text data including the emotion recognition results to generate minutes, means for analyzing the text data including the emotion recognition results to generate a to-do list, means for analyzing the text data including the emotion recognition results to generate a schedule, means for analyzing the text data including the emotion recognition results to generate a draft of an e-mail, and means for transmitting the generated data to a user terminal. This allows the content of a conversation with a customer to be accurately recorded, information including emotions and nuances to be automatically generated, and sales representatives to quickly and efficiently perform follow-up processing.

[0315] The "means for acquiring voice data" is a function for recording the contents of business negotiations via the terminal used by the user when visiting a customer, and transmitting the recorded data to the server.

[0316] The "means for converting acquired voice data into text data" is a function in which the server converts the received voice data into high-quality text format using a voice recognition engine.

[0317] The "means for analyzing the converted text data and recognizing emotions" is a function by which the server inputs the text data into an emotion analysis engine and extracts and recognizes the emotional elements contained in the text data.

[0318] The "means for generating minutes by analyzing text data including emotion recognition results" is a function in which the server analyzes text data that reflects emotion recognition results and automatically generates minutes that include the main points and emotional aspects of the business negotiations.

[0319] The "means for generating a to-do list by analyzing text data that includes emotion recognition results" is a function in which the server extracts tasks and homework from text data that reflects emotion recognition results, and generates a to-do list prioritized based on emotions.

[0320] The "means for generating schedules by analyzing text data including emotion recognition results" is a function in which the server automatically generates the next meeting or important schedule based on text data that reflects emotion recognition results.

[0321] The "means for analyzing text data including emotion recognition results and generating a draft of an e-mail" is a function in which the server automatically generates a draft of a thank-you e-mail including an appropriate greeting and words of gratitude based on the text data in which the emotion recognition results are reflected.

[0322] "Means for transmitting generated data to the user terminal" refers to a function for transmitting the minutes, to-do list, schedule, and draft thank-you email generated by the server to the user terminal.

[0323] MODE FOR CARRYING OUT THE INVENTION

[0324] This invention is an advanced sales support system incorporating an emotion engine that recognizes user emotions, and is intended to reduce the workload of sales representatives after customer visits. Specifically, it converts voice data into high-quality text data, analyzes that text data, and automatically generates drafts of meeting minutes, to-do lists, schedules, and thank-you emails. In addition, by adding emotion analysis using the emotion engine, it provides more flexible and personalized output.

[0325] When visiting a customer, the user records the details of the business meeting using a device such as a smartphone or tablet. Once the recording is complete, the user uploads the recording data from the device to the server. During this process, the server receives the recording data and prepares it for processing.

[0326] The server converts the received recording data into high-quality Japanese text. To do this, the server uses a speech recognition engine (e.g., Google Speech-to-Text API). The converted text data is stored on the server and used in the next step.

[0327] The server inputs the saved text data into an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotions. The emotion engine extracts emotional elements such as joy, sadness, and surprise from the text and reflects the results in the generative AI model.

[0328] When text data containing sentiment analysis results is input into a generative AI model (e.g., OpenAI GPT-4), the server generates the following outputs:

[0329] 1. Minutes:

[0330] The text data that reflects the emotion recognition results is analyzed to generate minutes that include the key points and emotional aspects of the business negotiations.

[0331] Example: "The key points of the discussion are..."

[0332] 2. To-Do List:

[0333] Tasks and homework are extracted from text data that reflects emotion recognition results, and a to-do list prioritized based on emotion is generated.

[0334] Example: "As homework for next time..."

[0335] 3. Plan:

[0336] Generate upcoming meetings and important appointments in Google Calendar format based on text data and sentiment analysis.

[0337] By taking emotional information into consideration, we ensure that users can comfortably complete their next appointment.

[0338] Example: "Our next meeting is scheduled for August 10th at 10am."

[0339] 4. Draft of a thank you email:

[0340] Draft a thank you email based on the deal details and sentiment analysis results.

[0341] Example: "Thank you for the business meeting the other day..."

[0342] The generated minutes, to-do lists, schedules, and draft thank-you emails are sent from the server to the user's device. The user receives a notification and can check and edit each piece of data. This significantly reduces the time spent on post-processing after returning to the office, allowing them to focus on their next sales activity.

[0343] Specific examples

[0344] A user visits a customer and conducts a business negotiation. During the negotiation, the contents of the negotiation are recorded on the device. After the negotiation is over, the user presses the "Stop" button to stop recording, and then presses the "Upload" button to send the recorded data to the server.

[0345] The server receives the recorded data and saves it in storage. The received recorded data is converted into high-quality Japanese text using a speech recognition engine (e.g., Google Speech-to-Text API), and the converted text data is also saved in storage. Next, the server inputs the text data into an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotions. The emotion analysis results are then passed to a generative AI model (e.g., OpenAI GPT-4) to generate the following data:

[0346] Minutes: Include emotional information, such as "The key points of the meeting were..."

[0347] To-do lists: Emotionally driven management, such as "Homework items for next time..."

[0348] Google Calendar Appointments: Emotionally-driven schedules, such as "Your next meeting is scheduled for August 10th at 10am."

[0349] Draft of a thank you email: A draft that reflects your feelings, such as "Thank you for the business meeting the other day..."

[0350] The generated data is collected and sent to the user's device, where the user receives a notification, can review and edit the generated data, and use it as needed. This system recognizes and analyzes the user's emotions and generates output that reflects them, thereby providing more personalized sales support and reducing the workload of sales representatives.

[0351] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0352] Step 1:

[0353] When a user visits a customer, the user uses a device such as a smartphone or tablet to record the details of the business meeting. Specifically, when the user presses the record button on the device, the device begins recording audio data. After the business meeting is over, the user presses the "Stop" button to stop recording, and then presses the "Upload" button to send the recorded data to the server. The input is the user's voice, and the output is the audio file sent to the server.

[0354] Step 2:

[0355] The server saves the voice data received from the device in initial storage. Next, the server uses a voice recognition engine (e.g., Google Speech-to-Text API) to convert the voice data into high-quality Japanese text. The specific operation of this process is to pass the voice data to the API and receive the converted text data. The input is an audio file, and the output is the converted text data.

[0356] Step 3:

[0357] The server inputs the converted text data into an emotion engine (e.g., IBM Watson Tone Analyzer) for emotion analysis. The emotion engine analyzes the content of the text data and extracts emotional elements such as joy, sadness, and surprise. The specific operation of this process is for the server to pass the text data to the emotion engine's API and receive the emotion analysis results. The input is text data, and the output is data containing the emotion analysis results.

[0358] Step 4:

[0359] The server inputs the text data including the sentiment analysis results into a generative AI model (e.g., OpenAI GPT-4) to generate various documents. The specific operation at this time is to pass the sentiment analysis results and text data to the generative AI model as prompt sentences, and obtain the following outputs:

[0360] 1. Meeting minutes - Minutes that include the key points and emotional aspects of the business meeting.

[0361] 2. To-Do List - A list that reflects emotion-based task management.

[0362] 3. Appointments - Schedule your next meeting or important appointment.

[0363] 4. Draft of a thank you email - An email containing appropriate greetings and thank you words that reflect the sentiment. The input is text data and sentiment analysis results, and the output is the respective documents.

[0364] Step 5:

[0365] The server sends the generated minutes, to-do list, schedule, and thank-you email drafts to the user's device. Specifically, the server sends each generated document to the device linked to the user's account and notifies them. The input is each generated document, and the output is the transmission and notification to the user's device.

[0366] Step 6:

[0367] The user can review and edit the drafts of meeting minutes, to-do lists, schedules, and thank-you emails generated on their device and use them as needed. Specifically, the user can open the application on their device and view and edit the contents of each document. The input is each document sent from the server, and the output is the final document edited by the user.

[0368] (Application example 2)

[0369] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0370] Conventional sales support systems and customer service systems have difficulty responding flexibly based on customer emotions, and have not been able to fully improve customer satisfaction or sales efficiency. In particular, there is no system that can analyze customer emotions in real time and provide appropriate responses to customer service staff based on that analysis, so there is a need for more advanced customer service support.

[0371] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring voice data, means for converting the acquired voice data into text data, means for analyzing the converted text data using an emotion engine, means for generating minutes using a generative AI model including the analyzed text data and emotion analysis results, means for generating a to-do list using the generative AI model, means for generating a schedule using the generative AI model, means for generating a draft email using the generative AI model, means for displaying the generated data and emotion analysis results on a user interface in real time, and means for transmitting the generated data to a user terminal. This makes it possible to analyze customer emotions in real time, present personalized countermeasures based on the results, and efficiently and quickly process data after customer service.

[0372] "Audio data" refers to audio signals recorded in digital format, and is information that can be converted into text data and analyzed.

[0373] "Text data" refers to written information converted from audio data and used as input for analytical and generative AI models.

[0374] An "emotion engine" is a software technology for extracting emotional elements from text data and analyzing emotions.

[0375] A "generative AI model" is an artificial intelligence technology that uses machine learning algorithms to automatically generate output such as minutes, to-do lists, schedules, and email drafts from input data.

[0376] "User interface" refers to the screen or operation panel that allows users to operate the system, and which displays sentiment analysis results and suggestions in real time.

[0377] A "user terminal" is a device that is directly operated by a user, and includes smartphones, tablets, smart glasses, and the like.

[0378] "Real-time" means that data acquisition, processing, and output are all done instantly, and information is provided without delay.

[0379] The present invention is a system for efficiently supporting customer service, and is implemented in the following procedure.

[0380] Program Generation

[0381] Hardware and software used

[0382] The system uses the following hardware and software:

[0383] Smart glasses (device example: general noun)

[0384] Speech recognition engine (software example: Google Cloud Speech-to-Text)

[0385] Sentiment analysis engine (software example: IBM Watson Tone Analyzer)

[0386] Generative AI model (software example: OpenAI GPT-3)

[0387] Cloud storage (software example: general noun)

[0388] Acquiring and uploading audio data

[0389] The user puts on the smart glasses and starts talking to the customer. The smart glasses record the conversation in real time and upload the audio data to the server.

[0390] Converting audio data to text

[0391] The server uses a speech recognition engine to convert the received voice data into high-quality text data, which is then stored in cloud storage.

[0392] Analysis by emotion engine

[0393] The server inputs the converted text data into a sentiment analysis engine to analyze customer sentiment in real time, and the analysis results are fed back to the smart glasses.

[0394] Data generation using generative AI models

[0395] Based on the analyzed text data and sentiment analysis results, the generative AI model generates the following data:

[0396] 1. Meeting minutes: Automatically generate meeting minutes that reflect the key points and sentiment of the conversation.

[0397] 2. To-Do List: Extract upcoming tasks and generate a prioritized to-do list.

[0398] 3. Schedule: Generate upcoming meetings and important events in Google Calendar format.

[0399] 4. Email Draft: Generate an emotionally relevant thank you email draft.

[0400] Real-time display and feedback

[0401] The generated data and sentiment analysis results are displayed in real time on the smart glasses' user interface, allowing users to immediately identify appropriate countermeasures. After the analysis is complete, the generated data is sent to the user's device for review and editing.

[0402] Specific examples

[0403] Customer Visit Scenario

[0404] A user visits a customer, puts on smart glasses, and begins a conversation. The smart glasses record the voice and upload it to the server in real time. The server converts the voice data into text and analyzes it using a sentiment analysis engine. The analysis results are displayed on the smart glasses' user interface, and the user can take appropriate action. The generative AI model then generates drafts of meeting minutes, to-do lists, schedules, and thank-you emails, which are saved in cloud storage.

[0405] Prompt Sentence Examples

[0406] Conversation:

[0407] "Hi, can you tell me more about this product? Is it really easy to use?"

[0408] Sentiment analysis results:

[0409] "The customer is interested. Please provide a detailed product description and use case."

[0410] Generated transcript:

[0411] "Service: The customer is interested in Product A and asks about its ease of use. A detailed explanation and examples of use are provided."

[0412] Generated thank you email draft:

[0413] "Thank you very much for visiting our store the other day. We received your question about Product A and we explained the details to you. If there is anything else you would like to ask, please feel free to contact us."

[0414] In this embodiment, the present invention can analyze customer emotions in real time, provide personalized countermeasures based on the analysis, and process data after the customer service quickly and efficiently.

[0415] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0416] Step 1:

[0417] The user puts on the smart glasses and starts a conversation with a customer. The smart glasses record the conversation in real time and capture the audio data. The captured audio data is then uploaded from the smart glasses to a server.

[0418] Input: Real-time audio data

[0419] Output: Uploaded audio data

[0420] Step 2:

[0421] The server inputs the uploaded voice data into a speech recognition engine (e.g., Google Cloud Speech-to-Text) and converts it into high-quality text data. The converted text data is then stored in the server's internal storage.

[0422] Input: Audio data

[0423] Output: Text data

[0424] Step 3:

[0425] The server inputs the converted text data into an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze emotions in real time. The analysis results are added to the text data and saved in storage as emotion analysis results.

[0426] Input: Text data

[0427] Output: Text data with sentiment analysis results

[0428] Step 4:

[0429] The server inputs the text data with sentiment analysis results into a generative AI model (e.g., OpenAI GPT-3) to generate the following data:

[0430] 1. Minutes

[0431] 2. To-Do List

[0432] 3. Schedule

[0433] 4. Draft the email

[0434] Each generated data is stored separately in storage.

[0435] Input: Text data with sentiment analysis results

[0436] Output: minutes, to-do lists, schedules, email drafts

[0437] Step 5:

[0438] The server provides feedback to the user by displaying the generated data and the emotion analysis results in real time on the user interface (smart glasses). The user can then check this information in real time and take appropriate action.

[0439] Input: Generated data, sentiment analysis results

[0440] Output: Real-time display on smart glasses

[0441] Step 6:

[0442] After the customer service is over, the server sends the generated data to the user's device. The user can check the data on the device and edit or add to it as needed. The data is also saved in cloud storage.

[0443] Input: Generated data

[0444] Output: Data sent to user devices, data saved to cloud storage

[0445] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0446] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0447] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0448] [Second embodiment]

[0449] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0450] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0451] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0452] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0453] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0454] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0455] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0456] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0457] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0458] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0459] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0460] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0461] This invention is a system to reduce the workload of sales representatives after customer visits. Specifically, it converts voice data into high-quality text data, and then uses a generative AI model to automatically generate meeting minutes, to-do lists, schedules, and thank-you emails.

[0462] Explanation of program processing

[0463] Acquiring and uploading audio data

[0464] When visiting a customer, the user records the details of the business meeting using a device such as a smartphone or tablet. Once the recording is complete, the user uploads the recorded data from the device to the server. The server then receives the recorded data and prepares it for the next process.

[0465] Japanese text

[0466] The server converts the received recording data into high-quality Japanese text using a speech recognition engine. The converted text data is stored in the server and used in the next step.

[0467] Processing of generated AI

[0468] Once the text data is created, the server uses a generative AI model to generate the following outputs:

[0469] 1. Generate transcripts:

[0470] The server analyzes the text data, extracts the key points of the sales negotiation, and generates minutes in bullet-point format. The minutes summarize the important points of the sales negotiation, which are useful for sales representatives to review later.

[0471] 2. Create a to-do list:

[0472] Tasks and homework are extracted from the text data and the server creates a to-do list, which lists the tasks that need to be completed before the next business meeting.

[0473] 3. Create an appointment:

[0474] By parsing the date and time information in the text data, the server generates a Google Calendar of upcoming meetings and important events, including the date and time of your next client visit and important deadlines.

[0475] 4. Draft thank you email:

[0476] The server creates a draft of a thank-you email based on the details of the sales meeting, allowing the sales representative to send a thank-you email quickly after the meeting.

[0477] feedback

[0478] The generated minutes, to-do lists, schedules, and draft thank-you emails are sent from the server to the user's device. The user receives a notification and can check and edit each piece of data. This significantly reduces the time spent on post-processing after returning to the office, allowing them to focus on their next sales activity.

[0479] Specific examples

[0480] Customer Visit Scenario

[0481] A user visits a customer and conducts a business negotiation. During the negotiation, the contents of the negotiation are recorded on the device. After the negotiation is over, the user presses the "Stop" button to stop recording, and then presses the "Upload" button to send the recorded data to the server.

[0482] Specific operations of server-side processing

[0483] 1. The server receives the recording data and saves it in storage.

[0484] 2. The server uses a speech recognition engine to convert the recorded data into high-quality Japanese text, and stores the converted text in storage.

[0485] 3. The server uses the generative AI model to generate the following data:

[0486] Minutes: "Key points of the meeting were..."

[0487] To-do list: "Homework items for next time..."

[0488] Google Calendar events: "Your next meeting is scheduled for August 10th at 10am."

[0489] A sample thank you email: "Thank you for the business meeting the other day..."

[0490] 4. The generated data is collected and sent to the user's device. The user receives a notification, checks and edits the generated data, and uses it as needed.

[0491] As described above, the system according to the present invention has the effect of largely automating the post-processing after a customer visit, thereby reducing the burden on sales staff.

[0492] The processing flow will be explained below.

[0493] Step 1:

[0494] When a user visits a customer, they use a device (smartphone or tablet) to record the details of the business meeting. They launch the device's voice recording app and press the "Record" button to start recording. When the business meeting is over, they press the "Stop" button to stop recording.

[0495] Step 2:

[0496] The user uploads the recorded data to the server from the voice recording app on their device. After pressing the "Upload" button on the app and confirming the connection to the server, the recorded data is sent to the server.

