System

Abstract

A system is provided.SOLUTION: The system includes a means for analyzing the received e-mail, a means for automatically generating a reply sentence based on the analysis result, a means for checking the grammar and expression of the generated reply sentence, a means for automatically classifying the e-mail based on the analyzed contents, and a means for notifying the analysis result and the extracted date and time or correspondence items.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] Business people and office workers are often faced with a large volume of emails in their busy daily routines, which places a significant burden on their time. Tasks such as writing replies, checking grammar, sorting emails, and extracting important dates and times and corresponding items require a significant amount of effort when performed individually. Conventional email systems lack the functionality to automate these tasks, significantly reducing user efficiency. Therefore, there is a need for a way to resolve these issues and improve the efficiency and quality of email communication. [Means for solving the problem]

[0005] The system of the present invention solves the above-mentioned problems by utilizing the following means. First, by providing a means for analyzing received e-mails, the content of the e-mails can be accurately understood. Second, by providing a means for automatically generating replies based on the analysis results, the user can reduce the time it takes to manually create a reply. Furthermore, by providing a means for checking the grammar and expression of the generated replies and correcting them as necessary, the system ensures that the replies are professional and error-free. Furthermore, by providing a means for automatically classifying e-mails based on the analyzed content, the system facilitates e-mail management. Finally, by providing a means for notifying the user of the analysis results, the extracted date and time, and the necessary actions, the system ensures that important information is not overlooked and that appropriate actions can be taken. These means can significantly improve the efficiency of overall e-mail processing.

[0006] "Means for analyzing received e-mail" refers to a mechanism that analyzes the content of e-mail received on the user's device and extracts the text of the e-mail and its components (subject, body, sender, recipient, date, etc.).

[0007] "Means for automatically generating a reply based on the analysis results" refers to a system that uses AI technology to automatically create an appropriate reply based on the analyzed content of the email.

[0008] The "means for checking the grammar and expressions of the generated reply message" refers to a mechanism for checking the grammar and expressions of the automatically generated reply message and correcting them as necessary.

[0009] "Means for automatically classifying e-mails based on analyzed content" refers to a system that automatically sorts e-mails into predetermined categories based on the analyzed content.

[0010] "Means for notifying the analysis results, extracted dates and times, and relevant actions" refers to a mechanism for notifying the user's device of important dates and times and relevant actions extracted from the results of email analysis.

[0011] The "means for correcting grammatical and phrasal errors in the generated reply text" is a mechanism for automatically correcting grammatical and phrasal errors when they are detected.

[0012] The "means for recording the date, time, and action items as calendar events" is a mechanism for recording and managing the date, time, and action items extracted from emails in calendar format. [Brief explanation of the drawings]

[0013] [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

[0014] 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.

[0015] First, the terms used in the following description will be explained.

[0016] 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).

[0017] 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.

[0018] 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.

[0019] 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.

[0020] 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."

[0021] [First embodiment]

[0022] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0023] 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.

[0024] 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).

[0025] 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.

[0026] 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.

[0027] 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.

[0028] 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.

[0029] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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."

[0034] This invention is an AI email assistant system that supports efficient and effective email communication. This system consists of three main components: a server, a terminal, and a user.

[0035] Email parsing and reply generation

[0036] When the server receives an email sent from a user's device, it automatically analyzes its contents. This analysis extracts important information (e.g., sender, recipient, subject, body, date). Based on the extracted information, the server uses a generative AI to automatically generate an appropriate reply. The generative AI generates a reply in natural-sounding sentences based on the context and content of the email.

[0037] Example: When a user receives an email proposing a new project, the server parses the email and generates a reply like this:

[0038] Thank you for your continued support.

[0039] I have reviewed your new project proposal. It is very interesting and I would like to give it a positive consideration. Could you please suggest a date for a detailed meeting?

[0040] Thank you very much.

[0041] Grammar and Expression Check

[0042] The server checks the generated reply for grammar and expression. It uses a grammar engine to detect grammatical errors and unnatural expressions and corrects them as necessary. Once the reply has been corrected, it notifies the user.

[0043] Email classification and notifications

[0044] Based on the analysis results, the server automatically classifies emails into pre-defined categories, such as "Meetings / Conferences," "Projects," and "Other." Once classification is complete, the server extracts important dates and times and matters to be addressed, and notifies the user's device.

[0045] Example: When an invitation email for a meeting is received, the server classifies the email into the "Meeting / Conference" category, extracts the date and time of the meeting, and notifies the user.

[0046] Subject: (Meeting Notice) Information about a new project meeting

[0047] Notification content:

[0048] A new project meeting is scheduled for the following dates:

[0049] Date and time: October 25, 2023 14:00

[0050] Location: Conference Room 1

[0051] User confirmation and correction

[0052] The user can check the notification from the server and the generated reply message through the terminal. If necessary, the user can modify the reply message and issue a final transmission instruction to the server.

[0053] Sending a reply

[0054] When the user issues a sending instruction, the terminal sends the instruction to the server, which then sends the final reply to the recipient.

[0055] In this way, the AI ​​email assistant system of the present invention enables users to reply to received emails efficiently and effectively, significantly reducing the burden of email processing in their daily work.

[0056] The processing flow will be explained below.

[0057] Step 1:

[0058] The user receives an email on the terminal, which then sends the received email to the server.

[0059] Step 2:

[0060] The server analyzes the content of the email received, which includes extracting information such as the email text, sender, recipient, subject, and date.

[0061] Step 3:

[0062] Based on the analysis results, the server uses generative AI to automatically generate a reply. Specifically, it selects a reply template based on the content and then uses natural language generation technology to construct the reply.

[0063] Step 4:

[0064] The server performs a grammar and expression check on the generated reply, using a grammar engine to detect grammatical errors and unnatural expressions and correct them as necessary.

[0065] Step 5:

[0066] The server automatically classifies emails based on the analysis results, for example, into categories such as "Meetings / Conferences," "Projects," and "Other."

[0067] Step 6:

[0068] The server extracts information from emails and sends it to the user's device, including meeting schedules and important tasks.

[0069] Step 7:

[0070] The user checks the notification content from the server and the generated reply text on the terminal, and can modify the reply text if necessary.

[0071] Step 8:

[0072] The user issues a transmission instruction for the reply message after editing, and the terminal sends the instruction to the server.

[0073] Step 9:

[0074] The server then sends the final reply to the recipient, allowing users to communicate efficiently via email.

[0075] Example 1

[0076] 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."

[0077] In today's business environment, email is an important means of communication, but processing it requires a great deal of time and effort. Responding appropriately and promptly to the large volume of emails received is particularly difficult, hindering efficient business operations. Manually crafting replies can lead to grammatical errors and unnatural expressions. Important emails can be overlooked, or follow-ups can be delayed because they are not properly categorized. These issues need to be addressed.

[0078] 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.

[0079] In this invention, the server includes means for analyzing received messages, means for automatically generating a response based on the analysis results, means for checking the grammar and expressions of the generated response, means for automatically classifying messages based on the analyzed content, and means for notifying the user of the analysis results, the extracted date and time, and the corresponding matters. This automates a series of processes from receiving emails to generating a reply, checking grammar, automatic classification, and notifying the user of important information, enabling the user to process emails quickly and accurately.

[0080] "Means for parsing received messages" refers to the process by which the server receives messages sent from the user's terminal and disassembles their contents to extract information in a structured format.

[0081] "Means for automatically generating a response based on the analysis results" refers to the process of automatically creating a natural and appropriate reply using a generative AI model based on the analyzed information.

[0082] "Means for checking the grammar and expression of the generated response sentence" refers to the process of detecting and correcting grammatical errors and unnatural expressions in the generated reply sentence.

[0083] "Means for automatically categorizing messages based on analyzed content" refers to a process for categorizing messages into predefined categories (e.g., meetings, projects, etc.) based on analyzed information.

[0084] "Means for notifying the analysis results, extracted date and time, and corresponding matters" refers to the process of notifying the user of important information extracted during the analysis process (date and time, corresponding matters, etc.).

[0085] This invention provides an AI message assistant system that supports efficient and effective electronic message communication. The system consists of three main components: a server, a terminal, and a user.

[0086] Receiving and parsing messages

[0087] server

[0088] The server receives messages sent from the user's device and automatically analyzes their contents. This analysis uses natural language processing libraries (e.g., spaCy and NLTK) to extract important information such as the sender, recipient, subject, body, and date. The received message data is saved in a database, and the analyzed information is stored in temporary storage.

[0089] Automatic generation of response sentences

[0090] server

[0091] Based on the analyzed information, the server uses a generative AI model (e.g., OpenAI's GPT-4) to generate appropriate prompts. The generative AI model creates natural-sounding sentences based on the context and content.

[0092] A specific example is the following prompt:

[0093] Sender: Sender name

[0094] Recipient: Recipient name

[0095] Subject: The subject of the message

[0096] Body: The body of the message

[0097] Prompt: Generate a reply to the following message. Responses should be polite and helpful.

[0098] Checking grammar and expressions in responses

[0099] server

[0100] The generated response sentence is checked for grammar and expression using a grammar engine (e.g., Grammarly API). Grammatical errors and unnatural expressions are detected and corrected as necessary. Once corrected, the response sentence is saved back to storage.

[0101] Automatic message classification

[0102] server

[0103] Based on the analysis results, the server classifies messages into predefined categories, such as "Meetings," "Projects," and "Other." Machine learning models (e.g., support vector machines and random forests) are used for classification, and the classification results are stored in a database as message metadata.

[0104] Important Information Notification

[0105] server

[0106] Important information (e.g., date, time, and action) is extracted during the analysis process, and a notification containing this information is generated and sent to the user's device, where the user can visually confirm it.

[0107] Terminal

[0108] The user's device receives the notification sent from the server and displays it to the user, including the date and time, location of the meeting, and any action to be taken.

[0109] User verification and correction

[0110] User

[0111] The user can check the generated response and notification through his / her own terminal, and if necessary, correct the response and issue a final transmission instruction to the server.

[0112] Sending a response

[0113] server

[0114] When the user issues a final command to send the message, the server receives the command and sends the finalized response via the mail server, using the SMTP protocol.

[0115] In this way, the AI ​​message assistant system automates the entire process from receiving messages to generating responses, grammar checking, automatic classification, and notifying users of important information, making message handling more efficient.

[0116] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0117] Step 1:

[0118] Parsing received messages

[0119] Operation explanation

[0120] The server receives messages sent from the user's device. This process uses an email protocol (e.g., IMAP or POP3) and uses a natural language processing library (e.g., spaCy or NLTK) to analyze the content of the received messages.

[0121] input

[0122] Received message data

[0123] output

[0124] Parsed information (sender, recipient, subject, body, date)

[0125] Specific actions

[0126] The server receives new messages, analyzes their contents, extracts important information, and stores it in temporary storage.

[0127] Step 2:

[0128] Automatically generate a response

[0129] Operation explanation

[0130] The server automatically generates an appropriate response sentence using a generative AI model (e.g., OpenAI's GPT-4) based on the analyzed information. In this process, the prompt sentence is input into the generative AI model, which outputs the response sentence.

[0131] input

[0132] Parsed information (sender, recipient, subject, body, date)

[0133] output

[0134] Auto-generated response

[0135] Specific actions

[0136] The server passes the following prompt to the spawned AI:

[0137] From: John Doe

[0138] Recipient: Taro Yamada

[0139] Subject: New project proposal

[0140] Body: I'd like to propose an idea for a new project. I'd like to discuss the details at our next meeting.

[0141] Prompt: Generate a reply to the following message. Responses should be polite and helpful.

[0142] The generation AI then generates a response sentence and stores it in storage.

[0143] Step 3:

[0144] Check the grammar and expression of the generated response

[0145] Operation explanation

[0146] The server passes the generated response to a grammar engine (e.g., Grammarly API) to check the grammar and expressions. If grammatical errors or unnatural expressions are detected, they are corrected.

[0147] input

[0148] Generated response

[0149] output

[0150] Responses with grammar and expression corrections

[0151] Specific actions

[0152] The server sends the generated response sentence to the grammar engine, receives the corrected response sentence, and stores it in storage.

[0153] Step 4:

[0154] Automatically categorize messages

[0155] Operation explanation

[0156] Based on the analyzed information, the server classifies messages into predefined categories using machine learning models (e.g., SVM or Random Forest).

[0157] input

[0158] Analyzed information

[0159] output

[0160] Category information (e.g., meeting, project, etc.)

[0161] Specific actions

[0162] The server categorizes messages and stores the categorization results in a database as message metadata.

[0163] Step 5:

[0164] Notify the extracted date and time and the action to be taken

[0165] Operation explanation

[0166] Important dates and times and action items are extracted during the analysis process and notified to the user. In this notification process, notification data is generated and sent to the user's terminal.

[0167] input

[0168] Analysis results and extracted information

[0169] output

[0170] Notification Data

[0171] Specific actions

[0172] The server generates a notification including the date and time of the meeting and the matters to be addressed, and sends it to the user's terminal.

[0173] Step 6:

[0174] User confirmation and correction

[0175] Operation explanation

[0176] The user can check the notification and the generated response text through the terminal, modify the response text as necessary, and issue a final transmission instruction to the server.

[0177] input

[0178] Notification data, generated response

[0179] output

[0180] Revised Final Response

[0181] Specific actions

[0182] The user checks the response text and notification generated on the terminal, corrects the response text as necessary, and clicks the send button.

[0183] Step 7:

[0184] Sending a final response

[0185] Operation explanation

[0186] When the user issues a final sending instruction, the server receives the instruction and sends the final response via the mail server using the SMTP protocol.

[0187] input

[0188] Revised Final Response

[0189] output

[0190] Sent messages

[0191] Specific actions

[0192] The server sends a final response to the SMTP server and the message is delivered to the recipient.

[0193] In this way, each step works together in a specific way to achieve efficient message processing as a whole system.

[0194] (Application example 1)

[0195] 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."

[0196] Conventional email management systems make it difficult for users to respond to received emails promptly and appropriately. Furthermore, they lack the functionality to analyze email content, automatically generate a reply, and check the grammar of that reply, making it difficult to significantly reduce the amount of work required by users. Especially in workplaces where real-time responses are required, delays in these tasks can lead to reduced work efficiency and potentially inadequate customer service.

[0197] 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.

[0198] In this invention, the server includes a means for analyzing received emails, a means for automatically generating a reply based on the analysis results, and a means for checking the grammar and expressions of the generated reply. This allows the user to be notified of the analysis results, the extracted date and time, and the corresponding action, and to check the reply in real time via a display device. Furthermore, by generating a reply using a prompt sentence using a generative AI model, a more natural and appropriate response is automatically provided, significantly reducing the burden on the user and improving work efficiency.

[0199] "Means for analyzing" is a function for analyzing the contents of received e-mail and extracting important information.

[0200] "Means for automatically generating replies" refers to a function that automatically creates appropriate replies using a generative AI model based on the analyzed content.

[0201] The "means for checking grammar and expressions" is a function for detecting and correcting grammatical errors and unnatural expressions in the generated reply text.

[0202] "Means for automatically classifying e-mails" refers to a function that automatically classifies e-mails into pre-defined categories based on the analyzed content.

[0203] "Means for notifying" is a function that notifies the user of the analysis results, the date and time of extraction, and the necessary measures.

[0204] "Means for displaying a reply message in real time via a display device" refers to a function for displaying the generated reply message to the user in real time using a display device (e.g., a head-mounted display).

[0205] A "generative AI model" is an artificial intelligence model used to generate natural-sounding sentences, taking a prompt as input and outputting an appropriate reply.

[0206] A "prompt sentence" is an input sentence given to a generative AI model, based on which a reply sentence is generated.

[0207] "Means for recording as a calendar event" is a function for recording and managing the analyzed date and time and corresponding matters as a calendar event in the system.