[0497] Step 3:

[0498] The server receives the uploaded recording, stores it in storage, and adds it to the conversion queue.

[0499] Step 4:

[0500] The server calls the speech recognition engine to convert the recorded data into high-quality Japanese text. The speech recognition engine analyzes the recorded data and generates Japanese text data. The converted text data is saved in storage.

[0501] Step 5:

[0502] The server inputs the stored Japanese text data into the generative AI model for analysis, analyzes the target text data, and generates various output data.

[0503] Step 6:

[0504] The server analyzes the text data, extracts the main points of the business negotiations, and generates minutes in bullet-point format. The generated minutes data is then saved in storage.

[0505] Step 7:

[0506] The server extracts action items (homework items) from the text data and generates a ToDo list. The generated ToDo list data is saved in storage.

[0507] Step 8:

[0508] The server extracts date and time information from the text data and adds the event to Google Calendar. It also calls the Google Calendar API to register upcoming meetings and important events.

[0509] Step 9:

[0510] The server creates a draft of the thank-you email based on the details of the business negotiation. It analyzes the text data and generates a draft email incorporating appropriate greetings and words of thanks. The generated draft email is then saved in storage.

[0511] Step 10:

[0512] The server sends the generated minutes, to-do list, Google Calendar schedule, and a draft of the thank-you email to the user's device. It also sends a notification to the user's device and provides a download link for the generated data.

[0513] Step 11:

[0514] The user receives a notification and clicks the provided link to download the generated data to their device. The user can provide feedback as needed, check and edit the data, and use it in their next sales activity.

[0515] Through the above processing steps, this system can improve the efficiency of the work that sales representatives perform after visiting customers and reduce their workload.

[0516] Example 1

[0517] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0518] The post-processing work that sales representatives perform after customer visits requires time and effort, hindering efficient sales activities. Furthermore, manual text generation and information organization have limited accuracy, leading to the risk of important information being overlooked or typographical errors. There is a need for a system that can solve these problems, reduce the workload of sales representatives, and enable fast and accurate information processing.

[0519] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0520] In this invention, the server includes means for acquiring voice data, means for converting the acquired voice data into text data, means for analyzing the converted text data to generate minutes, means for analyzing the converted text data to generate a to-do list, means for analyzing the converted text data to generate a schedule, means for analyzing the converted text data to generate a draft of an email, means for transmitting the generated data to a user terminal, means for converting the recorded data into Japanese text using a voice recognition engine, and means for generating various information from the text data using a generative AI model. This allows a user to simply record the contents of a business meeting and upload the data to the server, which automatically generates drafts of minutes, to-do lists, schedules, and thank-you emails, enabling efficient and accurate post-event processing.

[0521] "Voice data" refers to data that has been saved in digital format containing the content of conversations and dialogues recorded during customer visits, business negotiations, etc.

[0522] "Means of capture" refers to the combination of devices and software used by a user to record audio data and upload it to a server.

[0523] "Text data" refers to data obtained by converting acquired voice data into text format.

[0524] "Means for converting" refers collectively to a speech recognition engine and related software used to convert voice data into text data.

[0525] "Means of analysis" is a general term for the algorithms and generative AI models used to extract and organize key information from the converted text data.

[0526] Minutes are documents that summarize the main points of business negotiations, meetings, etc., and are used to confirm the contents later.

[0527] A "ToDo list" is a list of tasks or homework that need to be completed by the next time.

[0528] "Schedule" refers to schedule information such as the next date and time of a business meeting or meeting, or important deadlines.

[0529] An "email draft" is a draft of a greeting or thank you email to be used after a business meeting.

[0530] "Generative Means" refers collectively to generative AI models and related algorithms used to create meeting minutes, to-do lists, schedules, and email drafts from text data.

[0531] "Transmission means" is a general term for communication means and protocols for transmitting generated data to a user terminal.

[0532] This invention is a system that uses advanced voice data to automate the post-processing process after sales representatives visit customers. Specifically, it converts voice data into Japanese text data, and then uses a generative AI model based on that text data to automatically generate meeting minutes, to-do lists, schedules, and thank-you emails. This system is primarily comprised of components that play the roles of a server, terminal, and user.

[0533] Hardware and Software Overview

[0534] Terminal

[0535] Users use devices such as smartphones and tablets, which have an application with a recording function installed, allowing them to record the contents of business negotiations.

[0536] server

[0537] The server is the main device that receives, stores, analyzes, and distributes the generated audio data. The following software is installed on the server:

[0538] Speech Recognition Engine: Speech recognition software such as the Google Cloud Speech-to-Text API.

[0539] Generative AI models: Advanced generative AI software such as GPT-3.

[0540] User

[0541] The user is a sales representative who records voice data during a sales meeting and uploads the data to the server after the recording is complete. Post-processing is performed based on the generated information.

[0542] Data processing and calculation flow

[0543] 1. Acquiring and uploading audio data

[0544] When a user visits a customer, the user uses the device to record the details of the business meeting. After completing the recording, the user uploads the audio data from the device to the server. The server receives the recorded data via an HTTP request and stores it in storage.

[0545] 2. Japanese text

[0546] The server passes the received recorded data to a speech recognition engine and converts it into high-quality Japanese text data, which is then saved back into storage.

[0547] 3. Text Data Analysis and Generation

[0548] The server inputs the stored text data into a generative AI model to generate the following information:

[0549] Minutes: Summarize the main points of the business meeting in bullet points.

[0550] To-Do List: Make a list of tasks and homework that need to be done by the next time.

[0551] Appointments: Generate appointments in Google Calendar format based on the dates and times mentioned in the deal.

[0552] Draft a thank you email: Create a draft of a thank you email based on the details of the business meeting.

[0553] 4. Data distribution and feedback

[0554] The server sends the generated minutes, to-do lists, schedules, and thank-you email drafts to the user's device. The user receives a notification, checks and edits the generated data, and uses it as needed.

[0555] Examples and prompts

[0556] Specific examples

[0557] A user visits a customer and conducts a business meeting. The contents of the business meeting are recorded on the device, and after the business meeting is completed, the recording is stopped and the recorded data is uploaded to the server.

[0558] The server receives the recording, converts it into Japanese text using a speech recognition engine, and then uses a generative AI model to generate drafts of meeting minutes, to-do lists, schedules, and thank-you emails, which are then sent to the user's device.

[0559] Prompt Sentence Examples

[0560] By inputting the following prompt sentence into the generative AI model, we can get the corresponding output:

[0561] Prompt for generating transcripts:

[0562] "Please extract the key points from the following sales pitch in bullet points format:

[0563] 1. Customer needs

[0564] 2. Proposal details

[0565] 3. Questions and Answers

[0566] 4. Next assignment

[0567] Prompt for generating a ToDo list:

[0568] From the text below, please list the tasks you need to complete by next time:

[0569] Summary of business negotiations

[0570] Confirm next step

[0571] Preparing for follow-up

[0572] Prompt for generating an appointment:

[0573] "Extract important dates and times from the text below and generate a Google Calendar event."

[0574] Prompt for generating a draft thank you email:

[0575] "Please draft a thank you email based on the following business details:

[0576] Thank you message

[0577] Review the main points of the deal

[0578] Proposal for next steps

[0579] As a result, this system significantly improves the work efficiency of sales staff and reduces the time and effort required for post-sales processing.

[0580] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0581] Step 1:

[0582] Acquiring and uploading audio data

[0583] The user uses a device such as a smartphone or tablet to record the contents of a business meeting. They open a recording application and press the record button to record the contents of the business meeting. The input data is the audio spoken during the business meeting. After the business meeting is over, the user presses the "Stop" button to end recording, and then presses the "Upload" button to send the recorded data to the server. The server receives the HTTP request and saves the recorded data in storage. The output data is an audio file saved in storage.

[0584] Step 2:

[0585] Japanese text

[0586] The server passes the saved recording data to a speech recognition engine. The input data is an audio file. The server uses a speech recognition engine such as the Google Cloud Speech-to-Text API to convert this audio data into high-quality Japanese text data. Specifically, the speech recognition engine analyzes the audio waveform and converts it into text information. The converted text data is then saved back into storage within the server. The output data is Japanese text data.

[0587] Step 3:

[0588] Text data analysis and generation

[0589] The server inputs text data into the generative AI model to generate specific information. The input data is Japanese text data. The server uses the following prompt to have the generative AI model (e.g., GPT-3) analyze the data:

[0590] Prompt for generating transcripts:

[0591] "Please extract the key points from the following sales pitch in bullet points format:

[0592] 1. Customer needs

[0593] 2. Proposal details

[0594] 3. Questions and Answers

[0595] 4. Next assignment

[0596] Prompt for generating a ToDo list:

[0597] From the text below, please list the tasks you need to complete by next time:

[0598] Summary of business negotiations

[0599] Confirm next step

[0600] Preparing for follow-up

[0601] Prompt for generating an appointment:

[0602] "Extract important dates and times from the text below and generate a Google Calendar event."

[0603] Prompt for generating a draft thank you email:

[0604] "Please draft a thank you email based on the following business details:

[0605] Thank you message

[0606] Review the main points of the deal

[0607] Proposal for next steps

[0608] The server uses these prompts to have the generative AI model analyze the text data and generate information for each purpose. The generated information becomes the basis for meeting minutes, to-do lists, schedules, and thank-you emails, all of which are stored in the server's storage. The output data is the various types of generated information.

[0609] Step 4:

[0610] Data distribution and feedback

[0611] The server sends the generated data (minutes, to-do list, schedule, thank-you email draft) to the user's device. The input data is the various types of generated information. The server uses an HTTP response to send the generated data to the user's device. The user receives a notification on their device and checks and edits the generated data. Specifically, the user clicks the notification to open the application, checks the generated minutes, to-do list, schedule, and thank-you email, and edits them as necessary. The output data is the various types of information after it has been checked and edited by the user.

[0612] By following these steps, users can simply upload their voice data to the server, and minutes, to-do lists, schedules, and thank-you email drafts will be automatically generated and provided to them. This series of processes will greatly improve the efficiency of post-sales processing work for sales representatives.

[0613] (Application example 1)

[0614] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0615] During maintenance work in a factory, recording the details of the work and any problems that arise, and then creating work reports and to-do lists accurately and efficiently based on that information, places a heavy burden on workers. This creates the problem of time being spent creating reports and checking schedules, which reduces the efficiency of the work. Furthermore, mistakes are likely to occur in work reports and schedule settings, which can have an impact on factory operations. To solve these problems, an automated system is needed.

[0616] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0617] In this invention, the server includes means for acquiring voice data, means for converting the acquired voice data into text data, means for analyzing the converted text data to generate minutes, means for analyzing the converted text data to generate a to-do list, means for analyzing the converted text data to generate a schedule, means for analyzing the converted text data to generate a draft of an e-mail, means for sending the generated data to a user terminal, means for outputting the generated data as an in-factory work report, means for outputting the generated data as a work schedule in calendar format, and means for outputting the generated data as a work report e-mail. This automates in-factory maintenance work reporting and schedule setting, not only reducing the burden on workers but also improving work efficiency and reducing reporting and scheduling errors.

[0618] "Voice data" refers to audio information that records events, problems, and work conditions that occur during factory work.

[0619] "Text data" refers to data obtained by converting voice data into character information.

[0620] "Analysis" is the process of analyzing the converted text data and generating reports, lists, schedules, etc. based on it.

[0621] Minutes are documents that organize and summarize the work content, important points, and issues.

[0622] A "ToDo list" is a list in bulleted form that lists the next tasks to be done and items that need to be improved.

[0623] "Schedule" is a calendar listing the next work schedule and maintenance plan.

[0624] An "email draft" is a draft of an email containing a work report or contact information.

[0625] A "user terminal" is an electronic device such as a smartphone or tablet used by a worker.

[0626] The "Factory Work Report" is a report summarizing the details of the maintenance work carried out within the factory.

[0627] "Work schedule" is schedule information including the next maintenance work and check items.

[0628] A "work report email" is an email containing the report content to be sent to the relevant parties after the maintenance work has been completed.

[0629] "Cloud storage" is a service for storing data over the Internet.

[0630] This invention is a system for improving the efficiency of maintenance work in factories. Specifically, it converts voice data during work into text data, and then uses a generative AI model based on that text data to automatically generate drafts of work reports, to-do lists, work schedules, and thank-you emails. This system is implemented as follows.

[0631] First, the user, a worker, uses a device such as a smartphone or tablet to record the details of the maintenance work in the factory by voice. The hardware used for this is a smartphone or tablet. The recorded data is managed using a cloud server.

[0632] Once the audio data is acquired, the user uploads the recording to the cloud server via their device, and the cloud server stores the received recording data in its storage.

[0633] The server then uses a cloud-based voice recognition engine (e.g., Google Cloud Speech-to-Text) to convert the recorded data into high-quality text data, which is then stored in cloud storage.

[0634] The generated text data is used with a generative AI model (e.g., OpenAI GPT-4) to produce the following output:

[0635] 1. Generation of work reports: Analyze text data, extract key points of maintenance work, and generate reports.

[0636] 2. Generate a to-do list: Extract next tasks and areas for improvement from text data and create a to-do list.

[0637] 3. Generate work schedule: Analyze the date and time information in the text data and generate the next maintenance schedule in calendar format.

[0638] 4. Generate a thank you email: Draft a thank you email based on the maintenance work.

[0639] This generated data is sent from the server to the user's device. The user receives a notification, can check and edit the generated data, and use it as needed. This significantly reduces the time required for post-processing after maintenance work.

[0640] Examples of illustrative prompts include:

[0641] Generate meeting minutes, to-do lists, appointments, and thank you emails based on:

[0642] "The following problem was discovered during today's maintenance work. There is a small leak in the hydraulic system and the valve needs to be replaced. The next inspection date is October 20th. All other inspection items are normal."

[0643] The system according to the present invention has the effect of significantly reducing the burden on workers and improving work efficiency, and also reduces errors in reporting and schedule setting.

[0644] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0645] Step 1:

[0646] Users can use their smartphones or tablets to record their maintenance work by voice, and the recorded data is saved on the device.

[0647] Input: User voice input

[0648] Output: Recording data saved on the device

[0649] Specific operation: The user presses the start recording button, talks about the maintenance work and problems, and then presses the stop button to end the recording.

[0650] Step 2:

[0651] Users upload recorded data from their devices to the cloud server, which receives the data and stores it in its storage.

[0652] Input: Recording data stored on the device

[0653] Output: Recording data stored on a cloud server

[0654] Specific operation: The user presses the upload button to send the recorded data from the device to the cloud server. The server receives the recorded data and stores it in a database.

[0655] Step 3:

[0656] The server converts the recorded data into text data using a speech recognition engine, and stores the converted text data in cloud storage.

[0657] Input: Recording data stored on a cloud server

[0658] Output: Text data stored in cloud storage

[0659] Specific operation: The server uses a speech recognition engine such as Google Cloud Speech-to-Text to analyze the recorded data and convert it into high-quality Japanese text. The converted text data is then stored in cloud storage.

[0660] Step 4:

[0661] The server uses a generative AI model to generate the following outputs based on the text data:

[0662] business report

[0663] To-Do List

[0664] Work Schedule

[0665] Work report email

[0666] Input: Text data stored in cloud storage

[0667] Output: Generated work reports, ToDo lists, work schedules, work report emails

[0668] Specific operation: The server inputs text data into a generative AI model such as OpenAI GPT-4, and generates prompt sentences for minutes, to-do lists, schedules, and thank-you emails.

[0669] Step 5:

[0670] The server sends the generated data to the user's terminal, where the user receives the data and checks and edits it.

[0671] Input: Generated work reports, ToDo lists, work schedules, work report emails

[0672] Output: Data sent to the user's terminal

[0673] Specific operation: The server sends the generated data to the user's terminal. The user receives a notification, checks it, and edits it if necessary.

[0674] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0675] This invention is an advanced sales support system incorporating an emotion engine that recognizes user emotions, and is intended to reduce the workload of sales representatives after customer visits. Specifically, it converts voice data into high-quality text data, analyzes that text data, and automatically generates drafts of meeting minutes, to-do lists, schedules, and thank-you emails. In addition, by adding emotion analysis using the emotion engine, it provides more flexible and personalized output.

[0676] Explanation of program processing

[0677] Acquiring and uploading audio data

[0678] When a user visits a customer, the user records the details of the business meeting using a device such as a smartphone or tablet. Once the recording is complete, the user uploads the recording data from the device to the server. During this process, the server receives the recording data and prepares it for processing.

[0679] Japanese text

[0680] The server converts the received recording data into high-quality Japanese text using a speech recognition engine. The converted text data is stored in the server and used in the next step.

[0681] Analysis by emotion engine

[0682] The server inputs the text data into the emotion engine to analyze the user's emotions. The emotion engine extracts emotional elements (happiness, sadness, surprise, etc.) from the text and reflects the results in the next generation AI process.

[0683] Processing of generated AI

[0684] When text data containing sentiment analysis results is input into the generative AI, the server generates the following outputs:

[0685] 1. Generate transcripts:

[0686] The server analyzes the text data and generates minutes that reflect the key points of the business meeting and the user's emotions, allowing the system to record not only the important points of the business meeting but also the emotional aspects of the discussion.

[0687] 2. Create a to-do list:

[0688] It extracts tasks and homework from text data and creates a to-do list based on the user's emotions, allowing the user to manage task priorities so that they do not feel stressed.