[0208] This invention is an AI email assistant system that enables store staff in brick-and-mortar stores to efficiently handle email reservations and inquiries from customers. The system analyzes received emails, automatically generates appropriate replies, and checks their contents. It also notifies users of the analysis results, the extracted date and time, and the necessary actions, and displays the replies in real time on a display device.

[0209] Hardware and software used

[0210] Hardware: Head-mounted display worn by the store clerk (e.g., Microsoft® HoloLens®)

[0211] software:

[0212] Generative AI model: A model that generates natural language using TENSORFLOW®

[0213] Web server: Django framework

[0214] HMD Control: Uses OpenCV to manage display and notifications

[0215] System action

[0216] 1. Email analysis

[0217] The server has the means to parse the content of received emails, extracting important information such as sender, recipient, subject, body, date, etc.

[0218] 2. Automatically generate replies

[0219] Based on the analysis results, a generative AI model is used to automatically generate a reply, taking the prompt as input and generating an appropriate reply.

[0220] An example prompt is:

[0221] A customer is asking: Please make an appointment for October 25, 2023 at 2 PM. Please generate a suggested response.

[0222] 3. Grammar and Expression Check

[0223] The generated reply text is checked by a grammar engine on the server, and any errors or unnatural expressions detected are corrected.

[0224] 4. Email Classification and Notification

[0225] Based on the analysis results, emails are automatically categorized into predefined categories such as "reservations" and "inquiries," and important dates and times and actions are extracted.

[0226] 5. Real-time notifications and displays

[0227] The analysis results and reply text are sent to the store clerk in real time via the HMD, allowing the clerk to immediately check the reply and take action if necessary.

[0228] Specific examples

[0229] When a customer emails a store with a reservation, the server parses the email and generates an appropriate reply, such as:

[0230] Thank you for your help. We have accepted your reservation for October 25, 2023 at 2:00 PM. Thank you in advance.

[0231] This reply is displayed in real time on the HMD, allowing the store clerk to check its contents and reply to the customer.

[0232] In this way, the AI ​​email assistant system of the present invention enables store staff to efficiently and effectively handle customer inquiries, which is expected to significantly reduce the burden of email processing in daily work and improve work efficiency.

[0233] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0234] Step 1:

[0235] The server receives emails received from the user's device. To analyze the content of these emails, it extracts important information such as the sender, recipient, subject, body, date, etc. The email is the input, and the extracted information is the output.

[0236] Step 2:

[0237] The server automatically generates a reply based on the analyzed content. Using a generative AI model, it generates an appropriate reply using the prompt as input. Specifically, the prompt is input into the generative AI model, which outputs a natural reply. For example, the following prompt is input:

[0238] A customer is asking: Please make an appointment for October 25, 2023 at 2 PM. Please generate a suggested response.

[0239] The generative AI model outputs an appropriate reply based on this prompt.

[0240] Step 3:

[0241] The server performs grammar and expression checks on the generated reply. It uses a grammar engine to detect grammatical errors and unnatural expressions in the reply and corrects them as necessary. The input is the generated reply, and the output is the corrected or verified reply.

[0242] Step 4:

[0243] The server automatically classifies emails based on the analysis results. It sorts emails into appropriate folders according to predefined categories (e.g., "reservations" or "inquiries"). The input is the analyzed email information, and the output is the classified email.

[0244] Step 5:

[0245] The server notifies the user's device of the analysis results, the extracted date and time, and the corresponding action. Furthermore, this information, including the reply, is displayed in real time on a head-mounted display (HMD). The input is the analysis results and the reply, and the output is the notification and reply displayed on the HMD.

[0246] Step 6:

[0247] The user can check the reply message through the HMD and make corrections if necessary. If corrections are made, the information is sent back to the server, and the final reply message is confirmed. The input is the generated and corrected reply message, and the output is the final reply message.

[0248] Step 7:

[0249] When the server receives the final sending instruction from the user, it sends a confirmed reply to the customer. The input is the final reply, and the output is the completion of the email sending.

[0250] Step 8:

[0251] The server records the analysis results and notifications as a calendar in the system, and can review them later if needed. The input is important dates and times and actions, and the output is a record of the calendar events.

[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 AI email assistant system that supports efficient and effective email communication and combines it with an emotion engine that recognizes the user's emotions. This system consists of three main components: a server, a terminal, and a user.

[0254] Email parsing and reply generation

[0255] When the server receives an email sent from a user's device, it automatically analyzes its contents. This analysis extracts important information (e.g., sender, recipient, subject, body, date). Based on the extracted information, the server uses a generative AI to automatically generate an appropriate reply. The generative AI generates a reply in natural-sounding sentences based on the context and content of the email.

[0256] Example: When a user receives an email proposing a new project, the server parses the email and generates a reply like this:

[0257] Thank you for your continued support.

[0258] I have reviewed your new project proposal. It is very interesting and I would like to give it a positive consideration. Could you please suggest a date for a detailed meeting?

[0259] Thank you very much.

[0260] Use of emotion engine

[0261] During analysis, the server uses an emotion engine to recognize emotions based on the content of the email received by the user and the user's response. The recognized emotions are reflected in the reply generation process, allowing for the generation of natural-looking replies that take the user's emotional state into account.

[0262] Grammar and Expression Check

[0263] The server checks the generated reply for grammar and expression. It uses a grammar engine to detect grammatical errors and unnatural expressions and corrects them as necessary. Once the reply has been corrected, it notifies the user.

[0264] Email classification and notifications

[0265] Based on the analysis results, the server automatically classifies emails into pre-defined categories. For example, categories such as "Meetings / Conferences," "Projects," and "Other" can be set. Once classification is complete, the server extracts important dates and times and matters to be addressed, and notifies the user's device of these. Using an emotion engine, the content and method of notifications can be adjusted based on the user's emotions.

[0266] Example: When an invitation email for a meeting is received, the server classifies the email into the "Meeting / Conference" category, extracts the date and time of the meeting, and notifies the user.

[0267] Subject: (Meeting Notice) Information about a new project meeting

[0268] Notification content:

[0269] A new project meeting is scheduled for the following dates:

[0270] Date and time: October 25, 2023 14:00

[0271] Location: Conference Room 1

[0272] User confirmation and correction

[0273] The user can check the notification from the server and the generated reply message through the terminal. If necessary, the user can modify the reply message and issue a final transmission instruction to the server.

[0274] Sending a reply

[0275] When the user issues a sending instruction, the terminal sends the instruction to the server, which then sends the final reply to the recipient.

[0276] In this way, the AI ​​email assistant system of the present invention can efficiently and appropriately reply to emails received by the user. Furthermore, by utilizing the emotion engine, natural replies that reflect the user's emotions are possible, improving the efficiency and quality of the entire email process.

[0277] The processing flow will be explained below.

[0278] Step 1:

[0279] The user receives an email on the terminal, which then sends the received email to the server.

[0280] Step 2:

[0281] The server analyzes the content of the email received, extracting information such as the email text, sender, recipient, subject, and date.

[0282] Step 3:

[0283] The server uses an emotion engine to recognize the user's emotions, which are determined based on the content of the email and the user's past email history.

[0284] Step 4:

[0285] The server automatically generates a reply using generative AI based on the analysis results and the recognized emotions. The generative AI generates natural-sounding sentences taking into account the context and the user's emotions.

[0286] Example: If a user receives an email proposing a new project and the emotion engine recognizes that the user is "interested," the server generates a reply like this:

[0287] Thank you for your continued support.

[0288] I have reviewed your new project proposal. It is very interesting and I would like to give it a positive consideration. Could you please suggest a date for a detailed meeting?

[0289] Thank you very much.

[0290] Step 5:

[0291] The server performs a grammar and expression check on the generated reply, using a grammar engine to detect grammatical errors and unnatural expressions and correct them as necessary.

[0292] Step 6:

[0293] The server automatically classifies emails based on the analysis results, for example, into categories such as "Meetings / Conferences," "Projects," and "Other."

[0294] Step 7:

[0295] The server notifies the user's device of the date and time and action items extracted from the email. Notifications include meeting schedules and important tasks. The emotion engine adjusts the content and method of notifications based on the user's emotions.

[0296] Example: When a user receives an invitation email for a meeting and the emotion engine recognizes "tension," the server notifies the user of the meeting schedule in detail.

[0297] Subject: (Meeting Notice) Information about a new project meeting

[0298] Notification content:

[0299] A new project meeting is scheduled for the following dates:

[0300] Date and time: October 25, 2023 14:00

[0301] Location: Conference Room 1

[0302] Thank you for your preparation.

[0303] Step 8:

[0304] The user checks the notification content from the server and the generated reply text on the terminal, and can modify the reply text if necessary.

[0305] Step 9:

[0306] The user issues a transmission instruction for the reply message after editing, and the terminal sends the instruction to the server.

[0307] Step 10:

[0308] The server then sends the final reply to the recipient, allowing users to communicate efficiently via email.

[0309] Example 2

[0310] 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."

[0311] In conventional email processing systems, the content analysis of received emails, automatic generation of reply messages, grammar and expression checks, and email classification and notification are all performed separately, resulting in problems with consistency and accuracy. Furthermore, reply messages do not reflect the user's feelings, making it difficult to communicate naturally and appropriately.

[0312] 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 analyzing received emails, means for automatically generating reply messages based on the analysis results, means for recognizing the sentiment of the generated reply messages, means for checking grammar and expressions, means for automatically classifying emails based on the analyzed content, and means for notifying the analysis results, extracted date and time, and corresponding matters. This improves the consistency and accuracy of email processing and enables natural replies that reflect the user's sentiment.

[0313] The "means for analyzing received e-mail" is a means for automatically analyzing the contents of received e-mail and extracting important information.

[0314] The "means for automatically generating a reply message based on the analysis results" is a means for automatically generating an appropriate reply message based on the analyzed information.

[0315] The "means for recognizing the emotion of the generated reply message" is a means for analyzing and recognizing the emotion contained in the content of the generated reply message.

[0316] The "means for checking grammar and expressions" refers to a means for checking and correcting grammatical errors and unnatural expressions in the generated reply text.

[0317] The "means for automatically classifying e-mails based on the analyzed content" is a means for automatically classifying e-mails into pre-defined categories based on the analyzed content.

[0318] The "means for notifying the analysis results, extracted dates and times, and corresponding matters" refers to a means for notifying the user of the analysis results and important dates and times and corresponding matters extracted from them.

[0319] This invention is an AI email assistant system that supports efficient and effective email communication and combines it with an emotion engine that recognizes the user's emotions. This system consists of three main components: a server, a terminal, and a user.

[0320] Email parsing and reply generation

[0321] When the server receives an email sent from a user's device, it automatically analyzes the content of the email using NLP (Natural Language Processing) algorithms, particularly tools such as Google® Cloud Natural Language API, to extract important information such as the sender, recipient, subject, body, and date.

[0322] Based on the extracted information, the server automatically generates an appropriate reply using a generative AI (e.g., OpenAI's GPT-3 (registered trademark)). The generative AI generates a reply in natural-sounding sentences based on the context and content of the email.

[0323] As a concrete example, when a user receives an email proposing a new project, the following prompt sentence is input to the generative AI model:

[0324] Generate a sample project proposal email response using the information below.

[0325] Sender: Taro Tanaka

[0326] Recipient: Ichiro Suzuki

[0327] Subject: New Project Proposal

[0328] Body: I ​​would like to propose a new project.

[0329] Date: October 10, 2023

[0330] Based on this prompt, the following reply will be generated:

[0331] Thank you for your continued support.

[0332] I have reviewed your new project proposal. It is very interesting and I would like to give it a positive consideration. Could you please suggest a date for a detailed meeting?

[0333] Thank you very much.

[0334] Use of emotion engine

[0335] The server uses an emotion engine for analysis. This emotion engine uses, for example, the emotion analysis function of Microsoft Azure's (registered trademark) Text Analytics API. The emotion engine recognizes emotions based on the content of the email received by the user and the user's response. This recognized emotion is reflected in the reply generation process. This allows for the generation of natural-looking replies that take the user's emotional state into consideration.

[0336] Grammar and Expression Check

[0337] The server then checks the generated reply for grammar and expression. Specifically, it uses a grammar engine such as the Grammarly API to detect grammatical errors and unnatural expressions and corrects them as necessary. Once the corrections are complete, the user is notified.

[0338] Email classification and notifications

[0339] The server automatically classifies emails into predefined categories based on the analysis results. For example, a machine learning model (trained with scikit-learn) can be used. Once classification is complete, the server extracts important dates and times and actions to take and notifies the user's device. An emotion engine can be used to adjust the content and method of notifications based on the user's emotions.

[0340] For example, when a user receives an invitation email for a meeting, the server classifies the email into the "Meeting / Conference" category, extracts the date and time of the meeting, and notifies the user:

[0341] Subject: (Meeting Notice) Information about a new project meeting

[0342] Notification content:

[0343] A new project meeting is scheduled for the following dates:

[0344] Date and time: October 25, 2023 14:00

[0345] Location: Conference Room 1

[0346] User confirmation and correction

[0347] The user can check the notification from the server and the generated reply message through the terminal. If the reply message is not appropriate, the user can correct it and issue a final transmission instruction to the server.

[0348] Sending a reply

[0349] When a user issues a sending instruction, the device sends the instruction to the server, which then sends the final reply to the recipient. Through this series of processes, the AI ​​email assistant system efficiently and effectively supports users' email communication.

[0350] The above is a specific embodiment for carrying out the present invention. The combination of the generative AI model and emotion engine, which is a feature of this invention, significantly improves the efficiency and quality of email processing.

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

[0352] Step 1:

[0353] The server receives emails sent from the user's terminal. The input is the email information received from the terminal, and the output is the received email data. The server performs the specific operation of retrieving new emails from the email server using the IMAP or POP3 protocol.

[0354] Step 2:

[0355] The server analyzes the content of received emails. The input is the received email data, and the output is the analysis results (e.g., sender, recipient, subject, body, date). NLP algorithms are used for the analysis, and specific operations are performed using, for example, the Google Cloud Natural Language API.

[0356] Step 3:

[0357] The server generates a reply using a generative AI model (e.g., OpenAI's GPT-3) based on the analyzed email content. The input is the analysis result, and the output is the generated reply. Specifically, the following prompt is input to the generative AI model:

[0358] Generate a sample project proposal email response using the information below.

[0359] Sender: Taro Tanaka

[0360] Recipient: Ichiro Suzuki

[0361] Subject: New Project Proposal

[0362] Body: I ​​would like to propose a new project.

[0363] Date: October 10, 2023

[0364] Step 4:

[0365] The server uses an emotion engine to recognize emotions from the generated reply text. The input is the generated reply text, and the output is emotion data. Specifically, the server uses Microsoft Azure's Text Analytics API.

[0366] Step 5:

[0367] The server adjusts the naturalness and appropriateness of the reply based on the emotional data. The input is the emotional data, and the output is a modified reply that reflects the emotion. In this step, the tone and expression of the document generated by the generative AI model are changed to match the emotion.

[0368] Step 6:

[0369] The server checks the generated reply for grammar and expression. The input is the generated reply, and the output is the grammar- and expression-checked reply. Specifically, it uses the Grammarly API to detect and correct grammatical errors and unnatural expressions.

[0370] Step 7:

[0371] The server automatically classifies emails based on the analysis results. The input is the analysis results, and the output is the classified emails. Specifically, it uses a machine learning model (trained using scikit-learn) to classify emails into pre-defined categories.

[0372] Step 8:

[0373] The server notifies the user's device of the analysis results, the extracted date and time, and the necessary actions. The input is the analysis results and extracted important information, and the output is notification information. Using an emotion engine, the content and method of notification are adjusted based on the user's emotions.

[0374] Step 9:

[0375] The user checks the notification from the server and the generated reply through the terminal. The input is the notification from the server and the generated reply, and the output is the reply that has been checked and corrected by the user. The user corrects the reply as necessary and performs the specific action of issuing a final transmission instruction to the server.

[0376] Step 10:

[0377] The terminal transmits the user's transmission instruction to the server. The input is the user's transmission instruction, and the output is the transmission instruction data to the server.

[0378] Step 11:

[0379] The server sends the final reply to the recipient. The input is the revised final reply, and the output is the result sent to the recipient. Through this series of processes, the AI ​​email assistant system efficiently and effectively supports users' email communication.

[0380] (Application example 2)

[0381] 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."

[0382] There is a need for systems that can reduce the burden of responding to large volumes of inquiries via e-mail and other electronic communications, while also providing appropriate replies based on emotions. However, conventional systems have difficulty taking emotions into account, making it difficult to respond in a way that is sensitive to the user's feelings. It is also difficult to respond quickly while ensuring the appropriateness of grammar and expression in replies.

[0383] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing received electronic communications, means for automatically generating a reply message based on the analysis results, means for checking the grammar and expressions of the generated reply message, means for automatically classifying the electronic communications based on the analyzed content, means for notifying the analysis results and the extracted date and time and response matters, means for using an emotion engine for emotion analysis, and means for generating a reply message based on the emotion. This enables natural replies that are in line with the user's emotions and improves the efficiency and quality of inquiry responses.

[0384] "Means for analyzing received electronic communications" refers to the function of devices such as servers and terminals to analyze the contents of received emails and messages and extract important information.

[0385] "Means for automatically generating a reply based on the analysis results" refers to technology for automatically generating an appropriate reply based on the analyzed data.

[0386] "Means for checking the grammar and expressions of generated replies" refers to a function for detecting and correcting grammatical errors and unnatural expressions in automatically generated replies.

[0387] "Means for automatically classifying electronic communications based on analyzed content" refers to a function that automatically sorts emails and messages into specific categories based on the results of the analysis.

[0388] "Means for notifying the analysis results, extracted dates and times, and action items" refers to a notification function for informing the user of the analysis results, important dates, and action items.

[0389] "Means of using an emotion engine to perform emotion analysis" is a technology for recognizing and analyzing emotions from the text of emails and messages.

[0390] The "means for generating a reply based on emotion" is a technology for automatically generating a reply using expressions and writing styles that correspond to the recognized emotion.

[0391] MODE FOR CARRYING OUT THE INVENTION

[0392] The system for implementing the present invention is composed of three main components: a server, a terminal, and a user. A specific embodiment of the system is shown below.

[0393] Email parsing and reply generation

[0394] When the server receives an electronic communication sent from a user's device, it analyzes the content of the communication using natural language processing (NLP) technology. The analysis is performed with the aim of extracting important information from the electronic communication (e.g., sender, recipient, subject, body, date). Based on the extracted information, the server uses generative AI (e.g., GPT-4) to automatically generate an appropriate reply.

[0395] Example: When a user receives an inquiry email stating that "the product has not arrived," the server analyzes the email and generates a reply message like the following:

[0396] We are very sorry. We checked the tracking information for your order and it was shipped on XX date, but we will reconfirm the delivery status and respond accordingly.

[0397] Use of emotion engine

[0398] During analysis, the server uses an emotion engine (e.g., Google Cloud Natural Language API) to recognize emotions based on the content of the email received by the user and the user's response. The recognized emotions are reflected in the reply generation process. This allows for the generation of natural-looking replies that take the user's emotional state into account.

[0399] Example: For an angry customer, the server uses more polite and careful language. An example of a prompt for the AI ​​generation is, "Read the email about the customer's angry delivery of the item and generate a reply that includes an apology and reconfirmation."

[0400] Grammar and Expression Check

[0401] The server uses a grammar checker (e.g., Grammarly API) to detect grammatical errors and unnatural expressions in the generated reply, correcting them as necessary, and notifying the user of the completed reply.

[0402] Email classification and notifications

[0403] Based on the analysis results, the server automatically classifies electronic communications into pre-defined categories. For example, categories such as "Meetings / Consultations," "Projects," and "Other" may be set. Once classification is complete, the server extracts important dates and times and matters to be addressed, and notifies the user of these on their device. Using an emotion engine, the content and method of notifications can be adjusted based on the user's emotions.

[0404] Example: When an invitation email for a meeting is received, the server classifies the email into the "Meeting / Conference" category, extracts the date and time of the meeting, and notifies the user.

[0405] Subject: (Meeting Notice) Information about a new project meeting

[0406] Notification content:

[0407] A new project meeting is scheduled for the following dates:

[0408] Date and time: October 25, 2023 14:00

[0409] Location: Conference Room 1

[0410] User confirmation and correction

[0411] The user can check the notification from the server and the generated reply message through the terminal. If necessary, the user can modify the reply message and issue a final transmission instruction to the server.

[0412] Sending a reply

[0413] When the user issues a sending instruction, the terminal sends the instruction to the server, which then sends the final reply to the recipient.

[0414] In this way, the system of the present invention allows users to efficiently and appropriately reply to electronic communications received. Furthermore, by utilizing the emotion engine, natural replies that reflect the user's emotions are possible, thereby improving the efficiency and quality of the entire email process.

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

[0416] Step 1:

[0417] The server receives the incoming electronic communication. The input is an email or message sent from a user terminal. The server converts the received communication into a format suitable for analysis. The output of this step is text data that is used for analysis.

[0418] Step 2:

[0419] The server analyzes the content of the received electronic communication. The input is the text data generated in step 1. The server uses natural language processing (NLP) techniques to extract important information (sender, recipient, subject, body, date, etc.). Specifically, it uses a text analysis library (e.g., spaCy) for parsing and information extraction. The output of this step is a set of extracted important information.

[0420] Step 3:

[0421] The server uses an emotion engine to recognize emotions contained in the extracted information. The input is the information extracted in step 2. The emotion engine (e.g., Google Cloud Natural Language API) is used to parse emotions from the email body. The output of this step is the recognized emotion data.

[0422] Step 4:

[0423] The server automatically generates a reply based on the analysis results. The input is the set of important information extracted in step 2 and the emotion data obtained in step 3. The server uses a generative AI model (e.g., GPT-4) to set the prompt as follows:

[0424] A customer receives an email saying they didn't receive their item. You recognize their anger. Generate a response based on this information, including an apology and reassurance.

[0425] The output of this step is the generated reply.

[0426] Step 5:

[0427] The server checks the grammar and language of the generated reply. The input is the reply generated in step 4. A grammar checker (e.g., Grammarly API) is used to detect grammatical errors and unnatural language and correct them if necessary. The output of this step is the reply whose grammar and language have been checked and corrected.

[0428] Step 6:

[0429] The server automatically classifies the electronic communications based on the analysis results. The input is the information extracted in step 2. The server performs classification based on pre-defined categories (e.g., "meetings / discussions," "projects," "others"). The output of this step is a set of classified electronic communications.

[0430] Step 7:

[0431] The server notifies the analysis results, extracted dates and times, and actions to be taken. The input is the information extracted in step 2 and the classification results in step 6. The server extracts important dates and times and actions to be taken, and sends a notification to the user device. The output of this step is the notification sent to the user device.

[0432] Step 8:

[0433] The user checks the notification from the server and the generated reply through the terminal. The input is the notification sent in step 7 and the reply corrected in step 5. The user checks the reply and corrects it if necessary. The output of this step is the checked and corrected reply.

[0434] Step 9:

[0435] The user issues a sending instruction, and the terminal sends that instruction to the server. The input is the reply message that was checked and corrected in step 8. The server sends the final reply message to the recipient. The output of this step is the sent reply message.

[0436] 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.

[0437] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0438] 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.

[0439] [Second embodiment]

[0440] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0441] 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.

[0442] 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).

[0443] 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.

[0444] 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.

[0445] 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).

[0446] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0447] 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.

[0448] 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.

[0449] 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.

[0450] 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.

[0451] 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."

[0452] This invention is an AI email assistant system that supports efficient and effective email communication. This system consists of three main components: a server, a terminal, and a user.

[0453] Email parsing and reply generation

[0454] When the server receives an email sent from a user's device, it automatically analyzes its contents. This analysis extracts important information (e.g., sender, recipient, subject, body, date). Based on the extracted information, the server uses a generative AI to automatically generate an appropriate reply. The generative AI generates a reply in natural-sounding sentences based on the context and content of the email.

[0455] Example: When a user receives an email proposing a new project, the server parses the email and generates a reply like this:

[0456] Thank you for your continued support.

[0457] I have reviewed your new project proposal. It is very interesting and I would like to give it a positive consideration. Could you please suggest a date for a detailed meeting?

[0458] Thank you very much.

[0459] Grammar and Expression Check

[0460] The server checks the generated reply for grammar and expression. It uses a grammar engine to detect grammatical errors and unnatural expressions and corrects them as necessary. Once the reply has been corrected, it notifies the user.

[0461] Email classification and notifications

[0462] Based on the analysis results, the server automatically classifies emails into pre-defined categories, such as "Meetings / Conferences," "Projects," and "Other." Once classification is complete, the server extracts important dates and times and matters to be addressed, and notifies the user's device.

[0463] Example: When an invitation email for a meeting is received, the server classifies the email into the "Meeting / Conference" category, extracts the date and time of the meeting, and notifies the user.

[0464] Subject: (Meeting Notice) Information about a new project meeting

[0465] Notification content:

[0466] A new project meeting is scheduled for the following dates:

[0467] Date and time: October 25, 2023 14:00

[0468] Location: Conference Room 1

[0469] User confirmation and correction

[0470] The user can check the notification from the server and the generated reply message through the terminal. If necessary, the user can modify the reply message and issue a final transmission instruction to the server.

[0471] Sending a reply

[0472] When the user issues a sending instruction, the terminal sends the instruction to the server, which then sends the final reply to the recipient.

[0473] In this way, the AI ​​email assistant system of the present invention enables users to reply to received emails efficiently and effectively, significantly reducing the burden of email processing in their daily work.

[0474] The processing flow will be explained below.

[0475] Step 1:

[0476] The user receives an email on the terminal, which then sends the received email to the server.

[0477] Step 2:

[0478] The server analyzes the content of the email received, which includes extracting information such as the email text, sender, recipient, subject, and date.

[0479] Step 3:

[0480] Based on the analysis results, the server uses generative AI to automatically generate a reply. Specifically, it selects a reply template based on the content and then uses natural language generation technology to construct the reply.

[0481] Step 4:

[0482] The server performs a grammar and expression check on the generated reply, using a grammar engine to detect grammatical errors and unnatural expressions and correct them as necessary.

[0483] Step 5:

[0484] The server automatically classifies emails based on the analysis results, for example, into categories such as "Meetings / Conferences," "Projects," and "Other."

[0485] Step 6:

[0486] The server extracts information from emails and sends it to the user's device, including meeting schedules and important tasks.

[0487] Step 7:

[0488] The user checks the notification content from the server and the generated reply text on the terminal, and can modify the reply text if necessary.

[0489] Step 8:

[0490] The user issues a transmission instruction for the reply message after editing, and the terminal sends the instruction to the server.

[0491] Step 9:

[0492] The server then sends the final reply to the recipient, allowing users to communicate efficiently via email.

[0493] Example 1

[0494] 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."

[0495] In today's business environment, email is an important means of communication, but processing it requires a great deal of time and effort. Responding appropriately and promptly to the large volume of emails received is particularly difficult, hindering efficient business operations. Manually crafting replies can lead to grammatical errors and unnatural expressions. Important emails can be overlooked, or follow-ups can be delayed because they are not properly categorized. These issues need to be addressed.

[0496] 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.

[0497] In this invention, the server includes means for analyzing received messages, means for automatically generating a response based on the analysis results, means for checking the grammar and expressions of the generated response, means for automatically classifying messages based on the analyzed content, and means for notifying the user of the analysis results, the extracted date and time, and the corresponding matters. This automates a series of processes from receiving emails to generating a reply, checking grammar, automatic classification, and notifying the user of important information, enabling the user to process emails quickly and accurately.

[0498] "Means for parsing received messages" refers to the process by which the server receives messages sent from the user's terminal and disassembles their contents to extract information in a structured format.

[0499] "Means for automatically generating a response based on the analysis results" refers to the process of automatically creating a natural and appropriate reply using a generative AI model based on the analyzed information.

[0500] "Means for checking the grammar and expression of the generated response sentence" refers to the process of detecting and correcting grammatical errors and unnatural expressions in the generated reply sentence.

[0501] "Means for automatically categorizing messages based on analyzed content" refers to a process for categorizing messages into predefined categories (e.g., meetings, projects, etc.) based on analyzed information.

[0502] "Means for notifying the analysis results, extracted date and time, and corresponding matters" refers to the process of notifying the user of important information extracted during the analysis process (date and time, corresponding matters, etc.).

[0503] This invention provides an AI message assistant system that supports efficient and effective electronic message communication. The system consists of three main components: a server, a terminal, and a user.

[0504] Receiving and parsing messages

[0505] server

[0506] The server receives messages sent from the user's device and automatically analyzes their contents. This analysis uses natural language processing libraries (e.g., spaCy and NLTK) to extract important information such as the sender, recipient, subject, body, and date. The received message data is saved in a database, and the analyzed information is stored in temporary storage.

[0507] Automatic generation of response sentences

[0508] server

[0509] Based on the analyzed information, the server uses a generative AI model (e.g., OpenAI's GPT-4) to generate appropriate prompts, which create natural-sounding sentences based on context and content.

[0510] A specific example is the following prompt:

[0511] Sender: Sender name

[0512] Recipient: Recipient name

[0513] Subject: The subject of the message

[0514] Body: The body of the message

[0515] Prompt: Generate a reply to the following message. Responses should be polite and helpful.

[0516] Checking grammar and expressions in responses

[0517] server

[0518] The generated response sentence is checked for grammar and expression using a grammar engine (e.g., Grammarly API). Grammatical errors and unnatural expressions are detected and corrected as necessary. Once corrected, the response sentence is saved back to storage.

[0519] Automatic message classification

[0520] server

[0521] Based on the analysis results, the server classifies messages into predefined categories, such as "Meetings," "Projects," and "Other." Machine learning models (e.g., support vector machines and random forests) are used for classification, and the classification results are stored in a database as message metadata.

[0522] Important Information Notification

[0523] server

[0524] Important information (e.g., date, time, and action) is extracted during the analysis process, and a notification containing this information is generated and sent to the user's device, where the user can visually confirm it.

[0525] Terminal

[0526] The user's device receives the notification sent from the server and displays it to the user, including the date and time, location of the meeting, and any action to be taken.

[0527] User verification and correction

[0528] User

[0529] The user can check the generated response and notification through his / her own terminal, and if necessary, correct the response and issue a final transmission instruction to the server.

[0530] Sending a response

[0531] server

[0532] When the user issues a final command to send the message, the server receives the command and sends the finalized response via the mail server, using the SMTP protocol.

[0533] In this way, the AI ​​message assistant system automates the entire process from receiving messages to generating responses, grammar checking, automatic classification, and notifying users of important information, making message handling more efficient.

[0534] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0535] Step 1:

[0536] Parsing received messages

[0537] Operation explanation

[0538] The server receives messages sent from the user's device. This process uses an email protocol (e.g., IMAP or POP3) and uses a natural language processing library (e.g., spaCy or NLTK) to analyze the content of the received messages.

[0539] input

[0540] Received message data

[0541] output

[0542] Parsed information (sender, recipient, subject, body, date)

[0543] Specific actions

[0544] The server receives new messages, analyzes their contents, extracts important information, and stores it in temporary storage.

[0545] Step 2:

[0546] Automatically generate a response

[0547] Operation explanation

[0548] The server automatically generates an appropriate response sentence using a generative AI model (e.g., OpenAI's GPT-4) based on the analyzed information. In this process, the prompt sentence is input into the generative AI model, which outputs the response sentence.