[0689] 3. Create an appointment:

[0690] Based on text data and sentiment analysis, it generates upcoming meetings and important appointments in Google Calendar format. By taking into account sentiment information, it helps users to comfortably complete their next appointment.

[0691] 4. Draft thank you email:

[0692] The server creates a draft thank-you email based on the details of the business and the results of sentiment analysis, allowing it to generate an email that incorporates emotionally appropriate greetings and words of thanks.

[0693] feedback

[0694] The generated minutes, to-do lists, schedules, and draft thank-you emails are sent from the server to the user's device. The user receives a notification and can check and edit each piece of data. This significantly reduces the time spent on post-processing after returning to the office, allowing them to focus on their next sales activity.

[0695] Specific examples

[0696] Customer Visit Scenario

[0697] A user visits a customer and conducts a business negotiation. During the negotiation, the contents of the negotiation are recorded on the device. After the negotiation is over, the user presses the "Stop" button to stop recording, and then presses the "Upload" button to send the recorded data to the server.

[0698] Specific operations of server-side processing

[0699] 1. The server receives the recording data and saves it in storage.

[0700] 2. The server uses a speech recognition engine to convert the recorded data into high-quality Japanese text, and stores the converted text in storage.

[0701] 3. The server inputs the text data into the emotion engine to analyze the user's emotions, and passes the emotion analysis results to the generative AI model.

[0702] 4. The server uses the generative AI model to generate the following data:

[0703] Minutes: Include emotional information, such as "The key points of the meeting were..."

[0704] To-do lists: Emotionally driven management, such as "Homework items for next time..."

[0705] Google Calendar Appointments: Emotionally-driven schedules, such as "Your next meeting is scheduled for August 10th at 10am."

[0706] Draft of a thank you email: A draft that reflects your feelings, such as "Thank you for the business meeting the other day..."

[0707] 5. The generated data is collected and sent to the user's device. The user receives a notification, checks and edits the generated data, and uses it as needed.

[0708] In this way, the system of the present invention recognizes and analyzes the user's emotions and generates output that reflects them, thereby providing more personalized sales support and reducing the workload of sales representatives.

[0709] The processing flow will be explained below.

[0710] Step 1:

[0711] When a user visits a customer, they use a device (smartphone or tablet) to record the details of the business meeting. They launch the device's voice recording app and press the "Record" button to start recording. When the business meeting is over, they press the "Stop" button to stop recording.

[0712] Step 2:

[0713] The user uploads the recorded data to the server from the voice recording app on their device. After pressing the "Upload" button on the app and confirming the connection to the server, the recorded data is sent to the server.

[0714] Step 3:

[0715] The server receives the uploaded recording, stores it in storage, and adds it to the conversion queue.

[0716] Step 4:

[0717] The server calls the speech recognition engine to convert the recorded data into high-quality Japanese text. The speech recognition engine analyzes the recorded data and generates Japanese text data. The converted text data is saved in storage.

[0718] Step 5:

[0719] The server inputs the stored Japanese text data into the emotion engine to analyze the user's emotions. The emotion engine analyzes the text data and extracts emotional elements (joy, sadness, surprise, etc.). The emotion analysis results are passed to the generative AI model.

[0720] Step 6:

[0721] The server generates meeting minutes using the generative AI model. Taking into account the results of sentiment analysis, it extracts key points from the text data and creates minutes in bullet point format. The minutes also reflect emotional comments and important tones. The generated minutes data is saved in storage.

[0722] Step 7:

[0723] The server generates a to-do list using the generative AI model. Based on the results of sentiment analysis, action items (homework items) are extracted from the text data. A to-do list is created that takes into account the user's emotional state, task priority, and workload. The generated to-do list data is saved in storage.

[0724] Step 8:

[0725] The server generates schedules using a generative AI model. Taking into account the results of sentiment analysis, it extracts date and time information from the text data and creates upcoming meetings and important appointments in Google Calendar format. Based on the emotional information, it adjusts the schedule so that the user can complete it comfortably. The schedule data is automatically registered in Google Calendar.

[0726] Step 9:

[0727] The server uses the generative AI model to create a draft of a thank-you email. Based on the results of the sentiment analysis, it creates a draft email that includes a greeting and expression of gratitude appropriate to the business negotiation. It uses an appropriate tone and expressions depending on the emotional information. The generated email draft data is saved in storage.

[0728] Step 10:

[0729] The server sends the generated minutes, to-do list, Google Calendar schedule, and a draft of the thank-you email to the user's device. The server also sends a notification to the user's device and provides a download link for the generated data.

[0730] Step 11:

[0731] The user receives a notification and clicks the provided link to download the generated data to their device. The user can then review and edit the data as needed and use it for their next sales activity.

[0732] Through the above processing steps, the system can recognize the user's emotions and generate output that reflects them, thereby providing more personalized sales support and reducing the workload of sales representatives.

[0733] Example 2

[0734] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0735] Modern sales activities require a wide range of post-event processing, such as creating reports after customer visits, organizing tasks, setting up next-time appointments, and writing follow-up emails. These tasks require time and effort, placing a significant burden on sales representatives. Furthermore, accurate recording of customer interactions and conveying information that incorporates emotions and nuances is required, but doing this manually is difficult. There is a need for the development of a system that can solve these issues, reduce the workload of sales representatives, and provide efficient, personalized sales support.

[0736] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for acquiring voice data, means for converting the acquired voice data into text data, means for analyzing the converted text data to recognize emotions, means for analyzing the text data including the emotion recognition results to generate minutes, means for analyzing the text data including the emotion recognition results to generate a to-do list, means for analyzing the text data including the emotion recognition results to generate a schedule, means for analyzing the text data including the emotion recognition results to generate a draft of an e-mail, and means for transmitting the generated data to a user terminal. This allows the content of a conversation with a customer to be accurately recorded, information including emotions and nuances to be automatically generated, and sales representatives to quickly and efficiently perform follow-up processing.

[0737] The "means for acquiring voice data" is a function for recording the contents of business negotiations via the terminal used by the user when visiting a customer, and transmitting the recorded data to the server.

[0738] The "means for converting acquired voice data into text data" is a function in which the server converts the received voice data into high-quality text format using a voice recognition engine.

[0739] The "means for analyzing the converted text data and recognizing emotions" is a function by which the server inputs the text data into an emotion analysis engine and extracts and recognizes the emotional elements contained in the text data.

[0740] The "means for generating minutes by analyzing text data including emotion recognition results" is a function in which the server analyzes text data that reflects emotion recognition results and automatically generates minutes that include the main points and emotional aspects of the business negotiations.

[0741] The "means for generating a to-do list by analyzing text data that includes emotion recognition results" is a function in which the server extracts tasks and homework from text data that reflects emotion recognition results, and generates a to-do list prioritized based on emotions.

[0742] The "means for generating schedules by analyzing text data including emotion recognition results" is a function in which the server automatically generates the next meeting or important schedule based on text data that reflects emotion recognition results.

[0743] The "means for analyzing text data including emotion recognition results and generating a draft of an e-mail" is a function in which the server automatically generates a draft of a thank-you e-mail including an appropriate greeting and words of gratitude based on the text data in which the emotion recognition results are reflected.

[0744] "Means for transmitting generated data to the user terminal" refers to a function for transmitting the minutes, to-do list, schedule, and draft thank-you email generated by the server to the user terminal.

[0745] MODE FOR CARRYING OUT THE INVENTION

[0746] This invention is an advanced sales support system incorporating an emotion engine that recognizes user emotions, and is intended to reduce the workload of sales representatives after customer visits. Specifically, it converts voice data into high-quality text data, analyzes that text data, and automatically generates drafts of meeting minutes, to-do lists, schedules, and thank-you emails. In addition, by adding emotion analysis using the emotion engine, it provides more flexible and personalized output.

[0747] When visiting a customer, the user records the details of the business meeting using a device such as a smartphone or tablet. Once the recording is complete, the user uploads the recording data from the device to the server. During this process, the server receives the recording data and prepares it for processing.

[0748] The server converts the received recording data into high-quality Japanese text. To do this, the server uses a speech recognition engine (e.g., Google Speech-to-Text API). The converted text data is stored on the server and used in the next step.

[0749] The server inputs the saved text data into an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotions. The emotion engine extracts emotional elements such as joy, sadness, and surprise from the text and reflects the results in the generative AI model.

[0750] When text data containing sentiment analysis results is input into a generative AI model (e.g., OpenAI GPT-4), the server generates the following outputs:

[0751] 1. Minutes:

[0752] The text data that reflects the emotion recognition results is analyzed to generate minutes that include the key points and emotional aspects of the business negotiations.

[0753] Example: "The key points of the discussion are..."

[0754] 2. To-Do List:

[0755] Tasks and homework are extracted from text data that reflects emotion recognition results, and a to-do list prioritized based on emotion is generated.

[0756] Example: "As homework for next time..."

[0757] 3. Plan:

[0758] Generate upcoming meetings and important appointments in Google Calendar format based on text data and sentiment analysis.

[0759] By taking emotional information into consideration, we ensure that users can comfortably complete their next appointment.

[0760] Example: "Our next meeting is scheduled for August 10th at 10am."

[0761] 4. Draft of a thank you email:

[0762] Draft a thank you email based on the deal details and sentiment analysis results.

[0763] Example: "Thank you for the business meeting the other day..."

[0764] The generated minutes, to-do lists, schedules, and draft thank-you emails are sent from the server to the user's device. The user receives a notification and can check and edit each piece of data. This significantly reduces the time spent on post-processing after returning to the office, allowing them to focus on their next sales activity.

[0765] Specific examples

[0766] A user visits a customer and conducts a business negotiation. During the negotiation, the contents of the negotiation are recorded on the device. After the negotiation is over, the user presses the "Stop" button to stop recording, and then presses the "Upload" button to send the recorded data to the server.

[0767] The server receives the recorded data and saves it in storage. The received recorded data is converted into high-quality Japanese text using a speech recognition engine (e.g., Google Speech-to-Text API), and the converted text data is also saved in storage. Next, the server inputs the text data into an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotions. The emotion analysis results are then passed to a generative AI model (e.g., OpenAI GPT-4) to generate the following data:

[0768] Minutes: Include emotional information, such as "The key points of the meeting were..."

[0769] To-do lists: Emotionally driven management, such as "Homework items for next time..."

[0770] Google Calendar Appointments: Emotionally-driven schedules, such as "Your next meeting is scheduled for August 10th at 10am."

[0771] Draft of a thank you email: A draft that reflects your feelings, such as "Thank you for the business meeting the other day..."

[0772] The generated data is collected and sent to the user's device, where the user receives a notification, can review and edit the generated data, and use it as needed. This system recognizes and analyzes the user's emotions and generates output that reflects them, thereby providing more personalized sales support and reducing the workload of sales representatives.

[0773] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0774] Step 1:

[0775] When a user visits a customer, the user uses a device such as a smartphone or tablet to record the details of the business meeting. Specifically, when the user presses the record button on the device, the device begins recording audio data. After the business meeting is over, the user presses the "Stop" button to stop recording, and then presses the "Upload" button to send the recorded data to the server. The input is the user's voice, and the output is the audio file sent to the server.

[0776] Step 2:

[0777] The server saves the voice data received from the device in initial storage. Next, the server uses a voice recognition engine (e.g., Google Speech-to-Text API) to convert the voice data into high-quality Japanese text. The specific operation of this process is to pass the voice data to the API and receive the converted text data. The input is an audio file, and the output is the converted text data.

[0778] Step 3:

[0779] The server inputs the converted text data into an emotion engine (e.g., IBM Watson Tone Analyzer) for emotion analysis. The emotion engine analyzes the content of the text data and extracts emotional elements such as joy, sadness, and surprise. The specific operation of this process is for the server to pass the text data to the emotion engine's API and receive the emotion analysis results. The input is text data, and the output is data containing the emotion analysis results.

[0780] Step 4:

[0781] The server inputs the text data including the sentiment analysis results into a generative AI model (e.g., OpenAI GPT-4) to generate various documents. The specific operation at this time is to pass the sentiment analysis results and text data to the generative AI model as prompt sentences, and obtain the following outputs:

[0782] 1. Meeting minutes - Minutes that include the key points and emotional aspects of the business meeting.

[0783] 2. To-Do List - A list that reflects emotion-based task management.

[0784] 3. Appointments - Schedule your next meeting or important appointment.

[0785] 4. Draft of a thank you email - An email containing appropriate greetings and thank you words that reflect the sentiment. The input is text data and sentiment analysis results, and the output is the respective documents.

[0786] Step 5:

[0787] The server sends the generated minutes, to-do list, schedule, and thank-you email drafts to the user's device. Specifically, the server sends each generated document to the device linked to the user's account and notifies them. The input is each generated document, and the output is the transmission and notification to the user's device.

[0788] Step 6:

[0789] The user can review and edit the drafts of meeting minutes, to-do lists, schedules, and thank-you emails generated on their device and use them as needed. Specifically, the user can open the application on their device and view and edit the contents of each document. The input is each document sent from the server, and the output is the final document edited by the user.

[0790] (Application example 2)

[0791] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0792] Conventional sales support systems and customer service systems have difficulty responding flexibly based on customer emotions, and have not been able to fully improve customer satisfaction or sales efficiency. In particular, there is no system that can analyze customer emotions in real time and provide appropriate responses to customer service staff based on that analysis, so there is a need for more advanced customer service support.

[0793] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring voice data, means for converting the acquired voice data into text data, means for analyzing the converted text data using an emotion engine, means for generating minutes using a generative AI model including the analyzed text data and emotion analysis results, means for generating a to-do list using the generative AI model, means for generating a schedule using the generative AI model, means for generating a draft email using the generative AI model, means for displaying the generated data and emotion analysis results on a user interface in real time, and means for transmitting the generated data to a user terminal. This makes it possible to analyze customer emotions in real time, present personalized countermeasures based on the results, and efficiently and quickly process data after customer service.

[0794] "Audio data" refers to audio signals recorded in digital format, and is information that can be converted into text data and analyzed.

[0795] "Text data" refers to written information converted from audio data and used as input for analytical and generative AI models.

[0796] An "emotion engine" is a software technology for extracting emotional elements from text data and analyzing emotions.

[0797] A "generative AI model" is an artificial intelligence technology that uses machine learning algorithms to automatically generate output such as minutes, to-do lists, schedules, and email drafts from input data.

[0798] "User interface" refers to the screen or operation panel that allows users to operate the system, and which displays sentiment analysis results and suggestions in real time.

[0799] A "user terminal" is a device that is directly operated by a user, and includes smartphones, tablets, smart glasses, and the like.

[0800] "Real-time" means that data acquisition, processing, and output are all done instantly, and information is provided without delay.

[0801] The present invention is a system for efficiently supporting customer service, and is implemented in the following procedure.

[0802] Program Generation

[0803] Hardware and software used

[0804] The system uses the following hardware and software:

[0805] Smart glasses (device example: general noun)

[0806] Speech recognition engine (software example: Google Cloud Speech-to-Text)

[0807] Sentiment analysis engine (software example: IBM Watson Tone Analyzer)

[0808] Generative AI model (software example: OpenAI GPT-3)

[0809] Cloud storage (software example: general noun)

[0810] Acquiring and uploading audio data

[0811] The user puts on the smart glasses and starts talking to the customer. The smart glasses record the conversation in real time and upload the audio data to the server.

[0812] Converting audio data to text

[0813] The server uses a speech recognition engine to convert the received voice data into high-quality text data, which is then stored in cloud storage.

[0814] Analysis by emotion engine

[0815] The server inputs the converted text data into a sentiment analysis engine to analyze customer sentiment in real time, and the analysis results are fed back to the smart glasses.

[0816] Data generation using generative AI models

[0817] Based on the analyzed text data and sentiment analysis results, the generative AI model generates the following data:

[0818] 1. Meeting minutes: Automatically generate meeting minutes that reflect the key points and sentiment of the conversation.

[0819] 2. To-Do List: Extract upcoming tasks and generate a prioritized to-do list.

[0820] 3. Schedule: Generate upcoming meetings and important events in Google Calendar format.

[0821] 4. Email Draft: Generate an emotionally relevant thank you email draft.

[0822] Real-time display and feedback

[0823] The generated data and sentiment analysis results are displayed in real time on the smart glasses' user interface, allowing users to immediately identify appropriate countermeasures. After the analysis is complete, the generated data is sent to the user's device for review and editing.

[0824] Specific examples

[0825] Customer Visit Scenario

[0826] A user visits a customer, puts on smart glasses, and begins a conversation. The smart glasses record the voice and upload it to the server in real time. The server converts the voice data into text and analyzes it using a sentiment analysis engine. The analysis results are displayed on the smart glasses' user interface, and the user can take appropriate action. The generative AI model then generates drafts of meeting minutes, to-do lists, schedules, and thank-you emails, which are saved in cloud storage.

[0827] Prompt Sentence Examples

[0828] Conversation:

[0829] "Hi, can you tell me more about this product? Is it really easy to use?"

[0830] Sentiment analysis results:

[0831] "The customer is interested. Please provide a detailed product description and use case."

[0832] Generated transcript:

[0833] "Service: The customer is interested in Product A and asks about its ease of use. A detailed explanation and examples of use are provided."

[0834] Generated thank you email draft:

[0835] "Thank you very much for visiting our store the other day. We received your question about Product A and we explained the details to you. If there is anything else you would like to ask, please feel free to contact us."