[0549] input

[0550] Parsed information (sender, recipient, subject, body, date)

[0551] output

[0552] Auto-generated response

[0553] Specific actions

[0554] The server passes the following prompt to the spawned AI:

[0555] From: John Doe

[0556] Recipient: Taro Yamada

[0557] Subject: New project proposal

[0558] Body: I'd like to propose an idea for a new project. I'd like to discuss the details at our next meeting.

[0559] Prompt: Generate a reply to the following message. Responses should be polite and helpful.

[0560] The generation AI then generates a response sentence and stores it in storage.

[0561] Step 3:

[0562] Check the grammar and expression of the generated response

[0563] Operation explanation

[0564] The server passes the generated response to a grammar engine (e.g., Grammarly API) to check the grammar and expressions. If grammatical errors or unnatural expressions are detected, they are corrected.

[0565] input

[0566] Generated response

[0567] output

[0568] Responses with grammar and expression corrections

[0569] Specific actions

[0570] The server sends the generated response sentence to the grammar engine, receives the corrected response sentence, and stores it in storage.

[0571] Step 4:

[0572] Automatically categorize messages

[0573] Operation explanation

[0574] Based on the analyzed information, the server classifies messages into predefined categories using machine learning models (e.g., SVM or Random Forest).

[0575] input

[0576] Analyzed information

[0577] output

[0578] Category information (e.g., meeting, project, etc.)

[0579] Specific actions

[0580] The server categorizes messages and stores the categorization results in a database as message metadata.

[0581] Step 5:

[0582] Notify the extracted date and time and the action to be taken

[0583] Operation explanation

[0584] Important dates and times and action items are extracted during the analysis process and notified to the user. In this notification process, notification data is generated and sent to the user's terminal.

[0585] input

[0586] Analysis results and extracted information

[0587] output

[0588] Notification Data

[0589] Specific actions

[0590] The server generates a notification including the date and time of the meeting and the matters to be addressed, and sends it to the user's terminal.

[0591] Step 6:

[0592] User confirmation and correction

[0593] Operation explanation

[0594] The user can check the notification and the generated response text through the terminal, modify the response text as necessary, and issue a final transmission instruction to the server.

[0595] input

[0596] Notification data, generated response

[0597] output

[0598] Revised Final Response

[0599] Specific actions

[0600] The user checks the response text and notification generated on the terminal, corrects the response text as necessary, and clicks the send button.

[0601] Step 7:

[0602] Sending a final response

[0603] Operation explanation

[0604] When the user issues a final sending instruction, the server receives the instruction and sends the final response via the mail server using the SMTP protocol.

[0605] input

[0606] Revised Final Response

[0607] output

[0608] Sent messages

[0609] Specific actions

[0610] The server sends a final response to the SMTP server and the message is delivered to the recipient.

[0611] In this way, each step works together in a specific way to achieve efficient message processing as a whole system.

[0612] (Application example 1)

[0613] 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."

[0614] Conventional email management systems make it difficult for users to respond to received emails promptly and appropriately. Furthermore, they lack the functionality to analyze email content, automatically generate a reply, and check the grammar of that reply, making it difficult to significantly reduce the amount of work required by users. Especially in workplaces where real-time responses are required, delays in these tasks can lead to reduced work efficiency and potentially inadequate customer service.

[0615] 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.

[0616] In this invention, the server includes a means for analyzing received emails, a means for automatically generating a reply based on the analysis results, and a means for checking the grammar and expressions of the generated reply. This allows the user to be notified of the analysis results, the extracted date and time, and the corresponding action, and to check the reply in real time via a display device. Furthermore, by generating a reply using a prompt sentence using a generative AI model, a more natural and appropriate response is automatically provided, significantly reducing the burden on the user and improving work efficiency.

[0617] "Means for analyzing" is a function for analyzing the contents of received e-mail and extracting important information.

[0618] "Means for automatically generating replies" refers to a function that automatically creates appropriate replies using a generative AI model based on the analyzed content.

[0619] The "means for checking grammar and expressions" is a function for detecting and correcting grammatical errors and unnatural expressions in the generated reply text.

[0620] "Means for automatically classifying e-mails" refers to a function that automatically classifies e-mails into pre-defined categories based on the analyzed content.

[0621] "Means for notifying" is a function that notifies the user of the analysis results, the date and time of extraction, and the necessary measures.

[0622] "Means for displaying a reply message in real time via a display device" refers to a function for displaying the generated reply message to the user in real time using a display device (e.g., a head-mounted display).

[0623] A "generative AI model" is an artificial intelligence model used to generate natural-sounding sentences, taking a prompt as input and outputting an appropriate reply.

[0624] A "prompt sentence" is an input sentence given to a generative AI model, based on which a reply sentence is generated.

[0625] "Means for recording as a calendar event" is a function for recording and managing the analyzed date and time and corresponding matters as a calendar event in the system.

[0626] This invention is an AI email assistant system that enables store staff in brick-and-mortar stores to efficiently handle email reservations and inquiries from customers. The system analyzes received emails, automatically generates appropriate replies, and checks their contents. It also notifies users of the analysis results, the extracted date and time, and the necessary actions, and displays the replies in real time on a display device.

[0627] Hardware and software used

[0628] Hardware: Head-mounted display worn by the store clerk (e.g., Microsoft HoloLens)

[0629] software:

[0630] Generative AI model: A model that generates natural language using TensorFlow

[0631] Web server: Django framework

[0632] HMD Control: Uses OpenCV to manage display and notifications

[0633] System action

[0634] 1. Email analysis

[0635] The server has the means to parse the content of received emails, extracting important information such as sender, recipient, subject, body, date, etc.

[0636] 2. Automatically generate replies

[0637] Based on the analysis results, a generative AI model is used to automatically generate a reply, taking the prompt as input and generating an appropriate reply.

[0638] An example prompt is:

[0639] A customer is asking: Please make an appointment for October 25, 2023 at 2 PM. Please generate a suggested response.

[0640] 3. Grammar and Expression Check

[0641] The generated reply text is checked by a grammar engine on the server, and any errors or unnatural expressions detected are corrected.

[0642] 4. Email Classification and Notification

[0643] Based on the analysis results, emails are automatically categorized into predefined categories such as "reservations" and "inquiries," and important dates and times and actions are extracted.

[0644] 5. Real-time notifications and displays

[0645] The analysis results and reply text are sent to the store clerk in real time via the HMD, allowing the clerk to immediately check the reply and take action if necessary.

[0646] Specific examples

[0647] When a customer emails a store with a reservation, the server parses the email and generates an appropriate reply, such as:

[0648] Thank you for your help. We have accepted your reservation for October 25, 2023 at 2:00 PM. Thank you in advance.

[0649] This reply is displayed in real time on the HMD, allowing the store clerk to check its contents and reply to the customer.

[0650] In this way, the AI ​​email assistant system of the present invention enables store staff to efficiently and effectively handle customer inquiries, which is expected to significantly reduce the burden of email processing in daily work and improve work efficiency.

[0651] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0652] Step 1:

[0653] The server receives emails received from the user's device. To analyze the content of these emails, it extracts important information such as the sender, recipient, subject, body, date, etc. The email is the input, and the extracted information is the output.

[0654] Step 2:

[0655] The server automatically generates a reply based on the analyzed content. Using a generative AI model, it generates an appropriate reply using the prompt as input. Specifically, the prompt is input into the generative AI model, which outputs a natural reply. For example, the following prompt is input:

[0656] A customer is asking: Please make an appointment for October 25, 2023 at 2 PM. Please generate a suggested response.

[0657] The generative AI model outputs an appropriate reply based on this prompt.

[0658] Step 3:

[0659] The server performs grammar and expression checks on the generated reply. It uses a grammar engine to detect grammatical errors and unnatural expressions in the reply and corrects them as necessary. The input is the generated reply, and the output is the corrected or verified reply.

[0660] Step 4:

[0661] The server automatically classifies emails based on the analysis results. It sorts emails into appropriate folders according to predefined categories (e.g., "reservations" or "inquiries"). The input is the analyzed email information, and the output is the classified email.

[0662] Step 5:

[0663] The server notifies the user's device of the analysis results, the extracted date and time, and the corresponding action. Furthermore, this information, including the reply, is displayed in real time on a head-mounted display (HMD). The input is the analysis results and the reply, and the output is the notification and reply displayed on the HMD.

[0664] Step 6:

[0665] The user can check the reply message through the HMD and make corrections if necessary. If corrections are made, the information is sent back to the server, and the final reply message is confirmed. The input is the generated and corrected reply message, and the output is the final reply message.

[0666] Step 7:

[0667] When the server receives the final sending instruction from the user, it sends a confirmed reply to the customer. The input is the final reply, and the output is the completion of the email sending.

[0668] Step 8:

[0669] The server records the analysis results and notifications as a calendar in the system, and can review them later if needed. The input is important dates and times and actions, and the output is a record of the calendar events.

[0670] 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.

[0671] This invention is an AI email assistant system that supports efficient and effective email communication and combines it with an emotion engine that recognizes the user's emotions. This system consists of three main components: a server, a terminal, and a user.

[0672] Email parsing and reply generation

[0673] When the server receives an email sent from a user's device, it automatically analyzes its contents. This analysis extracts important information (e.g., sender, recipient, subject, body, date). Based on the extracted information, the server uses a generative AI to automatically generate an appropriate reply. The generative AI generates a reply in natural-sounding sentences based on the context and content of the email.

[0674] Example: When a user receives an email proposing a new project, the server parses the email and generates a reply like this:

[0675] Thank you for your continued support.

[0676] I have reviewed your new project proposal. It is very interesting and I would like to give it a positive consideration. Could you please suggest a date for a detailed meeting?

[0677] Thank you very much.

[0678] Use of emotion engine

[0679] During analysis, the server uses an emotion engine to recognize emotions based on the content of the email received by the user and the user's response. The recognized emotions are reflected in the reply generation process, allowing for the generation of natural-looking replies that take the user's emotional state into account.

[0680] Grammar and Expression Check

[0681] The server checks the generated reply for grammar and expression. It uses a grammar engine to detect grammatical errors and unnatural expressions and corrects them as necessary. Once the reply has been corrected, it notifies the user.

[0682] Email classification and notifications

[0683] Based on the analysis results, the server automatically classifies emails into pre-defined categories. For example, categories such as "Meetings / Conferences," "Projects," and "Other" can be set. Once classification is complete, the server extracts important dates and times and matters to be addressed, and notifies the user's device of these. Using an emotion engine, the content and method of notifications can be adjusted based on the user's emotions.

[0684] Example: When an invitation email for a meeting is received, the server classifies the email into the "Meeting / Conference" category, extracts the date and time of the meeting, and notifies the user.

[0685] Subject: (Meeting Notice) Information about a new project meeting

[0686] Notification content:

[0687] A new project meeting is scheduled for the following dates:

[0688] Date and time: October 25, 2023 14:00

[0689] Location: Conference Room 1

[0690] User confirmation and correction

[0691] The user can check the notification from the server and the generated reply message through the terminal. If necessary, the user can modify the reply message and issue a final transmission instruction to the server.

[0692] Sending a reply

[0693] When the user issues a sending instruction, the terminal sends the instruction to the server, which then sends the final reply to the recipient.

[0694] In this way, the AI ​​email assistant system of the present invention can efficiently and appropriately reply to emails received by the user. Furthermore, by utilizing the emotion engine, natural replies that reflect the user's emotions are possible, improving the efficiency and quality of the entire email process.

[0695] The processing flow will be explained below.

[0696] Step 1:

[0697] The user receives an email on the terminal, which then sends the received email to the server.

[0698] Step 2:

[0699] The server analyzes the content of the email received, extracting information such as the email text, sender, recipient, subject, and date.

[0700] Step 3:

[0701] The server uses an emotion engine to recognize the user's emotions, which are determined based on the content of the email and the user's past email history.

[0702] Step 4:

[0703] The server automatically generates a reply using generative AI based on the analysis results and the recognized emotions. The generative AI generates natural-sounding sentences taking into account the context and the user's emotions.

[0704] Example: If a user receives an email proposing a new project and the emotion engine recognizes that the user is "interested," the server generates a reply like this:

[0705] Thank you for your continued support.

[0706] I have reviewed your new project proposal. It is very interesting and I would like to give it a positive consideration. Could you please suggest a date for a detailed meeting?

[0707] Thank you very much.

[0708] Step 5:

[0709] The server performs a grammar and expression check on the generated reply, using a grammar engine to detect grammatical errors and unnatural expressions and correct them as necessary.

[0710] Step 6:

[0711] The server automatically classifies emails based on the analysis results, for example, into categories such as "Meetings / Conferences," "Projects," and "Other."

[0712] Step 7:

[0713] The server notifies the user's device of the date and time and action items extracted from the email. Notifications include meeting schedules and important tasks. The emotion engine adjusts the content and method of notifications based on the user's emotions.

[0714] Example: When a user receives an invitation email for a meeting and the emotion engine recognizes "tension," the server notifies the user of the meeting schedule in detail.

[0715] Subject: (Meeting Notice) Information about a new project meeting

[0716] Notification content:

[0717] A new project meeting is scheduled for the following dates:

[0718] Date and time: October 25, 2023 14:00

[0719] Location: Conference Room 1

[0720] Thank you for your preparation.

[0721] Step 8:

[0722] The user checks the notification content from the server and the generated reply text on the terminal, and can modify the reply text if necessary.

[0723] Step 9:

[0724] The user issues a transmission instruction for the reply message after editing, and the terminal sends the instruction to the server.

[0725] Step 10:

[0726] The server then sends the final reply to the recipient, allowing users to communicate efficiently via email.

[0727] Example 2

[0728] 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."

[0729] In conventional email processing systems, the content analysis of received emails, automatic generation of reply messages, grammar and expression checks, and email classification and notification are all performed separately, resulting in problems with consistency and accuracy. Furthermore, reply messages do not reflect the user's feelings, making it difficult to communicate naturally and appropriately.

[0730] 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 analyzing received emails, means for automatically generating reply messages based on the analysis results, means for recognizing the sentiment of the generated reply messages, means for checking grammar and expressions, means for automatically classifying emails based on the analyzed content, and means for notifying the analysis results, extracted date and time, and corresponding matters. This improves the consistency and accuracy of email processing and enables natural replies that reflect the user's sentiment.

[0731] The "means for analyzing received e-mail" is a means for automatically analyzing the contents of received e-mail and extracting important information.

[0732] The "means for automatically generating a reply message based on the analysis results" is a means for automatically generating an appropriate reply message based on the analyzed information.

[0733] The "means for recognizing the emotion of the generated reply message" is a means for analyzing and recognizing the emotion contained in the content of the generated reply message.

[0734] The "means for checking grammar and expressions" refers to a means for checking and correcting grammatical errors and unnatural expressions in the generated reply text.

[0735] The "means for automatically classifying e-mails based on the analyzed content" is a means for automatically classifying e-mails into pre-defined categories based on the analyzed content.

[0736] The "means for notifying the analysis results, extracted dates and times, and corresponding matters" refers to a means for notifying the user of the analysis results and important dates and times and corresponding matters extracted from them.

[0737] This invention is an AI email assistant system that supports efficient and effective email communication and combines it with an emotion engine that recognizes the user's emotions. This system consists of three main components: a server, a terminal, and a user.

[0738] Email parsing and reply generation

[0739] When the server receives an email sent from a user's device, it automatically analyzes the content using NLP (Natural Language Processing) algorithms, particularly tools such as Google Cloud Natural Language API, to extract important information such as the sender, recipient, subject, body, and date.

[0740] Based on the extracted information, the server automatically generates an appropriate reply using generative AI (e.g., OpenAI's GPT-3), which generates a natural-sounding reply based on the context and content of the email.