[0836] In this embodiment, the present invention can analyze customer emotions in real time, provide personalized countermeasures based on the analysis, and process data after the customer service quickly and efficiently.

[0837] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0838] Step 1:

[0839] The user puts on the smart glasses and starts a conversation with a customer. The smart glasses record the conversation in real time and capture the audio data. The captured audio data is then uploaded from the smart glasses to a server.

[0840] Input: Real-time audio data

[0841] Output: Uploaded audio data

[0842] Step 2:

[0843] The server inputs the uploaded voice data into a speech recognition engine (e.g., Google Cloud Speech-to-Text) and converts it into high-quality text data. The converted text data is then stored in the server's internal storage.

[0844] Input: Audio data

[0845] Output: Text data

[0846] Step 3:

[0847] The server inputs the converted text data into an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze emotions in real time. The analysis results are added to the text data and saved in storage as emotion analysis results.

[0848] Input: Text data

[0849] Output: Text data with sentiment analysis results

[0850] Step 4:

[0851] The server inputs the text data with sentiment analysis results into a generative AI model (e.g., OpenAI GPT-3) to generate the following data:

[0852] 1. Minutes

[0853] 2. To-Do List

[0854] 3. Schedule

[0855] 4. Draft the email

[0856] Each generated data is stored separately in storage.

[0857] Input: Text data with sentiment analysis results

[0858] Output: minutes, to-do lists, schedules, email drafts

[0859] Step 5:

[0860] The server provides feedback to the user by displaying the generated data and the emotion analysis results in real time on the user interface (smart glasses). The user can then check this information in real time and take appropriate action.

[0861] Input: Generated data, sentiment analysis results

[0862] Output: Real-time display on smart glasses

[0863] Step 6:

[0864] After the customer service is over, the server sends the generated data to the user's device. The user can check the data on the device and edit or add to it as needed. The data is also saved in cloud storage.

[0865] Input: Generated data

[0866] Output: Data sent to user devices, data saved to cloud storage

[0867] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0868] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0869] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0870] [Third embodiment]

[0871] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0872] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0873] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0874] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0875] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0876] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0877] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0878] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0879] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0880] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0881] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0882] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0883] This invention is a system to reduce the workload of sales representatives after customer visits. Specifically, it converts voice data into high-quality text data, and then uses a generative AI model to automatically generate meeting minutes, to-do lists, schedules, and thank-you emails.

[0884] Explanation of program processing

[0885] Acquiring and uploading audio data

[0886] When visiting a customer, the user records the details of the business meeting using a device such as a smartphone or tablet. Once the recording is complete, the user uploads the recorded data from the device to the server. The server then receives the recorded data and prepares it for the next process.

[0887] Japanese text

[0888] The server converts the received recording data into high-quality Japanese text using a speech recognition engine. The converted text data is stored in the server and used in the next step.

[0889] Processing of generated AI

[0890] Once the text data is created, the server uses a generative AI model to generate the following outputs:

[0891] 1. Generate transcripts:

[0892] The server analyzes the text data, extracts the key points of the sales negotiation, and generates minutes in bullet-point format. The minutes summarize the important points of the sales negotiation, which are useful for sales representatives to review later.

[0893] 2. Create a to-do list:

[0894] Tasks and homework are extracted from the text data and the server creates a to-do list, which lists the tasks that need to be completed before the next business meeting.

[0895] 3. Create an appointment:

[0896] By parsing the date and time information in the text data, the server generates a Google Calendar of upcoming meetings and important events, including the date and time of your next client visit and important deadlines.

[0897] 4. Draft thank you email:

[0898] The server creates a draft of a thank-you email based on the details of the sales meeting, allowing the sales representative to send a thank-you email quickly after the meeting.

[0899] feedback

[0900] The generated minutes, to-do lists, schedules, and draft thank-you emails are sent from the server to the user's device. The user receives a notification and can check and edit each piece of data. This significantly reduces the time spent on post-processing after returning to the office, allowing them to focus on their next sales activity.

[0901] Specific examples

[0902] Customer Visit Scenario

[0903] A user visits a customer and conducts a business negotiation. During the negotiation, the contents of the negotiation are recorded on the device. After the negotiation is over, the user presses the "Stop" button to stop recording, and then presses the "Upload" button to send the recorded data to the server.

[0904] Specific operations of server-side processing

[0905] 1. The server receives the recording data and saves it in storage.

[0906] 2. The server uses a speech recognition engine to convert the recorded data into high-quality Japanese text, and stores the converted text in storage.

[0907] 3. The server uses the generative AI model to generate the following data:

[0908] Minutes: "Key points of the meeting were..."

[0909] To-do list: "Homework items for next time..."

[0910] Google Calendar events: "Your next meeting is scheduled for August 10th at 10am."

[0911] A sample thank you email: "Thank you for the business meeting the other day..."

[0912] 4. The generated data is collected and sent to the user's device. The user receives a notification, checks and edits the generated data, and uses it as needed.

[0913] As described above, the system according to the present invention has the effect of largely automating the post-processing after a customer visit, thereby reducing the burden on sales staff.

[0914] The processing flow will be explained below.

[0915] Step 1:

[0916] When a user visits a customer, they use a device (smartphone or tablet) to record the details of the business meeting. They launch the device's voice recording app and press the "Record" button to start recording. When the business meeting is over, they press the "Stop" button to stop recording.

[0917] Step 2:

[0918] The user uploads the recorded data to the server from the voice recording app on their device. After pressing the "Upload" button on the app and confirming the connection to the server, the recorded data is sent to the server.

[0919] Step 3:

[0920] The server receives the uploaded recording, stores it in storage, and adds it to the conversion queue.

[0921] Step 4:

[0922] The server calls the speech recognition engine to convert the recorded data into high-quality Japanese text. The speech recognition engine analyzes the recorded data and generates Japanese text data. The converted text data is saved in storage.

[0923] Step 5:

[0924] The server inputs the stored Japanese text data into the generative AI model for analysis, analyzes the target text data, and generates various output data.

[0925] Step 6:

[0926] The server analyzes the text data, extracts the main points of the business negotiations, and generates minutes in bullet-point format. The generated minutes data is then saved in storage.

[0927] Step 7:

[0928] The server extracts action items (homework items) from the text data and generates a ToDo list. The generated ToDo list data is saved in storage.

[0929] Step 8:

[0930] The server extracts date and time information from the text data and adds the event to Google Calendar. It also calls the Google Calendar API to register upcoming meetings and important events.

[0931] Step 9:

[0932] The server creates a draft of the thank-you email based on the details of the business negotiation. It analyzes the text data and generates a draft email incorporating appropriate greetings and words of thanks. The generated draft email is then saved in storage.

[0933] Step 10:

[0934] The server sends the generated minutes, to-do list, Google Calendar schedule, and a draft of the thank-you email to the user's device. It also sends a notification to the user's device and provides a download link for the generated data.

[0935] Step 11:

[0936] The user receives a notification and clicks the provided link to download the generated data to their device. The user can provide feedback as needed, check and edit the data, and use it in their next sales activity.

[0937] Through the above processing steps, this system can improve the efficiency of the work that sales representatives perform after visiting customers and reduce their workload.

[0938] Example 1

[0939] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0940] The post-processing work that sales representatives perform after customer visits requires time and effort, hindering efficient sales activities. Furthermore, manual text generation and information organization have limited accuracy, leading to the risk of important information being overlooked or typographical errors. There is a need for a system that can solve these problems, reduce the workload of sales representatives, and enable fast and accurate information processing.

[0941] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0942] In this invention, the server includes means for acquiring voice data, means for converting the acquired voice data into text data, means for analyzing the converted text data to generate minutes, means for analyzing the converted text data to generate a to-do list, means for analyzing the converted text data to generate a schedule, means for analyzing the converted text data to generate a draft of an email, means for transmitting the generated data to a user terminal, means for converting the recorded data into Japanese text using a voice recognition engine, and means for generating various information from the text data using a generative AI model. This allows a user to simply record the contents of a business meeting and upload the data to the server, which automatically generates drafts of minutes, to-do lists, schedules, and thank-you emails, enabling efficient and accurate post-event processing.

[0943] "Voice data" refers to data that has been saved in digital format containing the content of conversations and dialogues recorded during customer visits, business negotiations, etc.

[0944] "Means of capture" refers to the combination of devices and software used by a user to record audio data and upload it to a server.

[0945] "Text data" refers to data obtained by converting acquired voice data into text format.

[0946] "Means for converting" refers collectively to a speech recognition engine and related software used to convert voice data into text data.

[0947] "Means of analysis" is a general term for the algorithms and generative AI models used to extract and organize key information from the converted text data.

[0948] Minutes are documents that summarize the main points of business negotiations, meetings, etc., and are used to confirm the contents later.

[0949] A "ToDo list" is a list of tasks or homework that need to be completed by the next time.

[0950] "Schedule" refers to schedule information such as the next date and time of a business meeting or meeting, or important deadlines.

[0951] An "email draft" is a draft of a greeting or thank you email to be used after a business meeting.

[0952] "Generative Means" refers collectively to generative AI models and related algorithms used to create meeting minutes, to-do lists, schedules, and email drafts from text data.

[0953] "Transmission means" is a general term for communication means and protocols for transmitting generated data to a user terminal.

[0954] This invention is a system that uses advanced voice data to automate the post-processing process after sales representatives visit customers. Specifically, it converts voice data into Japanese text data, and then uses a generative AI model based on that text data to automatically generate meeting minutes, to-do lists, schedules, and thank-you emails. This system is primarily comprised of components that play the roles of a server, terminal, and user.

[0955] Hardware and Software Overview

[0956] Terminal

[0957] Users use devices such as smartphones and tablets, which have an application with a recording function installed, allowing them to record the contents of business negotiations.

[0958] server

[0959] The server is the main device that receives, stores, analyzes, and distributes the generated audio data. The following software is installed on the server:

[0960] Speech Recognition Engine: Speech recognition software such as the Google Cloud Speech-to-Text API.

[0961] Generative AI models: Advanced generative AI software such as GPT-3.

[0962] User

[0963] The user is a sales representative who records voice data during a sales meeting and uploads the data to the server after the recording is complete. Post-processing is performed based on the generated information.

[0964] Data processing and calculation flow

[0965] 1. Acquiring and uploading audio data

[0966] When a user visits a customer, the user uses the device to record the details of the business meeting. After completing the recording, the user uploads the audio data from the device to the server. The server receives the recorded data via an HTTP request and stores it in storage.

[0967] 2. Japanese text

[0968] The server passes the received recorded data to a speech recognition engine and converts it into high-quality Japanese text data, which is then saved back into storage.

[0969] 3. Text Data Analysis and Generation

[0970] The server inputs the stored text data into a generative AI model to generate the following information:

[0971] Minutes: Summarize the main points of the business meeting in bullet points.

[0972] To-Do List: Make a list of tasks and homework that need to be done by the next time.

[0973] Appointments: Generate appointments in Google Calendar format based on the dates and times mentioned in the deal.

[0974] Draft a thank you email: Create a draft of a thank you email based on the details of the business meeting.

[0975] 4. Data distribution and feedback

[0976] The server sends the generated minutes, to-do lists, schedules, and thank-you email drafts to the user's device. The user receives a notification, checks and edits the generated data, and uses it as needed.

[0977] Examples and prompts

[0978] Specific examples

[0979] A user visits a customer and conducts a business meeting. The contents of the business meeting are recorded on the device, and after the business meeting is completed, the recording is stopped and the recorded data is uploaded to the server.

[0980] The server receives the recording, converts it into Japanese text using a speech recognition engine, and then uses a generative AI model to generate drafts of meeting minutes, to-do lists, schedules, and thank-you emails, which are then sent to the user's device.

[0981] Prompt Sentence Examples

[0982] By inputting the following prompt sentence into the generative AI model, we can get the corresponding output:

[0983] Prompt for generating transcripts:

[0984] "Please extract the key points from the following sales pitch in bullet points format:

[0985] 1. Customer needs

[0986] 2. Proposal details

[0987] 3. Questions and Answers

[0988] 4. Next assignment

[0989] Prompt for generating a ToDo list:

[0990] From the text below, please list the tasks you need to complete by next time:

[0991] Summary of business negotiations

[0992] Confirm next step

[0993] Preparing for follow-up

[0994] Prompt for generating an appointment:

[0995] "Extract important dates and times from the text below and generate a Google Calendar event."

[0996] Prompt for generating a draft thank you email:

[0997] "Please draft a thank you email based on the following business details:

[0998] Thank you message

[0999] Review the main points of the deal

[1000] Proposal for next steps

[1001] As a result, this system significantly improves the work efficiency of sales staff and reduces the time and effort required for post-sales processing.

[1002] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1003] Step 1:

[1004] Acquiring and uploading audio data

[1005] The user uses a device such as a smartphone or tablet to record the contents of a business meeting. They open a recording application and press the record button to record the contents of the business meeting. The input data is the audio spoken during the business meeting. After the business meeting is over, the user presses the "Stop" button to end recording, and then presses the "Upload" button to send the recorded data to the server. The server receives the HTTP request and saves the recorded data in storage. The output data is an audio file saved in storage.

[1006] Step 2:

[1007] Japanese text

[1008] The server passes the saved recording data to a speech recognition engine. The input data is an audio file. The server uses a speech recognition engine such as the Google Cloud Speech-to-Text API to convert this audio data into high-quality Japanese text data. Specifically, the speech recognition engine analyzes the audio waveform and converts it into text information. The converted text data is then saved back into storage within the server. The output data is Japanese text data.

[1009] Step 3:

[1010] Text data analysis and generation

[1011] The server inputs text data into the generative AI model to generate specific information. The input data is Japanese text data. The server uses the following prompt to have the generative AI model (e.g., GPT-3) analyze the data:

[1012] Prompt for generating transcripts:

[1013] "Please extract the key points from the following sales pitch in bullet points format:

[1014] 1. Customer needs

[1015] 2. Proposal details

[1016] 3. Questions and Answers

[1017] 4. Next assignment

[1018] Prompt for generating a ToDo list:

[1019] From the text below, please list the tasks you need to complete by next time:

[1020] Summary of business negotiations

[1021] Confirm next step

[1022] Preparing for follow-up

[1023] Prompt for generating an appointment:

[1024] "Extract important dates and times from the text below and generate a Google Calendar event."

[1025] Prompt for generating a draft thank you email:

[1026] "Please draft a thank you email based on the following business details:

[1027] Thank you message

[1028] Review the main points of the deal

[1029] Proposal for next steps

[1030] The server uses these prompts to have the generative AI model analyze the text data and generate information for each purpose. The generated information becomes the basis for meeting minutes, to-do lists, schedules, and thank-you emails, all of which are stored in the server's storage. The output data is the various types of generated information.

[1031] Step 4:

[1032] Data distribution and feedback

[1033] The server sends the generated data (minutes, to-do list, schedule, thank-you email draft) to the user's device. The input data is the various types of generated information. The server uses an HTTP response to send the generated data to the user's device. The user receives a notification on their device and checks and edits the generated data. Specifically, the user clicks the notification to open the application, checks the generated minutes, to-do list, schedule, and thank-you email, and edits them as necessary. The output data is the various types of information after it has been checked and edited by the user.

[1034] By following these steps, users can simply upload their voice data to the server, and minutes, to-do lists, schedules, and thank-you email drafts will be automatically generated and provided to them. This series of processes will greatly improve the efficiency of post-sales processing work for sales representatives.

[1035] (Application example 1)

[1036] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1037] During maintenance work in a factory, recording the details of the work and any problems that arise, and then creating work reports and to-do lists accurately and efficiently based on that information, places a heavy burden on workers. This creates the problem of time being spent creating reports and checking schedules, which reduces the efficiency of the work. Furthermore, mistakes are likely to occur in work reports and schedule settings, which can have an impact on factory operations. To solve these problems, an automated system is needed.

[1038] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1039] In this invention, the server includes means for acquiring voice data, means for converting the acquired voice data into text data, means for analyzing the converted text data to generate minutes, means for analyzing the converted text data to generate a to-do list, means for analyzing the converted text data to generate a schedule, means for analyzing the converted text data to generate a draft of an e-mail, means for sending the generated data to a user terminal, means for outputting the generated data as an in-factory work report, means for outputting the generated data as a work schedule in calendar format, and means for outputting the generated data as a work report e-mail. This automates in-factory maintenance work reporting and schedule setting, not only reducing the burden on workers but also improving work efficiency and reducing reporting and scheduling errors.

[1040] "Voice data" refers to audio information that records events, problems, and work conditions that occur during factory work.

[1041] "Text data" refers to data obtained by converting voice data into character information.

[1042] "Analysis" is the process of analyzing the converted text data and generating reports, lists, schedules, etc. based on it.

[1043] Minutes are documents that organize and summarize the work content, important points, and issues.

[1044] A "ToDo list" is a list in bulleted form that lists the next tasks to be done and items that need to be improved.

[1045] "Schedule" is a calendar listing the next work schedule and maintenance plan.

[1046] An "email draft" is a draft of an email containing a work report or contact information.

[1047] A "user terminal" is an electronic device such as a smartphone or tablet used by a worker.

[1048] The "Factory Work Report" is a report summarizing the details of the maintenance work carried out within the factory.

[1049] "Work schedule" is schedule information including the next maintenance work and check items.

[1050] A "work report email" is an email containing the report content to be sent to the relevant parties after the maintenance work has been completed.

[1051] "Cloud storage" is a service for storing data over the Internet.

[1052] This invention is a system for improving the efficiency of maintenance work in factories. Specifically, it converts voice data during work into text data, and then uses a generative AI model based on that text data to automatically generate drafts of work reports, to-do lists, work schedules, and thank-you emails. This system is implemented as follows.

[1053] First, the user, a worker, uses a device such as a smartphone or tablet to record the details of the maintenance work in the factory by voice. The hardware used for this is a smartphone or tablet. The recorded data is managed using a cloud server.

[1054] Once the audio data is acquired, the user uploads the recording to the cloud server via their device, and the cloud server stores the received recording data in its storage.

[1055] The server then uses a cloud-based voice recognition engine (e.g., Google Cloud Speech-to-Text) to convert the recorded data into high-quality text data, which is then stored in cloud storage.

[1056] The generated text data is used with a generative AI model (e.g., OpenAI GPT-4) to produce the following output:

[1057] 1. Generation of work reports: Analyze text data, extract key points of maintenance work, and generate reports.

[1058] 2. Generate a to-do list: Extract next tasks and areas for improvement from text data and create a to-do list.

[1059] 3. Generate work schedule: Analyze the date and time information in the text data and generate the next maintenance schedule in calendar format.

[1060] 4. Generate a thank you email: Draft a thank you email based on the maintenance work.

[1061] This generated data is sent from the server to the user's device. The user receives a notification, can check and edit the generated data, and use it as needed. This significantly reduces the time required for post-processing after maintenance work.

[1062] Examples of illustrative prompts include:

[1063] Generate meeting minutes, to-do lists, appointments, and thank you emails based on:

[1064] "The following problem was discovered during today's maintenance work. There is a small leak in the hydraulic system and the valve needs to be replaced. The next inspection date is October 20th. All other inspection items are normal."

[1065] The system according to the present invention has the effect of significantly reducing the burden on workers and improving work efficiency, and also reduces errors in reporting and schedule setting.

[1066] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1067] Step 1:

[1068] Users can use their smartphones or tablets to record their maintenance work by voice, and the recorded data is saved on the device.

[1069] Input: User voice input

[1070] Output: Recording data saved on the device

[1071] Specific operation: The user presses the start recording button, talks about the maintenance work and problems, and then presses the stop button to end the recording.

[1072] Step 2:

[1073] Users upload recorded data from their devices to the cloud server, which receives the data and stores it in its storage.

[1074] Input: Recording data stored on the device

[1075] Output: Recording data stored on a cloud server

[1076] Specific operation: The user presses the upload button to send the recorded data from the device to the cloud server. The server receives the recorded data and stores it in a database.

[1077] Step 3:

[1078] The server converts the recorded data into text data using a speech recognition engine, and stores the converted text data in cloud storage.

[1079] Input: Recording data stored on a cloud server

[1080] Output: Text data stored in cloud storage

[1081] Specific operation: The server uses a speech recognition engine such as Google Cloud Speech-to-Text to analyze the recorded data and convert it into high-quality Japanese text. The converted text data is then stored in cloud storage.

[1082] Step 4:

[1083] The server uses a generative AI model to generate the following outputs based on the text data:

[1084] business report

[1085] To-Do List

[1086] Work Schedule

[1087] Work report email

[1088] Input: Text data stored in cloud storage

[1089] Output: Generated work reports, ToDo lists, work schedules, work report emails

[1090] Specific operation: The server inputs text data into a generative AI model such as OpenAI GPT-4, and generates prompt sentences for minutes, to-do lists, schedules, and thank-you emails.

[1091] Step 5:

[1092] The server sends the generated data to the user's terminal, where the user receives the data and checks and edits it.

[1093] Input: Generated work reports, ToDo lists, work schedules, work report emails

[1094] Output: Data sent to the user's terminal

[1095] Specific operation: The server sends the generated data to the user's terminal. The user receives a notification, checks it, and edits it if necessary.

[1096] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1097] This invention is an advanced sales support system incorporating an emotion engine that recognizes user emotions, and is intended to reduce the workload of sales representatives after customer visits. Specifically, it converts voice data into high-quality text data, analyzes that text data, and automatically generates drafts of meeting minutes, to-do lists, schedules, and thank-you emails. In addition, by adding emotion analysis using the emotion engine, it provides more flexible and personalized output.

[1098] Explanation of program processing

[1099] Acquiring and uploading audio data

[1100] When a user visits a customer, the user records the details of the business meeting using a device such as a smartphone or tablet. Once the recording is complete, the user uploads the recording data from the device to the server. During this process, the server receives the recording data and prepares it for processing.

[1101] Japanese text

[1102] The server converts the received recording data into high-quality Japanese text using a speech recognition engine. The converted text data is stored in the server and used in the next step.

[1103] Analysis by emotion engine

[1104] The server inputs the text data into the emotion engine to analyze the user's emotions. The emotion engine extracts emotional elements (happiness, sadness, surprise, etc.) from the text and reflects the results in the next generation AI process.

[1105] Processing of generated AI

[1106] When text data containing sentiment analysis results is input into the generative AI, the server generates the following outputs:

[1107] 1. Generate transcripts:

[1108] The server analyzes the text data and generates minutes that reflect the key points of the business meeting and the user's emotions, allowing the system to record not only the important points of the business meeting but also the emotional aspects of the discussion.

[1109] 2. Create a to-do list:

[1110] It extracts tasks and homework from text data and creates a to-do list based on the user's emotions, allowing the user to manage task priorities so that they do not feel stressed.

[1111] 3. Create an appointment:

[1112] Based on text data and sentiment analysis, it generates upcoming meetings and important appointments in Google Calendar format. By taking into account sentiment information, it helps users to comfortably complete their next appointment.

[1113] 4. Draft thank you email:

[1114] The server creates a draft thank-you email based on the details of the business and the results of sentiment analysis, allowing it to generate an email that incorporates emotionally appropriate greetings and words of thanks.

[1115] feedback

[1116] The generated minutes, to-do lists, schedules, and draft thank-you emails are sent from the server to the user's device. The user receives a notification and can check and edit each piece of data. This significantly reduces the time spent on post-processing after returning to the office, allowing them to focus on their next sales activity.

[1117] Specific examples

[1118] Customer Visit Scenario

[1119] A user visits a customer and conducts a business negotiation. During the negotiation, the contents of the negotiation are recorded on the device. After the negotiation is over, the user presses the "Stop" button to stop recording, and then presses the "Upload" button to send the recorded data to the server.

[1120] Specific operations of server-side processing

[1121] 1. The server receives the recording data and saves it in storage.

[1122] 2. The server uses a speech recognition engine to convert the recorded data into high-quality Japanese text, and stores the converted text in storage.

[1123] 3. The server inputs the text data into the emotion engine to analyze the user's emotions, and passes the emotion analysis results to the generative AI model.

[1124] 4. The server uses the generative AI model to generate the following data:

[1125] Minutes: Include emotional information, such as "The key points of the meeting were..."

[1126] To-do lists: Emotionally driven management, such as "Homework items for next time..."

[1127] Google Calendar Appointments: Emotionally-driven schedules, such as "Your next meeting is scheduled for August 10th at 10am."

[1128] Draft of a thank you email: A draft that reflects your feelings, such as "Thank you for the business meeting the other day..."

[1129] 5. The generated data is collected and sent to the user's device. The user receives a notification, checks and edits the generated data, and uses it as needed.

[1130] In this way, the system of the present invention recognizes and analyzes the user's emotions and generates output that reflects them, thereby providing more personalized sales support and reducing the workload of sales representatives.

[1131] The processing flow will be explained below.

[1132] Step 1:

[1133] When a user visits a customer, they use a device (smartphone or tablet) to record the details of the business meeting. They launch the device's voice recording app and press the "Record" button to start recording. When the business meeting is over, they press the "Stop" button to stop recording.

[1134] Step 2:

[1135] The user uploads the recorded data to the server from the voice recording app on their device. After pressing the "Upload" button on the app and confirming the connection to the server, the recorded data is sent to the server.

[1136] Step 3:

[1137] The server receives the uploaded recording, stores it in storage, and adds it to the conversion queue.

[1138] Step 4:

[1139] The server calls the speech recognition engine to convert the recorded data into high-quality Japanese text. The speech recognition engine analyzes the recorded data and generates Japanese text data. The converted text data is saved in storage.

[1140] Step 5:

[1141] The server inputs the stored Japanese text data into the emotion engine to analyze the user's emotions. The emotion engine analyzes the text data and extracts emotional elements (joy, sadness, surprise, etc.). The emotion analysis results are passed to the generative AI model.

[1142] Step 6:

[1143] The server generates meeting minutes using the generative AI model. Taking into account the results of sentiment analysis, it extracts key points from the text data and creates minutes in bullet point format. The minutes also reflect emotional comments and important tones. The generated minutes data is saved in storage.

[1144] Step 7:

[1145] The server generates a to-do list using the generative AI model. Based on the results of sentiment analysis, action items (homework items) are extracted from the text data. A to-do list is created that takes into account the user's emotional state, task priority, and workload. The generated to-do list data is saved in storage.

[1146] Step 8:

[1147] The server generates schedules using a generative AI model. Taking into account the results of sentiment analysis, it extracts date and time information from the text data and creates upcoming meetings and important appointments in Google Calendar format. Based on the emotional information, it adjusts the schedule so that the user can complete it comfortably. The schedule data is automatically registered in Google Calendar.

[1148] Step 9:

[1149] The server uses the generative AI model to create a draft of a thank-you email. Based on the results of the sentiment analysis, it creates a draft email that includes a greeting and expression of gratitude appropriate to the business negotiation. It uses an appropriate tone and expressions depending on the emotional information. The generated email draft data is saved in storage.

[1150] Step 10:

[1151] The server sends the generated minutes, to-do list, Google Calendar schedule, and a draft of the thank-you email to the user's device. The server also sends a notification to the user's device and provides a download link for the generated data.

[1152] Step 11:

[1153] The user receives a notification and clicks the provided link to download the generated data to their device. The user can then review and edit the data as needed and use it for their next sales activity.

[1154] Through the above processing steps, the system can recognize the user's emotions and generate output that reflects them, thereby providing more personalized sales support and reducing the workload of sales representatives.

[1155] Example 2

[1156] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1157] Modern sales activities require a wide range of post-event processing, such as creating reports after customer visits, organizing tasks, setting up next-time appointments, and writing follow-up emails. These tasks require time and effort, placing a significant burden on sales representatives. Furthermore, accurate recording of customer interactions and conveying information that incorporates emotions and nuances is required, but doing this manually is difficult. There is a need for the development of a system that can solve these issues, reduce the workload of sales representatives, and provide efficient, personalized sales support.

[1158] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for acquiring voice data, means for converting the acquired voice data into text data, means for analyzing the converted text data to recognize emotions, means for analyzing the text data including the emotion recognition results to generate minutes, means for analyzing the text data including the emotion recognition results to generate a to-do list, means for analyzing the text data including the emotion recognition results to generate a schedule, means for analyzing the text data including the emotion recognition results to generate a draft of an e-mail, and means for transmitting the generated data to a user terminal. This allows the content of a conversation with a customer to be accurately recorded, information including emotions and nuances to be automatically generated, and sales representatives to quickly and efficiently perform follow-up processing.

[1159] The "means for acquiring voice data" is a function for recording the contents of business negotiations via the terminal used by the user when visiting a customer, and transmitting the recorded data to the server.

[1160] The "means for converting acquired voice data into text data" is a function in which the server converts the received voice data into high-quality text format using a voice recognition engine.

[1161] The "means for analyzing the converted text data and recognizing emotions" is a function by which the server inputs the text data into an emotion analysis engine and extracts and recognizes the emotional elements contained in the text data.

[1162] The "means for generating minutes by analyzing text data including emotion recognition results" is a function in which the server analyzes text data that reflects emotion recognition results and automatically generates minutes that include the main points and emotional aspects of the business negotiations.

[1163] The "means for generating a to-do list by analyzing text data that includes emotion recognition results" is a function in which the server extracts tasks and homework from text data that reflects emotion recognition results, and generates a to-do list prioritized based on emotions.

[1164] The "means for generating schedules by analyzing text data including emotion recognition results" is a function in which the server automatically generates the next meeting or important schedule based on text data that reflects emotion recognition results.

[1165] The "means for analyzing text data including emotion recognition results and generating a draft of an e-mail" is a function in which the server automatically generates a draft of a thank-you e-mail including an appropriate greeting and words of gratitude based on the text data in which the emotion recognition results are reflected.

[1166] "Means for transmitting generated data to the user terminal" refers to a function for transmitting the minutes, to-do list, schedule, and draft thank-you email generated by the server to the user terminal.

[1167] MODE FOR CARRYING OUT THE INVENTION

[1168] This invention is an advanced sales support system incorporating an emotion engine that recognizes user emotions, and is intended to reduce the workload of sales representatives after customer visits. Specifically, it converts voice data into high-quality text data, analyzes that text data, and automatically generates drafts of meeting minutes, to-do lists, schedules, and thank-you emails. In addition, by adding emotion analysis using the emotion engine, it provides more flexible and personalized output.

[1169] When visiting a customer, the user records the details of the business meeting using a device such as a smartphone or tablet. Once the recording is complete, the user uploads the recording data from the device to the server. During this process, the server receives the recording data and prepares it for processing.

[1170] The server converts the received recording data into high-quality Japanese text. To do this, the server uses a speech recognition engine (e.g., Google Speech-to-Text API). The converted text data is stored on the server and used in the next step.

[1171] The server inputs the saved text data into an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotions. The emotion engine extracts emotional elements such as joy, sadness, and surprise from the text and reflects the results in the generative AI model.

[1172] When text data containing sentiment analysis results is input into a generative AI model (e.g., OpenAI GPT-4), the server generates the following outputs:

[1173] 1. Minutes:

[1174] The text data that reflects the emotion recognition results is analyzed to generate minutes that include the key points and emotional aspects of the business negotiations.

[1175] Example: "The key points of the discussion are..."

[1176] 2. To-Do List:

[1177] Tasks and homework are extracted from text data that reflects emotion recognition results, and a to-do list prioritized based on emotion is generated.

[1178] Example: "As homework for next time..."

[1179] 3. Plan:

[1180] Generate upcoming meetings and important appointments in Google Calendar format based on text data and sentiment analysis.

[1181] By taking emotional information into consideration, we ensure that users can comfortably complete their next appointment.

[1182] Example: "Our next meeting is scheduled for August 10th at 10am."

[1183] 4. Draft of a thank you email:

[1184] Draft a thank you email based on the deal details and sentiment analysis results.

[1185] Example: "Thank you for the business meeting the other day..."

[1186] The generated minutes, to-do lists, schedules, and draft thank-you emails are sent from the server to the user's device. The user receives a notification and can check and edit each piece of data. This significantly reduces the time spent on post-processing after returning to the office, allowing them to focus on their next sales activity.

[1187] Specific examples

[1188] A user visits a customer and conducts a business negotiation. During the negotiation, the contents of the negotiation are recorded on the device. After the negotiation is over, the user presses the "Stop" button to stop recording, and then presses the "Upload" button to send the recorded data to the server.

[1189] The server receives the recorded data and saves it in storage. The received recorded data is converted into high-quality Japanese text using a speech recognition engine (e.g., Google Speech-to-Text API), and the converted text data is also saved in storage. Next, the server inputs the text data into an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotions. The emotion analysis results are then passed to a generative AI model (e.g., OpenAI GPT-4) to generate the following data:

[1190] Minutes: Include emotional information, such as "The key points of the meeting were..."

[1191] To-do lists: Emotionally driven management, such as "Homework items for next time..."

[1192] Google Calendar Appointments: Emotionally-driven schedules, such as "Your next meeting is scheduled for August 10th at 10am."

[1193] Draft of a thank you email: A draft that reflects your feelings, such as "Thank you for the business meeting the other day..."

[1194] The generated data is collected and sent to the user's device, where the user receives a notification, can review and edit the generated data, and use it as needed. This system recognizes and analyzes the user's emotions and generates output that reflects them, thereby providing more personalized sales support and reducing the workload of sales representatives.

[1195] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1196] Step 1:

[1197] When a user visits a customer, the user uses a device such as a smartphone or tablet to record the details of the business meeting. Specifically, when the user presses the record button on the device, the device begins recording audio data. After the business meeting is over, the user presses the "Stop" button to stop recording, and then presses the "Upload" button to send the recorded data to the server. The input is the user's voice, and the output is the audio file sent to the server.

[1198] Step 2:

[1199] The server saves the voice data received from the device in initial storage. Next, the server uses a voice recognition engine (e.g., Google Speech-to-Text API) to convert the voice data into high-quality Japanese text. The specific operation of this process is to pass the voice data to the API and receive the converted text data. The input is an audio file, and the output is the converted text data.

[1200] Step 3:

[1201] The server inputs the converted text data into an emotion engine (e.g., IBM Watson Tone Analyzer) for emotion analysis. The emotion engine analyzes the content of the text data and extracts emotional elements such as joy, sadness, and surprise. The specific operation of this process is for the server to pass the text data to the emotion engine's API and receive the emotion analysis results. The input is text data, and the output is data containing the emotion analysis results.

[1202] Step 4:

[1203] The server inputs the text data including the sentiment analysis results into a generative AI model (e.g., OpenAI GPT-4) to generate various documents. The specific operation at this time is to pass the sentiment analysis results and text data to the generative AI model as prompt sentences, and obtain the following outputs:

[1204] 1. Meeting minutes - Minutes that include the key points and emotional aspects of the business meeting.