[0741] As a concrete example, when a user receives an email proposing a new project, the following prompt sentence is input to the generative AI model:

[0742] Generate a sample project proposal email response using the information below.

[0743] Sender: Taro Tanaka

[0744] Recipient: Ichiro Suzuki

[0745] Subject: New Project Proposal

[0746] Body: I ​​would like to propose a new project.

[0747] Date: October 10, 2023

[0748] Based on this prompt, the following reply will be generated:

[0749] Thank you for your continued support.

[0750] I have reviewed your new project proposal. It is very interesting and I would like to give it a positive consideration. Could you please suggest a date for a detailed meeting?

[0751] Thank you very much.

[0752] Use of emotion engine

[0753] The server uses an emotion engine for analysis. This emotion engine uses, for example, the emotion analysis function of Microsoft Azure's Text Analytics API. The emotion engine recognizes emotions based on the content of the email received by the user and the user's response. This recognized emotion is reflected in the reply generation process. This allows for the generation of natural-looking replies that take the user's emotional state into account.

[0754] Grammar and Expression Check

[0755] The server then checks the generated reply for grammar and expression. Specifically, it uses a grammar engine such as the Grammarly API to detect grammatical errors and unnatural expressions and corrects them as necessary. Once the corrections are complete, the user is notified.

[0756] Email classification and notifications

[0757] The server automatically classifies emails into predefined categories based on the analysis results. For example, a machine learning model (trained with scikit-learn) can be used. Once classification is complete, the server extracts important dates and times and actions to take and notifies the user's device. An emotion engine can be used to adjust the content and method of notifications based on the user's emotions.

[0758] For example, when a user receives an invitation email for a meeting, the server classifies the email into the "Meeting / Conference" category, extracts the date and time of the meeting, and notifies the user:

[0759] Subject: (Meeting Notice) Information about a new project meeting

[0760] Notification content:

[0761] A new project meeting is scheduled for the following dates:

[0762] Date and time: October 25, 2023 14:00

[0763] Location: Conference Room 1

[0764] User confirmation and correction

[0765] The user can check the notification from the server and the generated reply message through the terminal. If the reply message is not appropriate, the user can correct it and issue a final transmission instruction to the server.

[0766] Sending a reply

[0767] When a user issues a sending instruction, the device sends the instruction to the server, which then sends the final reply to the recipient. Through this series of processes, the AI ​​email assistant system efficiently and effectively supports users' email communication.

[0768] The above is a specific embodiment for carrying out the present invention. The combination of the generative AI model and emotion engine, which is a feature of this invention, significantly improves the efficiency and quality of email processing.

[0769] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0770] Step 1:

[0771] The server receives emails sent from the user's terminal. The input is the email information received from the terminal, and the output is the received email data. The server performs the specific operation of retrieving new emails from the email server using the IMAP or POP3 protocol.

[0772] Step 2:

[0773] The server analyzes the content of received emails. The input is the received email data, and the output is the analysis results (e.g., sender, recipient, subject, body, date). NLP algorithms are used for the analysis, and specific operations are performed using, for example, the Google Cloud Natural Language API.

[0774] Step 3:

[0775] The server generates a reply using a generative AI model (e.g., OpenAI's GPT-3) based on the analyzed email content. The input is the analysis result, and the output is the generated reply. Specifically, the following prompt is input to the generative AI model:

[0776] Generate a sample project proposal email response using the information below.

[0777] Sender: Taro Tanaka

[0778] Recipient: Ichiro Suzuki

[0779] Subject: New Project Proposal

[0780] Body: I ​​would like to propose a new project.

[0781] Date: October 10, 2023

[0782] Step 4:

[0783] The server uses an emotion engine to recognize emotions from the generated reply text. The input is the generated reply text, and the output is emotion data. Specifically, the server uses Microsoft Azure's Text Analytics API.

[0784] Step 5:

[0785] The server adjusts the naturalness and appropriateness of the reply based on the emotional data. The input is the emotional data, and the output is a modified reply that reflects the emotion. In this step, the tone and expression of the document generated by the generative AI model are changed to match the emotion.

[0786] Step 6:

[0787] The server checks the generated reply for grammar and expression. The input is the generated reply, and the output is the grammar- and expression-checked reply. Specifically, it uses the Grammarly API to detect and correct grammatical errors and unnatural expressions.

[0788] Step 7:

[0789] The server automatically classifies emails based on the analysis results. The input is the analysis results, and the output is the classified emails. Specifically, it uses a machine learning model (trained using scikit-learn) to classify emails into pre-defined categories.

[0790] Step 8:

[0791] The server notifies the user's device of the analysis results, the extracted date and time, and the necessary actions. The input is the analysis results and extracted important information, and the output is notification information. Using an emotion engine, the content and method of notification are adjusted based on the user's emotions.

[0792] Step 9:

[0793] The user checks the notification from the server and the generated reply through the terminal. The input is the notification from the server and the generated reply, and the output is the reply that has been checked and corrected by the user. The user corrects the reply as necessary and performs the specific action of issuing a final transmission instruction to the server.

[0794] Step 10:

[0795] The terminal transmits the user's transmission instruction to the server. The input is the user's transmission instruction, and the output is the transmission instruction data to the server.

[0796] Step 11:

[0797] The server sends the final reply to the recipient. The input is the revised final reply, and the output is the result sent to the recipient. Through this series of processes, the AI ​​email assistant system efficiently and effectively supports users' email communication.

[0798] (Application example 2)

[0799] 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."

[0800] There is a need for systems that can reduce the burden of responding to large volumes of inquiries via e-mail and other electronic communications, while also providing appropriate replies based on emotions. However, conventional systems have difficulty taking emotions into account, making it difficult to respond in a way that is sensitive to the user's feelings. It is also difficult to respond quickly while ensuring the appropriateness of grammar and expression in replies.

[0801] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing received electronic communications, means for automatically generating a reply message based on the analysis results, means for checking the grammar and expressions of the generated reply message, means for automatically classifying the electronic communications based on the analyzed content, means for notifying the analysis results and the extracted date and time and response matters, means for using an emotion engine for emotion analysis, and means for generating a reply message based on the emotion. This enables natural replies that are in line with the user's emotions and improves the efficiency and quality of inquiry responses.

[0802] "Means for analyzing received electronic communications" refers to the function of devices such as servers and terminals to analyze the contents of received emails and messages and extract important information.

[0803] "Means for automatically generating a reply based on the analysis results" refers to technology for automatically generating an appropriate reply based on the analyzed data.

[0804] "Means for checking the grammar and expressions of generated replies" refers to a function for detecting and correcting grammatical errors and unnatural expressions in automatically generated replies.

[0805] "Means for automatically classifying electronic communications based on analyzed content" refers to a function that automatically sorts emails and messages into specific categories based on the results of the analysis.

[0806] "Means for notifying the analysis results, extracted dates and times, and action items" refers to a notification function for informing the user of the analysis results, important dates, and action items.

[0807] "Means of using an emotion engine to perform emotion analysis" is a technology for recognizing and analyzing emotions from the text of emails and messages.

[0808] The "means for generating a reply based on emotion" is a technology for automatically generating a reply using expressions and writing styles that correspond to the recognized emotion.

[0809] MODE FOR CARRYING OUT THE INVENTION

[0810] The system for implementing the present invention is composed of three main components: a server, a terminal, and a user. A specific embodiment of the system is shown below.

[0811] Email parsing and reply generation

[0812] When the server receives an electronic communication sent from a user's device, it analyzes the content of the communication using natural language processing (NLP) technology. The analysis is performed with the aim of extracting important information from the electronic communication (e.g., sender, recipient, subject, body, date). Based on the extracted information, the server uses generative AI (e.g., GPT-4) to automatically generate an appropriate reply.

[0813] Example: When a user receives an inquiry email stating that "the product has not arrived," the server analyzes the email and generates a reply message like the following:

[0814] We are very sorry. We checked the tracking information for your order and it was shipped on XX date, but we will reconfirm the delivery status and respond accordingly.

[0815] Use of emotion engine

[0816] During analysis, the server uses an emotion engine (e.g., Google Cloud Natural Language API) to recognize emotions based on the content of the email received by the user and the user's response. The recognized emotions are reflected in the reply generation process. This allows for the generation of natural-looking replies that take the user's emotional state into account.

[0817] Example: For an angry customer, the server uses more polite and careful language. An example of a prompt for the AI ​​generation is, "Read the email about the customer's angry delivery of the item and generate a reply that includes an apology and reconfirmation."

[0818] Grammar and Expression Check

[0819] The server uses a grammar checker (e.g., Grammarly API) to detect grammatical errors and unnatural expressions in the generated reply, correcting them as necessary, and notifying the user of the completed reply.

[0820] Email classification and notifications

[0821] Based on the analysis results, the server automatically classifies electronic communications into pre-defined categories. For example, categories such as "Meetings / Consultations," "Projects," and "Other" may be set. Once classification is complete, the server extracts important dates and times and matters to be addressed, and notifies the user of these on their device. Using an emotion engine, the content and method of notifications can be adjusted based on the user's emotions.

[0822] Example: When an invitation email for a meeting is received, the server classifies the email into the "Meeting / Conference" category, extracts the date and time of the meeting, and notifies the user.

[0823] Subject: (Meeting Notice) Information about a new project meeting

[0824] Notification content:

[0825] A new project meeting is scheduled for the following dates:

[0826] Date and time: October 25, 2023 14:00

[0827] Location: Conference Room 1

[0828] User confirmation and correction

[0829] The user can check the notification from the server and the generated reply message through the terminal. If necessary, the user can modify the reply message and issue a final transmission instruction to the server.

[0830] Sending a reply

[0831] When the user issues a sending instruction, the terminal sends the instruction to the server, which then sends the final reply to the recipient.

[0832] In this way, the system of the present invention allows users to efficiently and appropriately reply to electronic communications received. Furthermore, by utilizing the emotion engine, natural replies that reflect the user's emotions are possible, thereby improving the efficiency and quality of the entire email process.

[0833] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0834] Step 1:

[0835] The server receives the incoming electronic communication. The input is an email or message sent from a user terminal. The server converts the received communication into a format suitable for analysis. The output of this step is text data that is used for analysis.

[0836] Step 2:

[0837] The server analyzes the content of the received electronic communication. The input is the text data generated in step 1. The server uses natural language processing (NLP) techniques to extract important information (sender, recipient, subject, body, date, etc.). Specifically, it uses a text analysis library (e.g., spaCy) for parsing and information extraction. The output of this step is a set of extracted important information.

[0838] Step 3:

[0839] The server uses an emotion engine to recognize emotions contained in the extracted information. The input is the information extracted in step 2. The emotion engine (e.g., Google Cloud Natural Language API) is used to parse emotions from the email body. The output of this step is the recognized emotion data.

[0840] Step 4:

[0841] The server automatically generates a reply based on the analysis results. The input is the set of important information extracted in step 2 and the emotion data obtained in step 3. The server uses a generative AI model (e.g., GPT-4) to set the prompt as follows:

[0842] A customer receives an email saying they didn't receive their item. You recognize their anger. Generate a response based on this information, including an apology and reassurance.

[0843] The output of this step is the generated reply.

[0844] Step 5:

[0845] The server checks the grammar and language of the generated reply. The input is the reply generated in step 4. A grammar checker (e.g., Grammarly API) is used to detect grammatical errors and unnatural language and correct them if necessary. The output of this step is the reply whose grammar and language have been checked and corrected.

[0846] Step 6:

[0847] The server automatically classifies the electronic communications based on the analysis results. The input is the information extracted in step 2. The server performs classification based on pre-defined categories (e.g., "meetings / discussions," "projects," "others"). The output of this step is a set of classified electronic communications.

[0848] Step 7:

[0849] The server notifies the analysis results, extracted dates and times, and actions to be taken. The input is the information extracted in step 2 and the classification results in step 6. The server extracts important dates and times and actions to be taken, and sends a notification to the user device. The output of this step is the notification sent to the user device.

[0850] Step 8:

[0851] The user checks the notification from the server and the generated reply through the terminal. The input is the notification sent in step 7 and the reply corrected in step 5. The user checks the reply and corrects it if necessary. The output of this step is the checked and corrected reply.

[0852] Step 9:

[0853] The user issues a sending instruction, and the terminal sends that instruction to the server. The input is the reply message that was checked and corrected in step 8. The server sends the final reply message to the recipient. The output of this step is the sent reply message.

[0854] 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.

[0855] 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.

[0856] 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.

[0857] [Third embodiment]

[0858] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0859] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0860] 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).

[0861] 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.

[0862] 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.

[0863] 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).

[0864] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0865] 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.

[0866] 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.

[0867] 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.

[0868] 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.

[0869] 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."

[0870] This invention is an AI email assistant system that supports efficient and effective email communication. This system consists of three main components: a server, a terminal, and a user.

[0871] Email parsing and reply generation

[0872] When the server receives an email sent from a user's device, it automatically analyzes its contents. This analysis extracts important information (e.g., sender, recipient, subject, body, date). Based on the extracted information, the server uses a generative AI to automatically generate an appropriate reply. The generative AI generates a reply in natural-sounding sentences based on the context and content of the email.

[0873] Example: When a user receives an email proposing a new project, the server parses the email and generates a reply like this:

[0874] Thank you for your continued support.

[0875] I have reviewed your new project proposal. It is very interesting and I would like to give it a positive consideration. Could you please suggest a date for a detailed meeting?

[0876] Thank you very much.

[0877] Grammar and Expression Check

[0878] The server checks the generated reply for grammar and expression. It uses a grammar engine to detect grammatical errors and unnatural expressions and corrects them as necessary. Once the reply has been corrected, it notifies the user.

[0879] Email classification and notifications

[0880] Based on the analysis results, the server automatically classifies emails into pre-defined categories, such as "Meetings / Conferences," "Projects," and "Other." Once classification is complete, the server extracts important dates and times and matters to be addressed, and notifies the user's device.

[0881] Example: When an invitation email for a meeting is received, the server classifies the email into the "Meeting / Conference" category, extracts the date and time of the meeting, and notifies the user.

[0882] Subject: (Meeting Notice) Information about a new project meeting

[0883] Notification content:

[0884] A new project meeting is scheduled for the following dates:

[0885] Date and time: October 25, 2023 14:00

[0886] Location: Conference Room 1

[0887] User confirmation and correction

[0888] The user can check the notification from the server and the generated reply message through the terminal. If necessary, the user can modify the reply message and issue a final transmission instruction to the server.

[0889] Sending a reply

[0890] When the user issues a sending instruction, the terminal sends the instruction to the server, which then sends the final reply to the recipient.

[0891] In this way, the AI ​​email assistant system of the present invention enables users to reply to received emails efficiently and effectively, significantly reducing the burden of email processing in their daily work.

[0892] The processing flow will be explained below.

[0893] Step 1:

[0894] The user receives an email on the terminal, which then sends the received email to the server.

[0895] Step 2:

[0896] The server analyzes the content of the email received, which includes extracting information such as the email text, sender, recipient, subject, and date.

[0897] Step 3:

[0898] Based on the analysis results, the server uses generative AI to automatically generate a reply. Specifically, it selects a reply template based on the content and then uses natural language generation technology to construct the reply.

[0899] Step 4:

[0900] The server performs a grammar and expression check on the generated reply, using a grammar engine to detect grammatical errors and unnatural expressions and correct them as necessary.

[0901] Step 5:

[0902] The server automatically classifies emails based on the analysis results, for example, into categories such as "Meetings / Conferences," "Projects," and "Other."

[0903] Step 6:

[0904] The server extracts information from emails and sends it to the user's device, including meeting schedules and important tasks.

[0905] Step 7:

[0906] The user checks the notification content from the server and the generated reply text on the terminal, and can modify the reply text if necessary.

[0907] Step 8:

[0908] The user issues a transmission instruction for the reply message after editing, and the terminal sends the instruction to the server.