[1205] 2. To-Do List - A list that reflects emotion-based task management.

[1206] 3. Appointments - Schedule your next meeting or important appointment.

[1207] 4. Draft of a thank you email - An email containing appropriate greetings and thank you words that reflect the sentiment. The input is text data and sentiment analysis results, and the output is the respective documents.

[1208] Step 5:

[1209] The server sends the generated minutes, to-do list, schedule, and thank-you email drafts to the user's device. Specifically, the server sends each generated document to the device linked to the user's account and notifies them. The input is each generated document, and the output is the transmission and notification to the user's device.

[1210] Step 6:

[1211] The user can review and edit the drafts of meeting minutes, to-do lists, schedules, and thank-you emails generated on their device and use them as needed. Specifically, the user can open the application on their device and view and edit the contents of each document. The input is each document sent from the server, and the output is the final document edited by the user.

[1212] (Application example 2)

[1213] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1214] Conventional sales support systems and customer service systems have difficulty responding flexibly based on customer emotions, and have not been able to fully improve customer satisfaction or sales efficiency. In particular, there is no system that can analyze customer emotions in real time and provide appropriate responses to customer service staff based on that analysis, so there is a need for more advanced customer service support.

[1215] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring voice data, means for converting the acquired voice data into text data, means for analyzing the converted text data using an emotion engine, means for generating minutes using a generative AI model including the analyzed text data and emotion analysis results, means for generating a to-do list using the generative AI model, means for generating a schedule using the generative AI model, means for generating a draft email using the generative AI model, means for displaying the generated data and emotion analysis results on a user interface in real time, and means for transmitting the generated data to a user terminal. This makes it possible to analyze customer emotions in real time, present personalized countermeasures based on the results, and efficiently and quickly process data after customer service.

[1216] "Audio data" refers to audio signals recorded in digital format, and is information that can be converted into text data and analyzed.

[1217] "Text data" refers to written information converted from audio data and used as input for analytical and generative AI models.

[1218] An "emotion engine" is a software technology for extracting emotional elements from text data and analyzing emotions.

[1219] A "generative AI model" is an artificial intelligence technology that uses machine learning algorithms to automatically generate output such as minutes, to-do lists, schedules, and email drafts from input data.

[1220] "User interface" refers to the screen or operation panel that allows users to operate the system, and which displays sentiment analysis results and suggestions in real time.

[1221] A "user terminal" is a device that is directly operated by a user, and includes smartphones, tablets, smart glasses, and the like.

[1222] "Real-time" means that data acquisition, processing, and output are all done instantly, and information is provided without delay.

[1223] The present invention is a system for efficiently supporting customer service, and is implemented in the following procedure.

[1224] Program Generation

[1225] Hardware and software used

[1226] The system uses the following hardware and software:

[1227] Smart glasses (device example: general noun)

[1228] Speech recognition engine (software example: Google Cloud Speech-to-Text)

[1229] Sentiment analysis engine (software example: IBM Watson Tone Analyzer)

[1230] Generative AI model (software example: OpenAI GPT-3)

[1231] Cloud storage (software example: general noun)

[1232] Acquiring and uploading audio data

[1233] The user puts on the smart glasses and starts talking to the customer. The smart glasses record the conversation in real time and upload the audio data to the server.

[1234] Converting audio data to text

[1235] The server uses a speech recognition engine to convert the received voice data into high-quality text data, which is then stored in cloud storage.

[1236] Analysis by emotion engine

[1237] The server inputs the converted text data into a sentiment analysis engine to analyze customer sentiment in real time, and the analysis results are fed back to the smart glasses.

[1238] Data generation using generative AI models

[1239] Based on the analyzed text data and sentiment analysis results, the generative AI model generates the following data:

[1240] 1. Meeting minutes: Automatically generate meeting minutes that reflect the key points and sentiment of the conversation.

[1241] 2. To-Do List: Extract upcoming tasks and generate a prioritized to-do list.

[1242] 3. Schedule: Generate upcoming meetings and important events in Google Calendar format.

[1243] 4. Email Draft: Generate an emotionally relevant thank you email draft.

[1244] Real-time display and feedback

[1245] The generated data and sentiment analysis results are displayed in real time on the smart glasses' user interface, allowing users to immediately identify appropriate countermeasures. After the analysis is complete, the generated data is sent to the user's device for review and editing.

[1246] Specific examples

[1247] Customer Visit Scenario

[1248] A user visits a customer, puts on smart glasses, and begins a conversation. The smart glasses record the voice and upload it to the server in real time. The server converts the voice data into text and analyzes it using a sentiment analysis engine. The analysis results are displayed on the smart glasses' user interface, and the user can take appropriate action. The generative AI model then generates drafts of meeting minutes, to-do lists, schedules, and thank-you emails, which are saved in cloud storage.

[1249] Prompt Sentence Examples

[1250] Conversation:

[1251] "Hi, can you tell me more about this product? Is it really easy to use?"

[1252] Sentiment analysis results:

[1253] "The customer is interested. Please provide a detailed product description and use case."

[1254] Generated transcript:

[1255] "Service: The customer is interested in Product A and asks about its ease of use. A detailed explanation and examples of use are provided."

[1256] Generated thank you email draft:

[1257] "Thank you very much for visiting our store the other day. We received your question about Product A and we explained the details to you. If there is anything else you would like to ask, please feel free to contact us."

[1258] In this embodiment, the present invention can analyze customer emotions in real time, provide personalized countermeasures based on the analysis, and process data after the customer service quickly and efficiently.

[1259] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1260] Step 1:

[1261] The user puts on the smart glasses and starts a conversation with a customer. The smart glasses record the conversation in real time and capture the audio data. The captured audio data is then uploaded from the smart glasses to a server.

[1262] Input: Real-time audio data

[1263] Output: Uploaded audio data

[1264] Step 2:

[1265] The server inputs the uploaded voice data into a speech recognition engine (e.g., Google Cloud Speech-to-Text) and converts it into high-quality text data. The converted text data is then stored in the server's internal storage.

[1266] Input: Audio data

[1267] Output: Text data

[1268] Step 3:

[1269] The server inputs the converted text data into an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze emotions in real time. The analysis results are added to the text data and saved in storage as emotion analysis results.

[1270] Input: Text data

[1271] Output: Text data with sentiment analysis results

[1272] Step 4:

[1273] The server inputs the text data with sentiment analysis results into a generative AI model (e.g., OpenAI GPT-3) to generate the following data:

[1274] 1. Minutes

[1275] 2. To-Do List

[1276] 3. Schedule

[1277] 4. Draft the email

[1278] Each generated data is stored separately in storage.

[1279] Input: Text data with sentiment analysis results

[1280] Output: minutes, to-do lists, schedules, email drafts

[1281] Step 5:

[1282] The server provides feedback to the user by displaying the generated data and the emotion analysis results in real time on the user interface (smart glasses). The user can then check this information in real time and take appropriate action.

[1283] Input: Generated data, sentiment analysis results

[1284] Output: Real-time display on smart glasses

[1285] Step 6:

[1286] After the customer service is over, the server sends the generated data to the user's device. The user can check the data on the device and edit or add to it as needed. The data is also saved in cloud storage.

[1287] Input: Generated data

[1288] Output: Data sent to user devices, data saved to cloud storage

[1289] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1290] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1291] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1292] [Fourth embodiment]

[1293] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1294] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1295] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1296] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1297] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1298] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1299] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1300] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1301] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1302] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1303] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1304] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1305] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1306] This invention is a system to reduce the workload of sales representatives after customer visits. Specifically, it converts voice data into high-quality text data, and then uses a generative AI model to automatically generate meeting minutes, to-do lists, schedules, and thank-you emails.

[1307] Explanation of program processing

[1308] Acquiring and uploading audio data

[1309] When visiting a customer, the user records the details of the business meeting using a device such as a smartphone or tablet. Once the recording is complete, the user uploads the recorded data from the device to the server. The server then receives the recorded data and prepares it for the next process.

[1310] Japanese text

[1311] The server converts the received recording data into high-quality Japanese text using a speech recognition engine. The converted text data is stored in the server and used in the next step.

[1312] Processing of generated AI

[1313] Once the text data is created, the server uses a generative AI model to generate the following outputs:

[1314] 1. Generate transcripts:

[1315] The server analyzes the text data, extracts the key points of the sales negotiation, and generates minutes in bullet-point format. The minutes summarize the important points of the sales negotiation, which are useful for sales representatives to review later.

[1316] 2. Create a to-do list:

[1317] Tasks and homework are extracted from the text data and the server creates a to-do list, which lists the tasks that need to be completed before the next business meeting.

[1318] 3. Create an appointment:

[1319] By parsing the date and time information in the text data, the server generates a Google Calendar of upcoming meetings and important events, including the date and time of your next client visit and important deadlines.

[1320] 4. Draft thank you email:

[1321] The server creates a draft of a thank-you email based on the details of the sales meeting, allowing the sales representative to send a thank-you email quickly after the meeting.

[1322] feedback

[1323] The generated minutes, to-do lists, schedules, and draft thank-you emails are sent from the server to the user's device. The user receives a notification and can check and edit each piece of data. This significantly reduces the time spent on post-processing after returning to the office, allowing them to focus on their next sales activity.

[1324] Specific examples

[1325] Customer Visit Scenario

[1326] A user visits a customer and conducts a business negotiation. During the negotiation, the contents of the negotiation are recorded on the device. After the negotiation is over, the user presses the "Stop" button to stop recording, and then presses the "Upload" button to send the recorded data to the server.

[1327] Specific operations of server-side processing

[1328] 1. The server receives the recording data and saves it in storage.

[1329] 2. The server uses a speech recognition engine to convert the recorded data into high-quality Japanese text, and stores the converted text in storage.

[1330] 3. The server uses the generative AI model to generate the following data:

[1331] Minutes: "Key points of the meeting were..."

[1332] To-do list: "Homework items for next time..."

[1333] Google Calendar events: "Your next meeting is scheduled for August 10th at 10am."

[1334] A sample thank you email: "Thank you for the business meeting the other day..."

[1335] 4. The generated data is collected and sent to the user's device. The user receives a notification, checks and edits the generated data, and uses it as needed.

[1336] As described above, the system according to the present invention has the effect of largely automating the post-processing after a customer visit, thereby reducing the burden on sales staff.

[1337] The processing flow will be explained below.

[1338] Step 1:

[1339] When a user visits a customer, they use a device (smartphone or tablet) to record the details of the business meeting. They launch the device's voice recording app and press the "Record" button to start recording. When the business meeting is over, they press the "Stop" button to stop recording.

[1340] Step 2:

[1341] The user uploads the recorded data to the server from the voice recording app on their device. After pressing the "Upload" button on the app and confirming the connection to the server, the recorded data is sent to the server.

[1342] Step 3:

[1343] The server receives the uploaded recording, stores it in storage, and adds it to the conversion queue.

[1344] Step 4:

[1345] The server calls the speech recognition engine to convert the recorded data into high-quality Japanese text. The speech recognition engine analyzes the recorded data and generates Japanese text data. The converted text data is saved in storage.

[1346] Step 5:

[1347] The server inputs the stored Japanese text data into the generative AI model for analysis, analyzes the target text data, and generates various output data.

[1348] Step 6:

[1349] The server analyzes the text data, extracts the main points of the business negotiations, and generates minutes in bullet-point format. The generated minutes data is then saved in storage.

[1350] Step 7:

[1351] The server extracts action items (homework items) from the text data and generates a ToDo list. The generated ToDo list data is saved in storage.

[1352] Step 8:

[1353] The server extracts date and time information from the text data and adds the event to Google Calendar. It also calls the Google Calendar API to register upcoming meetings and important events.

[1354] Step 9:

[1355] The server creates a draft of the thank-you email based on the details of the business negotiation. It analyzes the text data and generates a draft email incorporating appropriate greetings and words of thanks. The generated draft email is then saved in storage.

[1356] Step 10:

[1357] The server sends the generated minutes, to-do list, Google Calendar schedule, and a draft of the thank-you email to the user's device. It also sends a notification to the user's device and provides a download link for the generated data.

[1358] Step 11:

[1359] The user receives a notification and clicks the provided link to download the generated data to their device. The user can provide feedback as needed, check and edit the data, and use it in their next sales activity.

[1360] Through the above processing steps, this system can improve the efficiency of the work that sales representatives perform after visiting customers and reduce their workload.

[1361] Example 1

[1362] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1363] The post-processing work that sales representatives perform after customer visits requires time and effort, hindering efficient sales activities. Furthermore, manual text generation and information organization have limited accuracy, leading to the risk of important information being overlooked or typographical errors. There is a need for a system that can solve these problems, reduce the workload of sales representatives, and enable fast and accurate information processing.

[1364] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1365] In this invention, the server includes means for acquiring voice data, means for converting the acquired voice data into text data, means for analyzing the converted text data to generate minutes, means for analyzing the converted text data to generate a to-do list, means for analyzing the converted text data to generate a schedule, means for analyzing the converted text data to generate a draft of an email, means for transmitting the generated data to a user terminal, means for converting the recorded data into Japanese text using a voice recognition engine, and means for generating various information from the text data using a generative AI model. This allows a user to simply record the contents of a business meeting and upload the data to the server, which automatically generates drafts of minutes, to-do lists, schedules, and thank-you emails, enabling efficient and accurate post-event processing.

[1366] "Voice data" refers to data that has been saved in digital format containing the content of conversations and dialogues recorded during customer visits, business negotiations, etc.

[1367] "Means of capture" refers to the combination of devices and software used by a user to record audio data and upload it to a server.

[1368] "Text data" refers to data obtained by converting acquired voice data into text format.

[1369] "Means for converting" refers collectively to a speech recognition engine and related software used to convert voice data into text data.

[1370] "Means of analysis" is a general term for the algorithms and generative AI models used to extract and organize key information from the converted text data.

[1371] Minutes are documents that summarize the main points of business negotiations, meetings, etc., and are used to confirm the contents later.

[1372] A "ToDo list" is a list of tasks or homework that need to be completed by the next time.

[1373] "Schedule" refers to schedule information such as the next date and time of a business meeting or meeting, or important deadlines.

[1374] An "email draft" is a draft of a greeting or thank you email to be used after a business meeting.

[1375] "Generative Means" refers collectively to generative AI models and related algorithms used to create meeting minutes, to-do lists, schedules, and email drafts from text data.

[1376] "Transmission means" is a general term for communication means and protocols for transmitting generated data to a user terminal.

[1377] This invention is a system that uses advanced voice data to automate the post-processing process after sales representatives visit customers. Specifically, it converts voice data into Japanese text data, and then uses a generative AI model based on that text data to automatically generate meeting minutes, to-do lists, schedules, and thank-you emails. This system is primarily comprised of components that play the roles of a server, terminal, and user.

[1378] Hardware and Software Overview

[1379] Terminal

[1380] Users use devices such as smartphones and tablets, which have an application with a recording function installed, allowing them to record the contents of business negotiations.

[1381] server

[1382] The server is the main device that receives, stores, analyzes, and distributes the generated audio data. The following software is installed on the server:

[1383] Speech Recognition Engine: Speech recognition software such as the Google Cloud Speech-to-Text API.

[1384] Generative AI models: Advanced generative AI software such as GPT-3.

[1385] User

[1386] The user is a sales representative who records voice data during a sales meeting and uploads the data to the server after the recording is complete. Post-processing is performed based on the generated information.

[1387] Data processing and calculation flow

[1388] 1. Acquiring and uploading audio data

[1389] When a user visits a customer, the user uses the device to record the details of the business meeting. After completing the recording, the user uploads the audio data from the device to the server. The server receives the recorded data via an HTTP request and stores it in storage.

[1390] 2. Japanese text

[1391] The server passes the received recorded data to a speech recognition engine and converts it into high-quality Japanese text data, which is then saved back into storage.

[1392] 3. Text Data Analysis and Generation

[1393] The server inputs the stored text data into a generative AI model to generate the following information:

[1394] Minutes: Summarize the main points of the business meeting in bullet points.

[1395] To-Do List: Make a list of tasks and homework that need to be done by the next time.

[1396] Appointments: Generate appointments in Google Calendar format based on the dates and times mentioned in the deal.

[1397] Draft a thank you email: Create a draft of a thank you email based on the details of the business meeting.

[1398] 4. Data distribution and feedback

[1399] The server sends the generated minutes, to-do lists, schedules, and thank-you email drafts to the user's device. The user receives a notification, checks and edits the generated data, and uses it as needed.

[1400] Examples and prompts

[1401] Specific examples

[1402] A user visits a customer and conducts a business meeting. The contents of the business meeting are recorded on the device, and after the business meeting is completed, the recording is stopped and the recorded data is uploaded to the server.

[1403] The server receives the recording, converts it into Japanese text using a speech recognition engine, and then uses a generative AI model to generate drafts of meeting minutes, to-do lists, schedules, and thank-you emails, which are then sent to the user's device.

[1404] Prompt Sentence Examples

[1405] By inputting the following prompt sentence into the generative AI model, we can get the corresponding output:

[1406] Prompt for generating transcripts:

[1407] "Please extract the key points from the following sales pitch in bullet points format:

[1408] 1. Customer needs

[1409] 2. Proposal details

[1410] 3. Questions and Answers

[1411] 4. Next assignment

[1412] Prompt for generating a ToDo list:

[1413] From the text below, please list the tasks you need to complete by next time:

[1414] Summary of business negotiations

[1415] Confirm next step

[1416] Preparing for follow-up

[1417] Prompt for generating an appointment:

[1418] "Extract important dates and times from the text below and generate a Google Calendar event."

[1419] Prompt for generating a draft thank you email:

[1420] "Please draft a thank you email based on the following business details:

[1421] Thank you message

[1422] Review the main points of the deal

[1423] Proposal for next steps

[1424] As a result, this system significantly improves the work efficiency of sales staff and reduces the time and effort required for post-sales processing.

[1425] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1426] Step 1:

[1427] Acquiring and uploading audio data

[1428] The user uses a device such as a smartphone or tablet to record the contents of a business meeting. They open a recording application and press the record button to record the contents of the business meeting. The input data is the audio spoken during the business meeting. After the business meeting is over, the user presses the "Stop" button to end recording, and then presses the "Upload" button to send the recorded data to the server. The server receives the HTTP request and saves the recorded data in storage. The output data is an audio file saved in storage.

[1429] Step 2:

[1430] Japanese text

[1431] The server passes the saved recording data to a speech recognition engine. The input data is an audio file. The server uses a speech recognition engine such as the Google Cloud Speech-to-Text API to convert this audio data into high-quality Japanese text data. Specifically, the speech recognition engine analyzes the audio waveform and converts it into text information. The converted text data is then saved back into storage within the server. The output data is Japanese text data.

[1432] Step 3:

[1433] Text data analysis and generation

[1434] The server inputs text data into the generative AI model to generate specific information. The input data is Japanese text data. The server uses the following prompt to have the generative AI model (e.g., GPT-3) analyze the data:

[1435] Prompt for generating transcripts:

[1436] "Please extract the key points from the following sales pitch in bullet points format:

[1437] 1. Customer needs

[1438] 2. Proposal details

[1439] 3. Questions and Answers

[1440] 4. Next assignment

[1441] Prompt for generating a ToDo list:

[1442] From the text below, please list the tasks you need to complete by next time:

[1443] Summary of business negotiations

[1444] Confirm next step

[1445] Preparing for follow-up

[1446] Prompt for generating an appointment:

[1447] "Extract important dates and times from the text below and generate a Google Calendar event."

[1448] Prompt for generating a draft thank you email:

[1449] "Please draft a thank you email based on the following business details:

[1450] Thank you message

[1451] Review the main points of the deal

[1452] Proposal for next steps

[1453] The server uses these prompts to have the generative AI model analyze the text data and generate information for each purpose. The generated information becomes the basis for meeting minutes, to-do lists, schedules, and thank-you emails, all of which are stored in the server's storage. The output data is the various types of generated information.

[1454] Step 4:

[1455] Data distribution and feedback

[1456] The server sends the generated data (minutes, to-do list, schedule, thank-you email draft) to the user's device. The input data is the various types of generated information. The server uses an HTTP response to send the generated data to the user's device. The user receives a notification on their device and checks and edits the generated data. Specifically, the user clicks the notification to open the application, checks the generated minutes, to-do list, schedule, and thank-you email, and edits them as necessary. The output data is the various types of information after it has been checked and edited by the user.

[1457] By following these steps, users can simply upload their voice data to the server, and minutes, to-do lists, schedules, and thank-you email drafts will be automatically generated and provided to them. This series of processes will greatly improve the efficiency of post-sales processing work for sales representatives.

[1458] (Application example 1)

[1459] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1460] During maintenance work in a factory, recording the details of the work and any problems that arise, and then creating work reports and to-do lists accurately and efficiently based on that information, places a heavy burden on workers. This creates the problem of time being spent creating reports and checking schedules, which reduces the efficiency of the work. Furthermore, mistakes are likely to occur in work reports and schedule settings, which can have an impact on factory operations. To solve these problems, an automated system is needed.

[1461] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1462] In this invention, the server includes means for acquiring voice data, means for converting the acquired voice data into text data, means for analyzing the converted text data to generate minutes, means for analyzing the converted text data to generate a to-do list, means for analyzing the converted text data to generate a schedule, means for analyzing the converted text data to generate a draft of an e-mail, means for sending the generated data to a user terminal, means for outputting the generated data as an in-factory work report, means for outputting the generated data as a work schedule in calendar format, and means for outputting the generated data as a work report e-mail. This automates in-factory maintenance work reporting and schedule setting, not only reducing the burden on workers but also improving work efficiency and reducing reporting and scheduling errors.

[1463] "Voice data" refers to audio information that records events, problems, and work conditions that occur during factory work.

[1464] "Text data" refers to data obtained by converting voice data into character information.

[1465] "Analysis" is the process of analyzing the converted text data and generating reports, lists, schedules, etc. based on it.

[1466] Minutes are documents that organize and summarize the work content, important points, and issues.

[1467] A "ToDo list" is a list in bulleted form that lists the next tasks to be done and items that need to be improved.

[1468] "Schedule" is a calendar listing the next work schedule and maintenance plan.

[1469] An "email draft" is a draft of an email containing a work report or contact information.

[1470] A "user terminal" is an electronic device such as a smartphone or tablet used by a worker.

[1471] The "Factory Work Report" is a report summarizing the details of the maintenance work carried out within the factory.

[1472] "Work schedule" is schedule information including the next maintenance work and check items.

[1473] A "work report email" is an email containing the report content to be sent to the relevant parties after the maintenance work has been completed.

[1474] "Cloud storage" is a service for storing data over the Internet.

[1475] This invention is a system for improving the efficiency of maintenance work in factories. Specifically, it converts voice data during work into text data, and then uses a generative AI model based on that text data to automatically generate drafts of work reports, to-do lists, work schedules, and thank-you emails. This system is implemented as follows.

[1476] First, the user, a worker, uses a device such as a smartphone or tablet to record the details of the maintenance work in the factory by voice. The hardware used for this is a smartphone or tablet. The recorded data is managed using a cloud server.

[1477] Once the audio data is acquired, the user uploads the recording to the cloud server via their device, and the cloud server stores the received recording data in its storage.

[1478] The server then uses a cloud-based voice recognition engine (e.g., Google Cloud Speech-to-Text) to convert the recorded data into high-quality text data, which is then stored in cloud storage.

[1479] The generated text data is used with a generative AI model (e.g., OpenAI GPT-4) to produce the following output:

[1480] 1. Generation of work reports: Analyze text data, extract key points of maintenance work, and generate reports.

[1481] 2. Generate a to-do list: Extract next tasks and areas for improvement from text data and create a to-do list.

[1482] 3. Generate work schedule: Analyze the date and time information in the text data and generate the next maintenance schedule in calendar format.

[1483] 4. Generate a thank you email: Draft a thank you email based on the maintenance work.

[1484] This generated data is sent from the server to the user's device. The user receives a notification, can check and edit the generated data, and use it as needed. This significantly reduces the time required for post-processing after maintenance work.

[1485] Examples of illustrative prompts include:

[1486] Generate meeting minutes, to-do lists, appointments, and thank you emails based on:

[1487] "The following problem was discovered during today's maintenance work. There is a small leak in the hydraulic system and the valve needs to be replaced. The next inspection date is October 20th. All other inspection items are normal."

[1488] The system according to the present invention has the effect of significantly reducing the burden on workers and improving work efficiency, and also reduces errors in reporting and schedule setting.

[1489] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1490] Step 1:

[1491] Users can use their smartphones or tablets to record their maintenance work by voice, and the recorded data is saved on the device.

[1492] Input: User voice input

[1493] Output: Recording data saved on the device

[1494] Specific operation: The user presses the start recording button, talks about the maintenance work and problems, and then presses the stop button to end the recording.

[1495] Step 2:

[1496] Users upload recorded data from their devices to the cloud server, which receives the data and stores it in its storage.

[1497] Input: Recording data stored on the device

[1498] Output: Recording data stored on a cloud server

[1499] Specific operation: The user presses the upload button to send the recorded data from the device to the cloud server. The server receives the recorded data and stores it in a database.

[1500] Step 3:

[1501] The server converts the recorded data into text data using a speech recognition engine, and stores the converted text data in cloud storage.

[1502] Input: Recording data stored on a cloud server

[1503] Output: Text data stored in cloud storage

[1504] Specific operation: The server uses a speech recognition engine such as Google Cloud Speech-to-Text to analyze the recorded data and convert it into high-quality Japanese text. The converted text data is then stored in cloud storage.

[1505] Step 4:

[1506] The server uses a generative AI model to generate the following outputs based on the text data:

[1507] business report

[1508] To-Do List

[1509] Work Schedule

[1510] Work report email

[1511] Input: Text data stored in cloud storage

[1512] Output: Generated work reports, ToDo lists, work schedules, work report emails

[1513] Specific operation: The server inputs text data into a generative AI model such as OpenAI GPT-4, and generates prompt sentences for minutes, to-do lists, schedules, and thank-you emails.

[1514] Step 5:

[1515] The server sends the generated data to the user's terminal, where the user receives the data and checks and edits it.

[1516] Input: Generated work reports, ToDo lists, work schedules, work report emails

[1517] Output: Data sent to the user's terminal

[1518] Specific operation: The server sends the generated data to the user's terminal. The user receives a notification, checks it, and edits it if necessary.

[1519] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1520] This invention is an advanced sales support system incorporating an emotion engine that recognizes user emotions, and is intended to reduce the workload of sales representatives after customer visits. Specifically, it converts voice data into high-quality text data, analyzes that text data, and automatically generates drafts of meeting minutes, to-do lists, schedules, and thank-you emails. In addition, by adding emotion analysis using the emotion engine, it provides more flexible and personalized output.

[1521] Explanation of program processing

[1522] Acquiring and uploading audio data

[1523] When a user visits a customer, the user records the details of the business meeting using a device such as a smartphone or tablet. Once the recording is complete, the user uploads the recording data from the device to the server. During this process, the server receives the recording data and prepares it for processing.

[1524] Japanese text

[1525] The server converts the received recording data into high-quality Japanese text using a speech recognition engine. The converted text data is stored in the server and used in the next step.

[1526] Analysis by emotion engine

[1527] The server inputs the text data into the emotion engine to analyze the user's emotions. The emotion engine extracts emotional elements (happiness, sadness, surprise, etc.) from the text and reflects the results in the next generation AI process.

[1528] Processing of generated AI

[1529] When text data containing sentiment analysis results is input into the generative AI, the server generates the following outputs:

[1530] 1. Generate transcripts:

[1531] The server analyzes the text data and generates minutes that reflect the key points of the business meeting and the user's emotions, allowing the system to record not only the important points of the business meeting but also the emotional aspects of the discussion.

[1532] 2. Create a to-do list:

[1533] It extracts tasks and homework from text data and creates a to-do list based on the user's emotions, allowing the user to manage task priorities so that they do not feel stressed.

[1534] 3. Create an appointment:

[1535] Based on text data and sentiment analysis, it generates upcoming meetings and important appointments in Google Calendar format. By taking into account sentiment information, it helps users to comfortably complete their next appointment.

[1536] 4. Draft thank you email:

[1537] The server creates a draft thank-you email based on the details of the business and the results of sentiment analysis, allowing it to generate an email that incorporates emotionally appropriate greetings and words of thanks.

[1538] feedback

[1539] The generated minutes, to-do lists, schedules, and draft thank-you emails are sent from the server to the user's device. The user receives a notification and can check and edit each piece of data. This significantly reduces the time spent on post-processing after returning to the office, allowing them to focus on their next sales activity.

[1540] Specific examples

[1541] Customer Visit Scenario

[1542] A user visits a customer and conducts a business negotiation. During the negotiation, the contents of the negotiation are recorded on the device. After the negotiation is over, the user presses the "Stop" button to stop recording, and then presses the "Upload" button to send the recorded data to the server.

[1543] Specific operations of server-side processing

[1544] 1. The server receives the recording data and saves it in storage.

[1545] 2. The server uses a speech recognition engine to convert the recorded data into high-quality Japanese text, and stores the converted text in storage.

[1546] 3. The server inputs the text data into the emotion engine to analyze the user's emotions, and passes the emotion analysis results to the generative AI model.

[1547] 4. The server uses the generative AI model to generate the following data:

[1548] Minutes: Include emotional information, such as "The key points of the meeting were..."

[1549] To-do lists: Emotionally driven management, such as "Homework items for next time..."

[1550] Google Calendar Appointments: Emotionally-driven schedules, such as "Your next meeting is scheduled for August 10th at 10am."

[1551] Draft of a thank you email: A draft that reflects your feelings, such as "Thank you for the business meeting the other day..."

[1552] 5. The generated data is collected and sent to the user's device. The user receives a notification, checks and edits the generated data, and uses it as needed.

[1553] In this way, the system of the present invention recognizes and analyzes the user's emotions and generates output that reflects them, thereby providing more personalized sales support and reducing the workload of sales representatives.

[1554] The processing flow will be explained below.

[1555] Step 1:

[1556] When a user visits a customer, they use a device (smartphone or tablet) to record the details of the business meeting. They launch the device's voice recording app and press the "Record" button to start recording. When the business meeting is over, they press the "Stop" button to stop recording.

[1557] Step 2:

[1558] The user uploads the recorded data to the server from the voice recording app on their device. After pressing the "Upload" button on the app and confirming the connection to the server, the recorded data is sent to the server.

[1559] Step 3:

[1560] The server receives the uploaded recording, stores it in storage, and adds it to the conversion queue.

[1561] Step 4:

[1562] The server calls the speech recognition engine to convert the recorded data into high-quality Japanese text. The speech recognition engine analyzes the recorded data and generates Japanese text data. The converted text data is saved in storage.

[1563] Step 5:

[1564] The server inputs the stored Japanese text data into the emotion engine to analyze the user's emotions. The emotion engine analyzes the text data and extracts emotional elements (joy, sadness, surprise, etc.). The emotion analysis results are passed to the generative AI model.

[1565] Step 6:

[1566] The server generates meeting minutes using the generative AI model. Taking into account the results of sentiment analysis, it extracts key points from the text data and creates minutes in bullet point format. The minutes also reflect emotional comments and important tones. The generated minutes data is saved in storage.

[1567] Step 7:

[1568] The server generates a to-do list using the generative AI model. Based on the results of sentiment analysis, action items (homework items) are extracted from the text data. A to-do list is created that takes into account the user's emotional state, task priority, and workload. The generated to-do list data is saved in storage.

[1569] Step 8:

[1570] The server generates schedules using a generative AI model. Taking into account the results of sentiment analysis, it extracts date and time information from the text data and creates upcoming meetings and important appointments in Google Calendar format. Based on the emotional information, it adjusts the schedule so that the user can complete it comfortably. The schedule data is automatically registered in Google Calendar.

[1571] Step 9:

[1572] The server uses the generative AI model to create a draft of a thank-you email. Based on the results of the sentiment analysis, it creates a draft email that includes a greeting and expression of gratitude appropriate to the business negotiation. It uses an appropriate tone and expressions depending on the emotional information. The generated email draft data is saved in storage.

[1573] Step 10:

[1574] The server sends the generated minutes, to-do list, Google Calendar schedule, and a draft of the thank-you email to the user's device. The server also sends a notification to the user's device and provides a download link for the generated data.

[1575] Step 11:

[1576] The user receives a notification and clicks the provided link to download the generated data to their device. The user can then review and edit the data as needed and use it for their next sales activity.

[1577] Through the above processing steps, the system can recognize the user's emotions and generate output that reflects them, thereby providing more personalized sales support and reducing the workload of sales representatives.

[1578] Example 2

[1579] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1580] Modern sales activities require a wide range of post-event processing, such as creating reports after customer visits, organizing tasks, setting up next-time appointments, and writing follow-up emails. These tasks require time and effort, placing a significant burden on sales representatives. Furthermore, accurate recording of customer interactions and conveying information that incorporates emotions and nuances is required, but doing this manually is difficult. There is a need for the development of a system that can solve these issues, reduce the workload of sales representatives, and provide efficient, personalized sales support.

[1581] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for acquiring voice data, means for converting the acquired voice data into text data, means for analyzing the converted text data to recognize emotions, means for analyzing the text data including the emotion recognition results to generate minutes, means for analyzing the text data including the emotion recognition results to generate a to-do list, means for analyzing the text data including the emotion recognition results to generate a schedule, means for analyzing the text data including the emotion recognition results to generate a draft of an e-mail, and means for transmitting the generated data to a user terminal. This allows the content of a conversation with a customer to be accurately recorded, information including emotions and nuances to be automatically generated, and sales representatives to quickly and efficiently perform follow-up processing.

[1582] The "means for acquiring voice data" is a function for recording the contents of business negotiations via the terminal used by the user when visiting a customer, and transmitting the recorded data to the server.

[1583] The "means for converting acquired voice data into text data" is a function in which the server converts the received voice data into high-quality text format using a voice recognition engine.

[1584] The "means for analyzing the converted text data and recognizing emotions" is a function by which the server inputs the text data into an emotion analysis engine and extracts and recognizes the emotional elements contained in the text data.

[1585] The "means for generating minutes by analyzing text data including emotion recognition results" is a function in which the server analyzes text data that reflects emotion recognition results and automatically generates minutes that include the main points and emotional aspects of the business negotiations.

[1586] The "means for generating a to-do list by analyzing text data that includes emotion recognition results" is a function in which the server extracts tasks and homework from text data that reflects emotion recognition results, and generates a to-do list prioritized based on emotions.

[1587] The "means for generating schedules by analyzing text data including emotion recognition results" is a function in which the server automatically generates the next meeting or important schedule based on text data that reflects emotion recognition results.