[0909] Step 9:

[0910] The server then sends the final reply to the recipient, allowing users to communicate efficiently via email.

[0911] Example 1

[0912] 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."

[0913] In today's business environment, email is an important means of communication, but processing it requires a great deal of time and effort. Responding appropriately and promptly to the large volume of emails received is particularly difficult, hindering efficient business operations. Manually crafting replies can lead to grammatical errors and unnatural expressions. Important emails can be overlooked, or follow-ups can be delayed because they are not properly categorized. These issues need to be addressed.

[0914] 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.

[0915] In this invention, the server includes means for analyzing received messages, means for automatically generating a response based on the analysis results, means for checking the grammar and expressions of the generated response, means for automatically classifying messages based on the analyzed content, and means for notifying the user of the analysis results, the extracted date and time, and the corresponding matters. This automates a series of processes from receiving emails to generating a reply, checking grammar, automatic classification, and notifying the user of important information, enabling the user to process emails quickly and accurately.

[0916] "Means for parsing received messages" refers to the process by which the server receives messages sent from the user's terminal and disassembles their contents to extract information in a structured format.

[0917] "Means for automatically generating a response based on the analysis results" refers to the process of automatically creating a natural and appropriate reply using a generative AI model based on the analyzed information.

[0918] "Means for checking the grammar and expression of the generated response sentence" refers to the process of detecting and correcting grammatical errors and unnatural expressions in the generated reply sentence.

[0919] "Means for automatically categorizing messages based on analyzed content" refers to a process for categorizing messages into predefined categories (e.g., meetings, projects, etc.) based on analyzed information.

[0920] "Means for notifying the analysis results, extracted date and time, and corresponding matters" refers to the process of notifying the user of important information extracted during the analysis process (date and time, corresponding matters, etc.).

[0921] This invention provides an AI message assistant system that supports efficient and effective electronic message communication. The system consists of three main components: a server, a terminal, and a user.

[0922] Receiving and parsing messages

[0923] server

[0924] The server receives messages sent from the user's device and automatically analyzes their contents. This analysis uses natural language processing libraries (e.g., spaCy and NLTK) to extract important information such as the sender, recipient, subject, body, and date. The received message data is saved in a database, and the analyzed information is stored in temporary storage.

[0925] Automatic generation of response sentences

[0926] server

[0927] Based on the analyzed information, the server uses a generative AI model (e.g., OpenAI's GPT-4) to generate appropriate prompts, which create natural-sounding sentences based on context and content.

[0928] A specific example is the following prompt:

[0929] Sender: Sender name

[0930] Recipient: Recipient name

[0931] Subject: The subject of the message

[0932] Body: The body of the message

[0933] Prompt: Generate a reply to the following message. Responses should be polite and helpful.

[0934] Checking grammar and expressions in responses

[0935] server

[0936] The generated response sentence is checked for grammar and expression using a grammar engine (e.g., Grammarly API). Grammatical errors and unnatural expressions are detected and corrected as necessary. Once corrected, the response sentence is saved back to storage.

[0937] Automatic message classification

[0938] server

[0939] Based on the analysis results, the server classifies messages into predefined categories, such as "Meetings," "Projects," and "Other." Machine learning models (e.g., support vector machines and random forests) are used for classification, and the classification results are stored in a database as message metadata.

[0940] Important Information Notification

[0941] server

[0942] Important information (e.g., date, time, and action) is extracted during the analysis process, and a notification containing this information is generated and sent to the user's device, where the user can visually confirm it.

[0943] Terminal

[0944] The user's device receives the notification sent from the server and displays it to the user, including the date and time, location of the meeting, and any action to be taken.

[0945] User verification and correction

[0946] User

[0947] The user can check the generated response and notification through his / her own terminal, and if necessary, correct the response and issue a final transmission instruction to the server.

[0948] Sending a response

[0949] server

[0950] When the user issues a final command to send the message, the server receives the command and sends the finalized response via the mail server, using the SMTP protocol.

[0951] In this way, the AI ​​message assistant system automates the entire process from receiving messages to generating responses, grammar checking, automatic classification, and notifying users of important information, making message handling more efficient.

[0952] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0953] Step 1:

[0954] Parsing received messages

[0955] Operation explanation

[0956] The server receives messages sent from the user's device. This process uses an email protocol (e.g., IMAP or POP3) and uses a natural language processing library (e.g., spaCy or NLTK) to analyze the content of the received messages.

[0957] input

[0958] Received message data

[0959] output

[0960] Parsed information (sender, recipient, subject, body, date)

[0961] Specific actions

[0962] The server receives new messages, analyzes their contents, extracts important information, and stores it in temporary storage.

[0963] Step 2:

[0964] Automatically generate a response

[0965] Operation explanation

[0966] The server automatically generates an appropriate response sentence using a generative AI model (e.g., OpenAI's GPT-4) based on the analyzed information. In this process, the prompt sentence is input into the generative AI model, which outputs the response sentence.

[0967] input

[0968] Parsed information (sender, recipient, subject, body, date)

[0969] output

[0970] Auto-generated response

[0971] Specific actions

[0972] The server passes the following prompt to the spawned AI:

[0973] From: John Doe

[0974] Recipient: Taro Yamada

[0975] Subject: New project proposal

[0976] Body: I'd like to propose an idea for a new project. I'd like to discuss the details at our next meeting.

[0977] Prompt: Generate a reply to the following message. Responses should be polite and helpful.

[0978] The generation AI then generates a response sentence and stores it in storage.

[0979] Step 3:

[0980] Check the grammar and expression of the generated response

[0981] Operation explanation

[0982] The server passes the generated response to a grammar engine (e.g., Grammarly API) to check the grammar and expressions. If grammatical errors or unnatural expressions are detected, they are corrected.

[0983] input

[0984] Generated response

[0985] output

[0986] Responses with grammar and expression corrections

[0987] Specific actions

[0988] The server sends the generated response sentence to the grammar engine, receives the corrected response sentence, and stores it in storage.

[0989] Step 4:

[0990] Automatically categorize messages

[0991] Operation explanation

[0992] Based on the analyzed information, the server classifies messages into predefined categories using machine learning models (e.g., SVM or Random Forest).

[0993] input

[0994] Analyzed information

[0995] output

[0996] Category information (e.g., meeting, project, etc.)

[0997] Specific actions

[0998] The server categorizes messages and stores the categorization results in a database as message metadata.

[0999] Step 5:

[1000] Notify the extracted date and time and the action to be taken

[1001] Operation explanation

[1002] Important dates and times and action items are extracted during the analysis process and notified to the user. In this notification process, notification data is generated and sent to the user's terminal.

[1003] input

[1004] Analysis results and extracted information

[1005] output

[1006] Notification Data

[1007] Specific actions

[1008] The server generates a notification including the date and time of the meeting and the matters to be addressed, and sends it to the user's terminal.

[1009] Step 6:

[1010] User confirmation and correction

[1011] Operation explanation

[1012] The user can check the notification and the generated response text through the terminal, modify the response text as necessary, and issue a final transmission instruction to the server.

[1013] input

[1014] Notification data, generated response

[1015] output

[1016] Revised Final Response

[1017] Specific actions

[1018] The user checks the response text and notification generated on the terminal, corrects the response text as necessary, and clicks the send button.

[1019] Step 7:

[1020] Sending a final response

[1021] Operation explanation

[1022] When the user issues a final sending instruction, the server receives the instruction and sends the final response via the mail server using the SMTP protocol.

[1023] input

[1024] Revised Final Response

[1025] output

[1026] Sent messages

[1027] Specific actions

[1028] The server sends a final response to the SMTP server and the message is delivered to the recipient.

[1029] In this way, each step works together in a specific way to achieve efficient message processing as a whole system.

[1030] (Application example 1)

[1031] 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."

[1032] Conventional email management systems make it difficult for users to respond to received emails promptly and appropriately. Furthermore, they lack the functionality to analyze email content, automatically generate a reply, and check the grammar of that reply, making it difficult to significantly reduce the amount of work required by users. Especially in workplaces where real-time responses are required, delays in these tasks can lead to reduced work efficiency and potentially inadequate customer service.

[1033] 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.

[1034] In this invention, the server includes a means for analyzing received emails, a means for automatically generating a reply based on the analysis results, and a means for checking the grammar and expressions of the generated reply. This allows the user to be notified of the analysis results, the extracted date and time, and the corresponding action, and to check the reply in real time via a display device. Furthermore, by generating a reply using a prompt sentence using a generative AI model, a more natural and appropriate response is automatically provided, significantly reducing the burden on the user and improving work efficiency.

[1035] "Means for analyzing" is a function for analyzing the contents of received e-mail and extracting important information.

[1036] "Means for automatically generating replies" refers to a function that automatically creates appropriate replies using a generative AI model based on the analyzed content.

[1037] The "means for checking grammar and expressions" is a function for detecting and correcting grammatical errors and unnatural expressions in the generated reply text.

[1038] "Means for automatically classifying e-mails" refers to a function that automatically classifies e-mails into pre-defined categories based on the analyzed content.

[1039] "Means for notifying" is a function that notifies the user of the analysis results, the date and time of extraction, and the necessary measures.

[1040] "Means for displaying a reply message in real time via a display device" refers to a function for displaying the generated reply message to the user in real time using a display device (e.g., a head-mounted display).

[1041] A "generative AI model" is an artificial intelligence model used to generate natural-sounding sentences, taking a prompt as input and outputting an appropriate reply.

[1042] A "prompt sentence" is an input sentence given to a generative AI model, based on which a reply sentence is generated.

[1043] "Means for recording as a calendar event" is a function for recording and managing the analyzed date and time and corresponding matters as a calendar event in the system.

[1044] This invention is an AI email assistant system that enables store staff in brick-and-mortar stores to efficiently handle email reservations and inquiries from customers. The system analyzes received emails, automatically generates appropriate replies, and checks their contents. It also notifies users of the analysis results, the extracted date and time, and the necessary actions, and displays the replies in real time on a display device.

[1045] Hardware and software used

[1046] Hardware: Head-mounted display worn by the store clerk (e.g., Microsoft HoloLens)

[1047] software:

[1048] Generative AI model: A model that generates natural language using TensorFlow

[1049] Web server: Django framework

[1050] HMD Control: Uses OpenCV to manage display and notifications

[1051] System action

[1052] 1. Email analysis

[1053] The server has the means to parse the content of received emails, extracting important information such as sender, recipient, subject, body, date, etc.

[1054] 2. Automatically generate replies

[1055] Based on the analysis results, a generative AI model is used to automatically generate a reply, taking the prompt as input and generating an appropriate reply.

[1056] An example prompt is:

[1057] A customer is asking: Please make an appointment for October 25, 2023 at 2 PM. Please generate a suggested response.

[1058] 3. Grammar and Expression Check

[1059] The generated reply text is checked by a grammar engine on the server, and any errors or unnatural expressions detected are corrected.

[1060] 4. Email Classification and Notification

[1061] Based on the analysis results, emails are automatically categorized into predefined categories such as "reservations" and "inquiries," and important dates and times and actions are extracted.

[1062] 5. Real-time notifications and displays

[1063] The analysis results and reply text are sent to the store clerk in real time via the HMD, allowing the clerk to immediately check the reply and take action if necessary.

[1064] Specific examples

[1065] When a customer emails a store with a reservation, the server parses the email and generates an appropriate reply, such as:

[1066] Thank you for your help. We have accepted your reservation for October 25, 2023 at 2:00 PM. Thank you in advance.

[1067] This reply is displayed in real time on the HMD, allowing the store clerk to check its contents and reply to the customer.

[1068] In this way, the AI ​​email assistant system of the present invention enables store staff to efficiently and effectively handle customer inquiries, which is expected to significantly reduce the burden of email processing in daily work and improve work efficiency.

[1069] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1070] Step 1:

[1071] The server receives emails received from the user's device. To analyze the content of these emails, it extracts important information such as the sender, recipient, subject, body, date, etc. The email is the input, and the extracted information is the output.

[1072] Step 2:

[1073] The server automatically generates a reply based on the analyzed content. Using a generative AI model, it generates an appropriate reply using the prompt as input. Specifically, the prompt is input into the generative AI model, which outputs a natural reply. For example, the following prompt is input:

[1074] A customer is asking: Please make an appointment for October 25, 2023 at 2 PM. Please generate a suggested response.

[1075] The generative AI model outputs an appropriate reply based on this prompt.

[1076] Step 3:

[1077] The server performs grammar and expression checks on the generated reply. It uses a grammar engine to detect grammatical errors and unnatural expressions in the reply and corrects them as necessary. The input is the generated reply, and the output is the corrected or verified reply.

[1078] Step 4:

[1079] The server automatically classifies emails based on the analysis results. It sorts emails into appropriate folders according to predefined categories (e.g., "reservations" or "inquiries"). The input is the analyzed email information, and the output is the classified email.

[1080] Step 5:

[1081] The server notifies the user's device of the analysis results, the extracted date and time, and the corresponding action. Furthermore, this information, including the reply, is displayed in real time on a head-mounted display (HMD). The input is the analysis results and the reply, and the output is the notification and reply displayed on the HMD.

[1082] Step 6:

[1083] The user can check the reply message through the HMD and make corrections if necessary. If corrections are made, the information is sent back to the server, and the final reply message is confirmed. The input is the generated and corrected reply message, and the output is the final reply message.

[1084] Step 7:

[1085] When the server receives the final sending instruction from the user, it sends a confirmed reply to the customer. The input is the final reply, and the output is the completion of the email sending.

[1086] Step 8:

[1087] The server records the analysis results and notifications as a calendar in the system, and can review them later if needed. The input is important dates and times and actions, and the output is a record of the calendar events.

[1088] 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.

[1089] This invention is an AI email assistant system that supports efficient and effective email communication and combines it with an emotion engine that recognizes the user's emotions. This system consists of three main components: a server, a terminal, and a user.

[1090] Email parsing and reply generation

[1091] When the server receives an email sent from a user's device, it automatically analyzes its contents. This analysis extracts important information (e.g., sender, recipient, subject, body, date). Based on the extracted information, the server uses a generative AI to automatically generate an appropriate reply. The generative AI generates a reply in natural-sounding sentences based on the context and content of the email.

[1092] Example: When a user receives an email proposing a new project, the server parses the email and generates a reply like this:

[1093] Thank you for your continued support.

[1094] I have reviewed your new project proposal. It is very interesting and I would like to give it a positive consideration. Could you please suggest a date for a detailed meeting?

[1095] Thank you very much.

[1096] Use of emotion engine

[1097] During analysis, the server uses an emotion engine to recognize emotions based on the content of the email received by the user and the user's response. The recognized emotions are reflected in the reply generation process, allowing for the generation of natural-looking replies that take the user's emotional state into account.

[1098] Grammar and Expression Check

[1099] The server checks the generated reply for grammar and expression. It uses a grammar engine to detect grammatical errors and unnatural expressions and corrects them as necessary. Once the reply has been corrected, it notifies the user.

[1100] Email classification and notifications

[1101] Based on the analysis results, the server automatically classifies emails into pre-defined categories. For example, categories such as "Meetings / Conferences," "Projects," and "Other" can be set. Once classification is complete, the server extracts important dates and times and matters to be addressed, and notifies the user's device of these. Using an emotion engine, the content and method of notifications can be adjusted based on the user's emotions.

[1102] Example: When an invitation email for a meeting is received, the server classifies the email into the "Meeting / Conference" category, extracts the date and time of the meeting, and notifies the user.

[1103] Subject: (Meeting Notice) Information about a new project meeting

[1104] Notification content:

[1105] A new project meeting is scheduled for the following dates:

[1106] Date and time: October 25, 2023 14:00

[1107] Location: Conference Room 1

[1108] User confirmation and correction

[1109] The user can check the notification from the server and the generated reply message through the terminal. If necessary, the user can modify the reply message and issue a final transmission instruction to the server.