[1588] The "means for analyzing text data including emotion recognition results and generating a draft of an e-mail" is a function in which the server automatically generates a draft of a thank-you e-mail including an appropriate greeting and words of gratitude based on the text data in which the emotion recognition results are reflected.

[1589] "Means for transmitting generated data to the user terminal" refers to a function for transmitting the minutes, to-do list, schedule, and draft thank-you email generated by the server to the user terminal.

[1590] MODE FOR CARRYING OUT THE INVENTION

[1591] This invention is an advanced sales support system incorporating an emotion engine that recognizes user emotions, and is intended to reduce the workload of sales representatives after customer visits. Specifically, it converts voice data into high-quality text data, analyzes that text data, and automatically generates drafts of meeting minutes, to-do lists, schedules, and thank-you emails. In addition, by adding emotion analysis using the emotion engine, it provides more flexible and personalized output.

[1592] When visiting a customer, the user records the details of the business meeting using a device such as a smartphone or tablet. Once the recording is complete, the user uploads the recording data from the device to the server. During this process, the server receives the recording data and prepares it for processing.

[1593] The server converts the received recording data into high-quality Japanese text. To do this, the server uses a speech recognition engine (e.g., Google Speech-to-Text API). The converted text data is stored on the server and used in the next step.

[1594] The server inputs the saved text data into an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotions. The emotion engine extracts emotional elements such as joy, sadness, and surprise from the text and reflects the results in the generative AI model.

[1595] When text data containing sentiment analysis results is input into a generative AI model (e.g., OpenAI GPT-4), the server generates the following outputs:

[1596] 1. Minutes:

[1597] The text data that reflects the emotion recognition results is analyzed to generate minutes that include the key points and emotional aspects of the business negotiations.

[1598] Example: "The key points of the discussion are..."

[1599] 2. To-Do List:

[1600] Tasks and homework are extracted from text data that reflects emotion recognition results, and a to-do list prioritized based on emotion is generated.

[1601] Example: "As homework for next time..."

[1602] 3. Plan:

[1603] Generate upcoming meetings and important appointments in Google Calendar format based on text data and sentiment analysis.

[1604] By taking emotional information into consideration, we ensure that users can comfortably complete their next appointment.

[1605] Example: "Our next meeting is scheduled for August 10th at 10am."

[1606] 4. Draft of a thank you email:

[1607] Draft a thank you email based on the deal details and sentiment analysis results.

[1608] Example: "Thank you for the business meeting the other day..."

[1609] The generated minutes, to-do lists, schedules, and draft thank-you emails are sent from the server to the user's device. The user receives a notification and can check and edit each piece of data. This significantly reduces the time spent on post-processing after returning to the office, allowing them to focus on their next sales activity.

[1610] Specific examples

[1611] A user visits a customer and conducts a business negotiation. During the negotiation, the contents of the negotiation are recorded on the device. After the negotiation is over, the user presses the "Stop" button to stop recording, and then presses the "Upload" button to send the recorded data to the server.

[1612] The server receives the recorded data and saves it in storage. The received recorded data is converted into high-quality Japanese text using a speech recognition engine (e.g., Google Speech-to-Text API), and the converted text data is also saved in storage. Next, the server inputs the text data into an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotions. The emotion analysis results are then passed to a generative AI model (e.g., OpenAI GPT-4) to generate the following data:

[1613] Minutes: Include emotional information, such as "The key points of the meeting were..."

[1614] To-do lists: Emotionally driven management, such as "Homework items for next time..."

[1615] Google Calendar Appointments: Emotionally-driven schedules, such as "Your next meeting is scheduled for August 10th at 10am."

[1616] Draft of a thank you email: A draft that reflects your feelings, such as "Thank you for the business meeting the other day..."

[1617] The generated data is collected and sent to the user's device, where the user receives a notification, can review and edit the generated data, and use it as needed. This system recognizes and analyzes the user's emotions and generates output that reflects them, thereby providing more personalized sales support and reducing the workload of sales representatives.

[1618] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1619] Step 1:

[1620] When a user visits a customer, the user uses a device such as a smartphone or tablet to record the details of the business meeting. Specifically, when the user presses the record button on the device, the device begins recording audio data. After the business meeting is over, the user presses the "Stop" button to stop recording, and then presses the "Upload" button to send the recorded data to the server. The input is the user's voice, and the output is the audio file sent to the server.

[1621] Step 2:

[1622] The server saves the voice data received from the device in initial storage. Next, the server uses a voice recognition engine (e.g., Google Speech-to-Text API) to convert the voice data into high-quality Japanese text. The specific operation of this process is to pass the voice data to the API and receive the converted text data. The input is an audio file, and the output is the converted text data.

[1623] Step 3:

[1624] The server inputs the converted text data into an emotion engine (e.g., IBM Watson Tone Analyzer) for emotion analysis. The emotion engine analyzes the content of the text data and extracts emotional elements such as joy, sadness, and surprise. The specific operation of this process is for the server to pass the text data to the emotion engine's API and receive the emotion analysis results. The input is text data, and the output is data containing the emotion analysis results.

[1625] Step 4:

[1626] The server inputs the text data including the sentiment analysis results into a generative AI model (e.g., OpenAI GPT-4) to generate various documents. The specific operation at this time is to pass the sentiment analysis results and text data to the generative AI model as prompt sentences, and obtain the following outputs:

[1627] 1. Meeting minutes - Minutes that include the key points and emotional aspects of the business meeting.

[1628] 2. To-Do List - A list that reflects emotion-based task management.

[1629] 3. Appointments - Schedule your next meeting or important appointment.

[1630] 4. Draft of a thank you email - An email containing appropriate greetings and thank you words that reflect the sentiment. The input is text data and sentiment analysis results, and the output is the respective documents.

[1631] Step 5:

[1632] The server sends the generated minutes, to-do list, schedule, and thank-you email drafts to the user's device. Specifically, the server sends each generated document to the device linked to the user's account and notifies them. The input is each generated document, and the output is the transmission and notification to the user's device.

[1633] Step 6:

[1634] The user can review and edit the drafts of meeting minutes, to-do lists, schedules, and thank-you emails generated on their device and use them as needed. Specifically, the user can open the application on their device and view and edit the contents of each document. The input is each document sent from the server, and the output is the final document edited by the user.

[1635] (Application example 2)

[1636] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1637] Conventional sales support systems and customer service systems have difficulty responding flexibly based on customer emotions, and have not been able to fully improve customer satisfaction or sales efficiency. In particular, there is no system that can analyze customer emotions in real time and provide appropriate responses to customer service staff based on that analysis, so there is a need for more advanced customer service support.

[1638] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring voice data, means for converting the acquired voice data into text data, means for analyzing the converted text data using an emotion engine, means for generating minutes using a generative AI model including the analyzed text data and emotion analysis results, means for generating a to-do list using the generative AI model, means for generating a schedule using the generative AI model, means for generating a draft email using the generative AI model, means for displaying the generated data and emotion analysis results on a user interface in real time, and means for transmitting the generated data to a user terminal. This makes it possible to analyze customer emotions in real time, present personalized countermeasures based on the results, and efficiently and quickly process data after customer service.

[1639] "Audio data" refers to audio signals recorded in digital format, and is information that can be converted into text data and analyzed.

[1640] "Text data" refers to written information converted from audio data and used as input for analytical and generative AI models.

[1641] An "emotion engine" is a software technology for extracting emotional elements from text data and analyzing emotions.

[1642] A "generative AI model" is an artificial intelligence technology that uses machine learning algorithms to automatically generate output such as minutes, to-do lists, schedules, and email drafts from input data.

[1643] "User interface" refers to the screen or operation panel that allows users to operate the system, and which displays sentiment analysis results and suggestions in real time.

[1644] A "user terminal" is a device that is directly operated by a user, and includes smartphones, tablets, smart glasses, and the like.

[1645] "Real-time" means that data acquisition, processing, and output are all done instantly, and information is provided without delay.

[1646] The present invention is a system for efficiently supporting customer service, and is implemented in the following procedure.

[1647] Program Generation

[1648] Hardware and software used

[1649] The system uses the following hardware and software:

[1650] Smart glasses (device example: general noun)

[1651] Speech recognition engine (software example: Google Cloud Speech-to-Text)

[1652] Sentiment analysis engine (software example: IBM Watson Tone Analyzer)

[1653] Generative AI model (software example: OpenAI GPT-3)

[1654] Cloud storage (software example: general noun)

[1655] Acquiring and uploading audio data

[1656] The user puts on the smart glasses and starts talking to the customer. The smart glasses record the conversation in real time and upload the audio data to the server.

[1657] Converting audio data to text

[1658] The server uses a speech recognition engine to convert the received voice data into high-quality text data, which is then stored in cloud storage.

[1659] Analysis by emotion engine

[1660] The server inputs the converted text data into a sentiment analysis engine to analyze customer sentiment in real time, and the analysis results are fed back to the smart glasses.

[1661] Data generation using generative AI models

[1662] Based on the analyzed text data and sentiment analysis results, the generative AI model generates the following data:

[1663] 1. Meeting minutes: Automatically generate meeting minutes that reflect the key points and sentiment of the conversation.

[1664] 2. To-Do List: Extract upcoming tasks and generate a prioritized to-do list.

[1665] 3. Schedule: Generate upcoming meetings and important events in Google Calendar format.

[1666] 4. Email Draft: Generate an emotionally relevant thank you email draft.

[1667] Real-time display and feedback

[1668] The generated data and sentiment analysis results are displayed in real time on the smart glasses' user interface, allowing users to immediately identify appropriate countermeasures. After the analysis is complete, the generated data is sent to the user's device for review and editing.

[1669] Specific examples

[1670] Customer Visit Scenario

[1671] A user visits a customer, puts on smart glasses, and begins a conversation. The smart glasses record the voice and upload it to the server in real time. The server converts the voice data into text and analyzes it using a sentiment analysis engine. The analysis results are displayed on the smart glasses' user interface, and the user can take appropriate action. The generative AI model then generates drafts of meeting minutes, to-do lists, schedules, and thank-you emails, which are saved in cloud storage.

[1672] Prompt Sentence Examples

[1673] Conversation:

[1674] "Hi, can you tell me more about this product? Is it really easy to use?"

[1675] Sentiment analysis results:

[1676] "The customer is interested. Please provide a detailed product description and use case."

[1677] Generated transcript:

[1678] "Service: The customer is interested in Product A and asks about its ease of use. A detailed explanation and examples of use are provided."

[1679] Generated thank you email draft:

[1680] "Thank you very much for visiting our store the other day. We received your question about Product A and we explained the details to you. If there is anything else you would like to ask, please feel free to contact us."

[1681] In this embodiment, the present invention can analyze customer emotions in real time, provide personalized countermeasures based on the analysis, and process data after the customer service quickly and efficiently.

[1682] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1683] Step 1:

[1684] The user puts on the smart glasses and starts a conversation with a customer. The smart glasses record the conversation in real time and capture the audio data. The captured audio data is then uploaded from the smart glasses to a server.

[1685] Input: Real-time audio data

[1686] Output: Uploaded audio data

[1687] Step 2:

[1688] The server inputs the uploaded voice data into a speech recognition engine (e.g., Google Cloud Speech-to-Text) and converts it into high-quality text data. The converted text data is then stored in the server's internal storage.

[1689] Input: Audio data

[1690] Output: Text data

[1691] Step 3:

[1692] The server inputs the converted text data into an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze emotions in real time. The analysis results are added to the text data and saved in storage as emotion analysis results.

[1693] Input: Text data

[1694] Output: Text data with sentiment analysis results

[1695] Step 4:

[1696] The server inputs the text data with sentiment analysis results into a generative AI model (e.g., OpenAI GPT-3) to generate the following data:

[1697] 1. Minutes

[1698] 2. To-Do List

[1699] 3. Schedule

[1700] 4. Draft the email

[1701] Each generated data is stored separately in storage.

[1702] Input: Text data with sentiment analysis results

[1703] Output: minutes, to-do lists, schedules, email drafts

[1704] Step 5:

[1705] The server provides feedback to the user by displaying the generated data and the emotion analysis results in real time on the user interface (smart glasses). The user can then check this information in real time and take appropriate action.

[1706] Input: Generated data, sentiment analysis results

[1707] Output: Real-time display on smart glasses

[1708] Step 6:

[1709] After the customer service is over, the server sends the generated data to the user's device. The user can check the data on the device and edit or add to it as needed. The data is also saved in cloud storage.

[1710] Input: Generated data

[1711] Output: Data sent to user devices, data saved to cloud storage

[1712] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1713] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1714] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1715] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1716] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1717] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1718] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1719] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1720] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1721] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1722] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1723] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1724] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1725] 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.

[1726] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1727] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1728] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1729] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1730] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1731] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1732] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1733] The following is further disclosed regarding the above embodiment.

[1734] (Claim 1)

[1735] means for acquiring audio data;

[1736] A means for converting the acquired voice data into text data;

[1737] A means for analyzing the converted text data to generate minutes;

[1738] A means for analyzing the converted text data and generating a to-do list;

[1739] means for analyzing the converted text data to generate a schedule;

[1740] means for analyzing the converted text data to generate a draft of an email;

[1741] means for transmitting the generated data to a user terminal;

[1742] A system including:

[1743] (Claim 2)

[1744] 10. The system of claim 1, further comprising means for notifying the user terminal of the transmitted data.

[1745] (Claim 3)

[1746] 10. The system of claim 1, further comprising means for storing the generated data in cloud storage.

[1747] "Example 1"

[1748] (Claim 1)

[1749] means for acquiring audio data;

[1750] A means for converting the acquired voice data into text data;

[1751] A means for analyzing the converted text data to generate minutes;

[1752] A means for analyzing the converted text data and generating a to-do list;

[1753] means for analyzing the converted text data to generate a schedule;

[1754] means for analyzing the converted text data to generate a draft of an email;

[1755] means for transmitting the generated data to a user terminal;

[1756] A means for converting the recorded data into Japanese text using a speech recognition engine;

[1757] A means for generating various information from text data using a generative AI model;

[1758] A system including:

[1759] (Claim 2)

[1760] 10. The system of claim 1, further comprising means for notifying the user terminal of the transmitted data.

[1761] (Claim 3)

[1762] 10. The system of claim 1, further comprising means for storing the generated data in cloud storage.

[1763] "Application Example 1"

[1764] (Claim 1)

[1765] means for acquiring audio data;

[1766] A means for converting the acquired voice data into text data;

[1767] A means for analyzing the converted text data to generate minutes;

[1768] A means for analyzing the converted text data and generating a to-do list;

[1769] means for analyzing the converted text data to generate a schedule;

[1770] means for analyzing the converted text data to generate a draft of an email;

[1771] means for transmitting the generated data to a user terminal;

[1772] A means for outputting the generated data as an in-factory business report;

[1773] a means for outputting the generated data as a work schedule in a calendar format;

[1774] means for outputting the generated data as a work report email;

[1775] A system including:

[1776] (Claim 2)

[1777] 10. The system of claim 1, further comprising means for notifying the user terminal of the transmitted data.

[1778] (Claim 3)

[1779] 10. The system of claim 1, further comprising means for storing the generated data in cloud storage.

[1780] "Example 2: Combining Emotion Engines"

[1781] (Claim 1)

[1782] means for acquiring audio data;

[1783] A means for converting the acquired voice data into text data;

[1784] A means for analyzing the converted text data to recognize emotions;

[1785] A means for analyzing text data including emotion recognition results and generating minutes;

[1786] A means for analyzing text data including emotion recognition results and generating a to-do list;

[1787] means for analyzing text data including emotion recognition results and generating a schedule;

[1788] means for analyzing text data including emotion recognition results to generate a draft of an email;

[1789] means for transmitting the generated data to a user terminal;

[1790] A system including:

[1791] (Claim 2)

[1792] 10. The system of claim 1, further comprising means for notifying the user terminal of the transmitted data.

[1793] (Claim 3)

[1794] 10. The system of claim 1, further comprising means for storing the generated data in cloud storage.

[1795] "Application example 2 when combining emotion engines"

[1796] (Claim 1)

[1797] means for acquiring audio data;

[1798] A means for converting the acquired voice data into text data;

[1799] A means for analyzing the converted text data using an emotion engine;

[1800] A means for generating minutes using a generative AI model including the analyzed text data and sentiment analysis results;

[1801] A means for generating a to-do list using a generative AI model;

[1802] A means for generating a schedule using a generative AI model;

[1803] A means for generating a draft email using a generative AI model; and

[1804] means for displaying the generated data and sentiment analysis results in a user interface in real time;

[1805] means for transmitting the generated data to a user terminal;

[1806] A system including:

[1807] (Claim 2)

[1808] 10. The system of claim 1, further comprising means for notifying the user terminal of the transmitted data.

[1809] (Claim 3)

[1810] 10. The system of claim 1, further comprising means for storing the generated data in cloud storage. [Explanation of symbols]

[1811] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for acquiring audio data; A means for converting the acquired voice data into text data; A means for analyzing the converted text data to generate minutes; A means for analyzing the converted text data and generating a to-do list; means for analyzing the converted text data to generate a schedule; means for analyzing the converted text data to generate a draft of an email; means for transmitting the generated data to a user terminal; A system including:

2. 10. The system of claim 1, further comprising means for notifying the user terminal of the transmitted data.

3. The system of claim 1 , further comprising means for storing the generated data in cloud storage.

Images