[1110] Sending a reply

[1111] When the user issues a sending instruction, the terminal sends the instruction to the server, which then sends the final reply to the recipient.

[1112] In this way, the AI ​​email assistant system of the present invention can efficiently and appropriately reply to emails received by the user. Furthermore, by utilizing the emotion engine, natural replies that reflect the user's emotions are possible, improving the efficiency and quality of the entire email process.

[1113] The processing flow will be explained below.

[1114] Step 1:

[1115] The user receives an email on the terminal, which then sends the received email to the server.

[1116] Step 2:

[1117] The server analyzes the content of the email received, extracting information such as the email text, sender, recipient, subject, and date.

[1118] Step 3:

[1119] The server uses an emotion engine to recognize the user's emotions, which are determined based on the content of the email and the user's past email history.

[1120] Step 4:

[1121] The server automatically generates a reply using generative AI based on the analysis results and the recognized emotions. The generative AI generates natural-sounding sentences taking into account the context and the user's emotions.

[1122] Example: If a user receives an email proposing a new project and the emotion engine recognizes that the user is "interested," the server generates a reply like this:

[1123] Thank you for your continued support.

[1124] I have reviewed your new project proposal. It is very interesting and I would like to give it a positive consideration. Could you please suggest a date for a detailed meeting?

[1125] Thank you very much.

[1126] Step 5:

[1127] The server performs a grammar and expression check on the generated reply, using a grammar engine to detect grammatical errors and unnatural expressions and correct them as necessary.

[1128] Step 6:

[1129] The server automatically classifies emails based on the analysis results, for example, into categories such as "Meetings / Conferences," "Projects," and "Other."

[1130] Step 7:

[1131] The server notifies the user's device of the date and time and action items extracted from the email. Notifications include meeting schedules and important tasks. The emotion engine adjusts the content and method of notifications based on the user's emotions.

[1132] Example: When a user receives an invitation email for a meeting and the emotion engine recognizes "tension," the server notifies the user of the meeting schedule in detail.

[1133] Subject: (Meeting Notice) Information about a new project meeting

[1134] Notification content:

[1135] A new project meeting is scheduled for the following dates:

[1136] Date and time: October 25, 2023 14:00

[1137] Location: Conference Room 1

[1138] Thank you for your preparation.

[1139] Step 8:

[1140] The user checks the notification content from the server and the generated reply text on the terminal, and can modify the reply text if necessary.

[1141] Step 9:

[1142] The user issues a transmission instruction for the reply message after editing, and the terminal sends the instruction to the server.

[1143] Step 10:

[1144] The server then sends the final reply to the recipient, allowing users to communicate efficiently via email.

[1145] Example 2

[1146] 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."

[1147] In conventional email processing systems, the content analysis of received emails, automatic generation of reply messages, grammar and expression checks, and email classification and notification are all performed separately, resulting in problems with consistency and accuracy. Furthermore, reply messages do not reflect the user's feelings, making it difficult to communicate naturally and appropriately.

[1148] 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 analyzing received emails, means for automatically generating reply messages based on the analysis results, means for recognizing the sentiment of the generated reply messages, means for checking grammar and expressions, means for automatically classifying emails based on the analyzed content, and means for notifying the analysis results, extracted date and time, and corresponding matters. This improves the consistency and accuracy of email processing and enables natural replies that reflect the user's sentiment.

[1149] The "means for analyzing received e-mail" is a means for automatically analyzing the contents of received e-mail and extracting important information.

[1150] The "means for automatically generating a reply message based on the analysis results" is a means for automatically generating an appropriate reply message based on the analyzed information.

[1151] The "means for recognizing the emotion of the generated reply message" is a means for analyzing and recognizing the emotion contained in the content of the generated reply message.

[1152] The "means for checking grammar and expressions" refers to a means for checking and correcting grammatical errors and unnatural expressions in the generated reply text.

[1153] The "means for automatically classifying e-mails based on the analyzed content" is a means for automatically classifying e-mails into pre-defined categories based on the analyzed content.

[1154] The "means for notifying the analysis results, extracted dates and times, and corresponding matters" refers to a means for notifying the user of the analysis results and important dates and times and corresponding matters extracted from them.

[1155] This invention is an AI email assistant system that supports efficient and effective email communication and combines it with an emotion engine that recognizes the user's emotions. This system consists of three main components: a server, a terminal, and a user.

[1156] Email parsing and reply generation

[1157] When the server receives an email sent from a user's device, it automatically analyzes the content using NLP (Natural Language Processing) algorithms, particularly tools such as Google Cloud Natural Language API, to extract important information such as the sender, recipient, subject, body, and date.

[1158] Based on the extracted information, the server automatically generates an appropriate reply using generative AI (e.g., OpenAI's GPT-3), which generates a natural-sounding reply based on the context and content of the email.

[1159] As a concrete example, when a user receives an email proposing a new project, the following prompt sentence is input to the generative AI model:

[1160] Generate a sample project proposal email response using the information below.

[1161] Sender: Taro Tanaka

[1162] Recipient: Ichiro Suzuki

[1163] Subject: New Project Proposal

[1164] Body: I ​​would like to propose a new project.

[1165] Date: October 10, 2023

[1166] Based on this prompt, the following reply will be generated:

[1167] Thank you for your continued support.

[1168] I have reviewed your new project proposal. It is very interesting and I would like to give it a positive consideration. Could you please suggest a date for a detailed meeting?

[1169] Thank you very much.

[1170] Use of emotion engine

[1171] The server uses an emotion engine for analysis. This emotion engine uses, for example, the emotion analysis function of Microsoft Azure's Text Analytics API. The emotion engine recognizes emotions based on the content of the email received by the user and the user's response. This recognized emotion is reflected in the reply generation process. This allows for the generation of natural-looking replies that take the user's emotional state into account.

[1172] Grammar and Expression Check

[1173] The server then checks the generated reply for grammar and expression. Specifically, it uses a grammar engine such as the Grammarly API to detect grammatical errors and unnatural expressions and corrects them as necessary. Once the corrections are complete, the user is notified.

[1174] Email classification and notifications

[1175] The server automatically classifies emails into predefined categories based on the analysis results. For example, a machine learning model (trained with scikit-learn) can be used. Once classification is complete, the server extracts important dates and times and actions to take and notifies the user's device. An emotion engine can be used to adjust the content and method of notifications based on the user's emotions.

[1176] For example, when a user receives an invitation email for a meeting, the server classifies the email into the "Meeting / Conference" category, extracts the date and time of the meeting, and notifies the user:

[1177] Subject: (Meeting Notice) Information about a new project meeting

[1178] Notification content:

[1179] A new project meeting is scheduled for the following dates:

[1180] Date and time: October 25, 2023 14:00

[1181] Location: Conference Room 1

[1182] User confirmation and correction

[1183] The user can check the notification from the server and the generated reply message through the terminal. If the reply message is not appropriate, the user can correct it and issue a final transmission instruction to the server.

[1184] Sending a reply

[1185] When a user issues a sending instruction, the device sends the instruction to the server, which then sends the final reply to the recipient. Through this series of processes, the AI ​​email assistant system efficiently and effectively supports users' email communication.

[1186] The above is a specific embodiment for carrying out the present invention. The combination of the generative AI model and emotion engine, which is a feature of this invention, significantly improves the efficiency and quality of email processing.

[1187] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1188] Step 1:

[1189] The server receives emails sent from the user's terminal. The input is the email information received from the terminal, and the output is the received email data. The server performs the specific operation of retrieving new emails from the email server using the IMAP or POP3 protocol.

[1190] Step 2:

[1191] The server analyzes the content of received emails. The input is the received email data, and the output is the analysis results (e.g., sender, recipient, subject, body, date). NLP algorithms are used for the analysis, and specific operations are performed using, for example, the Google Cloud Natural Language API.

[1192] Step 3:

[1193] The server generates a reply using a generative AI model (e.g., OpenAI's GPT-3) based on the analyzed email content. The input is the analysis result, and the output is the generated reply. Specifically, the following prompt is input to the generative AI model:

[1194] Generate a sample project proposal email response using the information below.

[1195] Sender: Taro Tanaka

[1196] Recipient: Ichiro Suzuki

[1197] Subject: New Project Proposal

[1198] Body: I ​​would like to propose a new project.

[1199] Date: October 10, 2023

[1200] Step 4:

[1201] The server uses an emotion engine to recognize emotions from the generated reply text. The input is the generated reply text, and the output is emotion data. Specifically, the server uses Microsoft Azure's Text Analytics API.

[1202] Step 5:

[1203] The server adjusts the naturalness and appropriateness of the reply based on the emotional data. The input is the emotional data, and the output is a modified reply that reflects the emotion. In this step, the tone and expression of the document generated by the generative AI model are changed to match the emotion.

[1204] Step 6:

[1205] The server checks the generated reply for grammar and expression. The input is the generated reply, and the output is the grammar- and expression-checked reply. Specifically, it uses the Grammarly API to detect and correct grammatical errors and unnatural expressions.

[1206] Step 7:

[1207] The server automatically classifies emails based on the analysis results. The input is the analysis results, and the output is the classified emails. Specifically, it uses a machine learning model (trained using scikit-learn) to classify emails into pre-defined categories.

[1208] Step 8:

[1209] The server notifies the user's device of the analysis results, the extracted date and time, and the necessary actions. The input is the analysis results and extracted important information, and the output is notification information. Using an emotion engine, the content and method of notification are adjusted based on the user's emotions.

[1210] Step 9:

[1211] The user checks the notification from the server and the generated reply through the terminal. The input is the notification from the server and the generated reply, and the output is the reply that has been checked and corrected by the user. The user corrects the reply as necessary and performs the specific action of issuing a final transmission instruction to the server.

[1212] Step 10:

[1213] The terminal transmits the user's transmission instruction to the server. The input is the user's transmission instruction, and the output is the transmission instruction data to the server.

[1214] Step 11:

[1215] The server sends the final reply to the recipient. The input is the revised final reply, and the output is the result sent to the recipient. Through this series of processes, the AI ​​email assistant system efficiently and effectively supports users' email communication.

[1216] (Application example 2)

[1217] 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."

[1218] There is a need for systems that can reduce the burden of responding to large volumes of inquiries via e-mail and other electronic communications, while also providing appropriate replies based on emotions. However, conventional systems have difficulty taking emotions into account, making it difficult to respond in a way that is sensitive to the user's feelings. It is also difficult to respond quickly while ensuring the appropriateness of grammar and expression in replies.

[1219] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing received electronic communications, means for automatically generating a reply message based on the analysis results, means for checking the grammar and expressions of the generated reply message, means for automatically classifying the electronic communications based on the analyzed content, means for notifying the analysis results and the extracted date and time and response matters, means for using an emotion engine for emotion analysis, and means for generating a reply message based on the emotion. This enables natural replies that are in line with the user's emotions and improves the efficiency and quality of inquiry responses.

[1220] "Means for analyzing received electronic communications" refers to the function of devices such as servers and terminals to analyze the contents of received emails and messages and extract important information.

[1221] "Means for automatically generating a reply based on the analysis results" refers to technology for automatically generating an appropriate reply based on the analyzed data.

[1222] "Means for checking the grammar and expressions of generated replies" refers to a function for detecting and correcting grammatical errors and unnatural expressions in automatically generated replies.

[1223] "Means for automatically classifying electronic communications based on analyzed content" refers to a function that automatically sorts emails and messages into specific categories based on the results of the analysis.

[1224] "Means for notifying the analysis results, extracted dates and times, and action items" refers to a notification function for informing the user of the analysis results, important dates, and action items.

[1225] "Means of using an emotion engine to perform emotion analysis" is a technology for recognizing and analyzing emotions from the text of emails and messages.

[1226] The "means for generating a reply based on emotion" is a technology for automatically generating a reply using expressions and writing styles that correspond to the recognized emotion.

[1227] MODE FOR CARRYING OUT THE INVENTION

[1228] The system for implementing the present invention is composed of three main components: a server, a terminal, and a user. A specific embodiment of the system is shown below.

[1229] Email parsing and reply generation

[1230] When the server receives an electronic communication sent from a user's device, it analyzes the content of the communication using natural language processing (NLP) technology. The analysis is performed with the aim of extracting important information from the electronic communication (e.g., sender, recipient, subject, body, date). Based on the extracted information, the server uses generative AI (e.g., GPT-4) to automatically generate an appropriate reply.

[1231] Example: When a user receives an inquiry email stating that "the product has not arrived," the server analyzes the email and generates a reply message like the following:

[1232] We are very sorry. We checked the tracking information for your order and it was shipped on XX date, but we will reconfirm the delivery status and respond accordingly.

[1233] Use of emotion engine

[1234] During analysis, the server uses an emotion engine (e.g., Google Cloud Natural Language API) to recognize emotions based on the content of the email received by the user and the user's response. The recognized emotions are reflected in the reply generation process. This allows for the generation of natural-looking replies that take the user's emotional state into account.

[1235] Example: For an angry customer, the server uses more polite and careful language. An example of a prompt for the AI ​​generation is, "Read the email about the customer's angry delivery of the item and generate a reply that includes an apology and reconfirmation."

[1236] Grammar and Expression Check

[1237] The server uses a grammar checker (e.g., Grammarly API) to detect grammatical errors and unnatural expressions in the generated reply, correcting them as necessary, and notifying the user of the completed reply.

[1238] Email classification and notifications

[1239] Based on the analysis results, the server automatically classifies electronic communications into pre-defined categories. For example, categories such as "Meetings / Consultations," "Projects," and "Other" may be set. Once classification is complete, the server extracts important dates and times and matters to be addressed, and notifies the user of these on their device. Using an emotion engine, the content and method of notifications can be adjusted based on the user's emotions.

[1240] Example: When an invitation email for a meeting is received, the server classifies the email into the "Meeting / Conference" category, extracts the date and time of the meeting, and notifies the user.

[1241] Subject: (Meeting Notice) Information about a new project meeting

[1242] Notification content:

[1243] A new project meeting is scheduled for the following dates:

[1244] Date and time: October 25, 2023 14:00

[1245] Location: Conference Room 1

[1246] User confirmation and correction

[1247] The user can check the notification from the server and the generated reply message through the terminal. If necessary, the user can modify the reply message and issue a final transmission instruction to the server.

[1248] Sending a reply

[1249] When the user issues a sending instruction, the terminal sends the instruction to the server, which then sends the final reply to the recipient.

[1250] In this way, the system of the present invention allows users to efficiently and appropriately reply to electronic communications received. Furthermore, by utilizing the emotion engine, natural replies that reflect the user's emotions are possible, thereby improving the efficiency and quality of the entire email process.

[1251] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1252] Step 1:

[1253] The server receives the incoming electronic communication. The input is an email or message sent from a user terminal. The server converts the received communication into a format suitable for analysis. The output of this step is text data that is used for analysis.

[1254] Step 2:

[1255] The server analyzes the content of the received electronic communication. The input is the text data generated in step 1. The server uses natural language processing (NLP) techniques to extract important information (sender, recipient, subject, body, date, etc.). Specifically, it uses a text analysis library (e.g., spaCy) for parsing and information extraction. The output of this step is a set of extracted important information.

[1256] Step 3:

[1257] The server uses an emotion engine to recognize emotions contained in the extracted information. The input is the information extracted in step 2. The emotion engine (e.g., Google Cloud Natural Language API) is used to parse emotions from the email body. The output of this step is the recognized emotion data.

[1258] Step 4:

[1259] The server automatically generates a reply based on the analysis results. The input is the set of important information extracted in step 2 and the emotion data obtained in step 3. The server uses a generative AI model (e.g., GPT-4) to set the prompt as follows:

[1260] A customer receives an email saying they didn't receive their item. You recognize their anger. Generate a response based on this information, including an apology and reassurance.

[1261] The output of this step is the generated reply.

[1262] Step 5:

[1263] The server checks the grammar and language of the generated reply. The input is the reply generated in step 4. A grammar checker (e.g., Grammarly API) is used to detect grammatical errors and unnatural language and correct them if necessary. The output of this step is the reply whose grammar and language have been checked and corrected.

[1264] Step 6:

[1265] The server automatically classifies the electronic communications based on the analysis results. The input is the information extracted in step 2. The server performs classification based on pre-defined categories (e.g., "meetings / discussions," "projects," "others"). The output of this step is a set of classified electronic communications.

[1266] Step 7:

[1267] The server notifies the analysis results, extracted dates and times, and actions to be taken. The input is the information extracted in step 2 and the classification results in step 6. The server extracts important dates and times and actions to be taken, and sends a notification to the user device. The output of this step is the notification sent to the user device.

[1268] Step 8:

[1269] The user checks the notification from the server and the generated reply through the terminal. The input is the notification sent in step 7 and the reply corrected in step 5. The user checks the reply and corrects it if necessary. The output of this step is the checked and corrected reply.

[1270] Step 9:

[1271] The user issues a sending instruction, and the terminal sends that instruction to the server. The input is the reply message that was checked and corrected in step 8. The server sends the final reply message to the recipient. The output of this step is the sent reply message.

[1272] 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.

[1273] 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.

[1274] 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.

[1275] [Fourth embodiment]

[1276] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1277] 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.

[1278] 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).

[1279] 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.

[1280] 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.

[1281] 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).

[1282] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1283] 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.

[1284] 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.

[1285] 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.

[1286] 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.

[1287] 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.

[1288] 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."

[1289] This invention is an AI email assistant system that supports efficient and effective email communication. This system consists of three main components: a server, a terminal, and a user.

[1290] Email parsing and reply generation

[1291] When the server receives an email sent from a user's device, it automatically analyzes its contents. This analysis extracts important information (e.g., sender, recipient, subject, body, date). Based on the extracted information, the server uses a generative AI to automatically generate an appropriate reply. The generative AI generates a reply in natural-sounding sentences based on the context and content of the email.

[1292] Example: When a user receives an email proposing a new project, the server parses the email and generates a reply like this:

[1293] Thank you for your continued support.

[1294] I have reviewed your new project proposal. It is very interesting and I would like to give it a positive consideration. Could you please suggest a date for a detailed meeting?

[1295] Thank you very much.

[1296] Grammar and Expression Check

[1297] The server checks the generated reply for grammar and expression. It uses a grammar engine to detect grammatical errors and unnatural expressions and corrects them as necessary. Once the reply has been corrected, it notifies the user.

[1298] Email classification and notifications

[1299] Based on the analysis results, the server automatically classifies emails into pre-defined categories, such as "Meetings / Conferences," "Projects," and "Other." Once classification is complete, the server extracts important dates and times and matters to be addressed, and notifies the user's device.

[1300] Example: When an invitation email for a meeting is received, the server classifies the email into the "Meeting / Conference" category, extracts the date and time of the meeting, and notifies the user.

[1301] Subject: (Meeting Notice) Information about a new project meeting

[1302] Notification content:

[1303] A new project meeting is scheduled for the following dates:

[1304] Date and time: October 25, 2023 14:00

[1305] Location: Conference Room 1

[1306] User confirmation and correction

[1307] The user can check the notification from the server and the generated reply message through the terminal. If necessary, the user can modify the reply message and issue a final transmission instruction to the server.

[1308] Sending a reply

[1309] When the user issues a sending instruction, the terminal sends the instruction to the server, which then sends the final reply to the recipient.

[1310] In this way, the AI ​​email assistant system of the present invention enables users to reply to received emails efficiently and effectively, significantly reducing the burden of email processing in their daily work.

[1311] The processing flow will be explained below.

[1312] Step 1:

[1313] The user receives an email on the terminal, which then sends the received email to the server.

[1314] Step 2:

[1315] The server analyzes the content of the email received, which includes extracting information such as the email text, sender, recipient, subject, and date.

[1316] Step 3:

[1317] Based on the analysis results, the server uses generative AI to automatically generate a reply. Specifically, it selects a reply template based on the content and then uses natural language generation technology to construct the reply.

[1318] Step 4:

[1319] The server performs a grammar and expression check on the generated reply, using a grammar engine to detect grammatical errors and unnatural expressions and correct them as necessary.

[1320] Step 5:

[1321] The server automatically classifies emails based on the analysis results, for example, into categories such as "Meetings / Conferences," "Projects," and "Other."

[1322] Step 6:

[1323] The server extracts information from emails and sends it to the user's device, including meeting schedules and important tasks.

[1324] Step 7:

[1325] The user checks the notification content from the server and the generated reply text on the terminal, and can modify the reply text if necessary.

[1326] Step 8:

[1327] The user issues a transmission instruction for the reply message after editing, and the terminal sends the instruction to the server.

[1328] Step 9:

[1329] The server then sends the final reply to the recipient, allowing users to communicate efficiently via email.

[1330] Example 1

[1331] 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."

[1332] In today's business environment, email is an important means of communication, but processing it requires a great deal of time and effort. Responding appropriately and promptly to the large volume of emails received is particularly difficult, hindering efficient business operations. Manually crafting replies can lead to grammatical errors and unnatural expressions. Important emails can be overlooked, or follow-ups can be delayed because they are not properly categorized. These issues need to be addressed.

[1333] 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.

[1334] In this invention, the server includes means for analyzing received messages, means for automatically generating a response based on the analysis results, means for checking the grammar and expressions of the generated response, means for automatically classifying messages based on the analyzed content, and means for notifying the user of the analysis results, the extracted date and time, and the corresponding matters. This automates a series of processes from receiving emails to generating a reply, checking grammar, automatic classification, and notifying the user of important information, enabling the user to process emails quickly and accurately.

[1335] "Means for parsing received messages" refers to the process by which the server receives messages sent from the user's terminal and disassembles their contents to extract information in a structured format.

[1336] "Means for automatically generating a response based on the analysis results" refers to the process of automatically creating a natural and appropriate reply using a generative AI model based on the analyzed information.

[1337] "Means for checking the grammar and expression of the generated response sentence" refers to the process of detecting and correcting grammatical errors and unnatural expressions in the generated reply sentence.

[1338] "Means for automatically categorizing messages based on analyzed content" refers to a process for categorizing messages into predefined categories (e.g., meetings, projects, etc.) based on analyzed information.

[1339] "Means for notifying the analysis results, extracted date and time, and corresponding matters" refers to the process of notifying the user of important information extracted during the analysis process (date and time, corresponding matters, etc.).

[1340] This invention provides an AI message assistant system that supports efficient and effective electronic message communication. The system consists of three main components: a server, a terminal, and a user.

[1341] Receiving and parsing messages

[1342] server

[1343] The server receives messages sent from the user's device and automatically analyzes their contents. This analysis uses natural language processing libraries (e.g., spaCy and NLTK) to extract important information such as the sender, recipient, subject, body, and date. The received message data is saved in a database, and the analyzed information is stored in temporary storage.

[1344] Automatic generation of response sentences

[1345] server

[1346] Based on the analyzed information, the server uses a generative AI model (e.g., OpenAI's GPT-4) to generate appropriate prompts, which create natural-sounding sentences based on context and content.

[1347] A specific example is the following prompt:

[1348] Sender: Sender name

[1349] Recipient: Recipient name

[1350] Subject: The subject of the message

[1351] Body: The body of the message

[1352] Prompt: Generate a reply to the following message. Responses should be polite and helpful.

[1353] Checking grammar and expressions in responses

[1354] server

[1355] The generated response sentence is checked for grammar and expression using a grammar engine (e.g., Grammarly API). Grammatical errors and unnatural expressions are detected and corrected as necessary. Once corrected, the response sentence is saved back to storage.

[1356] Automatic message classification

[1357] server

[1358] Based on the analysis results, the server classifies messages into predefined categories, such as "Meetings," "Projects," and "Other." Machine learning models (e.g., support vector machines and random forests) are used for classification, and the classification results are stored in a database as message metadata.

[1359] Important Information Notification

[1360] server

[1361] Important information (e.g., date, time, and action) is extracted during the analysis process, and a notification containing this information is generated and sent to the user's device, where the user can visually confirm it.

[1362] Terminal

[1363] The user's device receives the notification sent from the server and displays it to the user, including the date and time, location of the meeting, and any action to be taken.

[1364] User verification and correction

[1365] User

[1366] The user can check the generated response and notification through his / her own terminal, and if necessary, correct the response and issue a final transmission instruction to the server.

[1367] Sending a response

[1368] server

[1369] When the user issues a final command to send the message, the server receives the command and sends the finalized response via the mail server, using the SMTP protocol.

[1370] In this way, the AI ​​message assistant system automates the entire process from receiving messages to generating responses, grammar checking, automatic classification, and notifying users of important information, making message handling more efficient.

[1371] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1372] Step 1:

[1373] Parsing received messages

[1374] Operation explanation

[1375] The server receives messages sent from the user's device. This process uses an email protocol (e.g., IMAP or POP3) and uses a natural language processing library (e.g., spaCy or NLTK) to analyze the content of the received messages.

[1376] input

[1377] Received message data

[1378] output

[1379] Parsed information (sender, recipient, subject, body, date)

[1380] Specific actions

[1381] The server receives new messages, analyzes their contents, extracts important information, and stores it in temporary storage.

[1382] Step 2:

[1383] Automatically generate a response

[1384] Operation explanation

[1385] The server automatically generates an appropriate response sentence using a generative AI model (e.g., OpenAI's GPT-4) based on the analyzed information. In this process, the prompt sentence is input into the generative AI model, which outputs the response sentence.

[1386] input

[1387] Parsed information (sender, recipient, subject, body, date)

[1388] output

[1389] Auto-generated response

[1390] Specific actions

[1391] The server passes the following prompt to the spawned AI:

[1392] From: John Doe

[1393] Recipient: Taro Yamada

[1394] Subject: New project proposal

[1395] Body: I'd like to propose an idea for a new project. I'd like to discuss the details at our next meeting.

[1396] Prompt: Generate a reply to the following message. Responses should be polite and helpful.

[1397] The generation AI then generates a response sentence and stores it in storage.

[1398] Step 3:

[1399] Check the grammar and expression of the generated response

[1400] Operation explanation

[1401] The server passes the generated response to a grammar engine (e.g., Grammarly API) to check the grammar and expressions. If grammatical errors or unnatural expressions are detected, they are corrected.

[1402] input

[1403] Generated response

[1404] output

[1405] Responses with grammar and expression corrections

[1406] Specific actions

[1407] The server sends the generated response sentence to the grammar engine, receives the corrected response sentence, and stores it in storage.

[1408] Step 4:

[1409] Automatically categorize messages

[1410] Operation explanation

[1411] Based on the analyzed information, the server classifies messages into predefined categories using machine learning models (e.g., SVM or Random Forest).

[1412] input

[1413] Analyzed information

[1414] output

[1415] Category information (e.g., meeting, project, etc.)

[1416] Specific actions

[1417] The server categorizes messages and stores the categorization results in a database as message metadata.

[1418] Step 5:

[1419] Notify the extracted date and time and the action to be taken

[1420] Operation explanation

[1421] Important dates and times and action items are extracted during the analysis process and notified to the user. In this notification process, notification data is generated and sent to the user's terminal.

[1422] input

[1423] Analysis results and extracted information

[1424] output

[1425] Notification Data

[1426] Specific actions

[1427] The server generates a notification including the date and time of the meeting and the matters to be addressed, and sends it to the user's terminal.

[1428] Step 6:

[1429] User confirmation and correction

[1430] Operation explanation

[1431] The user can check the notification and the generated response text through the terminal, modify the response text as necessary, and issue a final transmission instruction to the server.

[1432] input

[1433] Notification data, generated response

[1434] output

[1435] Revised Final Response

[1436] Specific actions

[1437] The user checks the response text and notification generated on the terminal, corrects the response text as necessary, and clicks the send button.

[1438] Step 7:

[1439] Sending a final response

[1440] Operation explanation

[1441] When the user issues a final sending instruction, the server receives the instruction and sends the final response via the mail server using the SMTP protocol.

[1442] input

[1443] Revised Final Response

[1444] output

[1445] Sent messages

[1446] Specific actions

[1447] The server sends a final response to the SMTP server and the message is delivered to the recipient.

[1448] In this way, each step works together in a specific way to achieve efficient message processing as a whole system.

[1449] (Application example 1)

[1450] 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."

[1451] Conventional email management systems make it difficult for users to respond to received emails promptly and appropriately. Furthermore, they lack the functionality to analyze email content, automatically generate a reply, and check the grammar of that reply, making it difficult to significantly reduce the amount of work required by users. Especially in workplaces where real-time responses are required, delays in these tasks can lead to reduced work efficiency and potentially inadequate customer service.

[1452] 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.

[1453] In this invention, the server includes a means for analyzing received emails, a means for automatically generating a reply based on the analysis results, and a means for checking the grammar and expressions of the generated reply. This allows the user to be notified of the analysis results, the extracted date and time, and the corresponding action, and to check the reply in real time via a display device. Furthermore, by generating a reply using a prompt sentence using a generative AI model, a more natural and appropriate response is automatically provided, significantly reducing the burden on the user and improving work efficiency.

[1454] "Means for analyzing" is a function for analyzing the contents of received e-mail and extracting important information.

[1455] "Means for automatically generating replies" refers to a function that automatically creates appropriate replies using a generative AI model based on the analyzed content.

[1456] The "means for checking grammar and expressions" is a function for detecting and correcting grammatical errors and unnatural expressions in the generated reply text.

[1457] "Means for automatically classifying e-mails" refers to a function that automatically classifies e-mails into pre-defined categories based on the analyzed content.

[1458] "Means for notifying" is a function that notifies the user of the analysis results, the date and time of extraction, and the necessary measures.

[1459] "Means for displaying a reply message in real time via a display device" refers to a function for displaying the generated reply message to the user in real time using a display device (e.g., a head-mounted display).

[1460] A "generative AI model" is an artificial intelligence model used to generate natural-sounding sentences, taking a prompt as input and outputting an appropriate reply.

[1461] A "prompt sentence" is an input sentence given to a generative AI model, based on which a reply sentence is generated.

[1462] "Means for recording as a calendar event" is a function for recording and managing the analyzed date and time and corresponding matters as a calendar event in the system.

[1463] This invention is an AI email assistant system that enables store staff in brick-and-mortar stores to efficiently handle email reservations and inquiries from customers. The system analyzes received emails, automatically generates appropriate replies, and checks their contents. It also notifies users of the analysis results, the extracted date and time, and the necessary actions, and displays the replies in real time on a display device.

[1464] Hardware and software used

[1465] Hardware: Head-mounted display worn by the store clerk (e.g., Microsoft HoloLens)

[1466] software:

[1467] Generative AI model: A model that generates natural language using TensorFlow

[1468] Web server: Django framework

[1469] HMD Control: Uses OpenCV to manage display and notifications

[1470] System action

[1471] 1. Email analysis

[1472] The server has the means to parse the content of received emails, extracting important information such as sender, recipient, subject, body, date, etc.

[1473] 2. Automatically generate replies

[1474] Based on the analysis results, a generative AI model is used to automatically generate a reply, taking the prompt as input and generating an appropriate reply.

[1475] An example prompt is:

[1476] A customer is asking: Please make an appointment for October 25, 2023 at 2 PM. Please generate a suggested response.

[1477] 3. Grammar and Expression Check

[1478] The generated reply text is checked by a grammar engine on the server, and any errors or unnatural expressions detected are corrected.

[1479] 4. Email Classification and Notification

[1480] Based on the analysis results, emails are automatically categorized into predefined categories such as "reservations" and "inquiries," and important dates and times and actions are extracted.

[1481] 5. Real-time notifications and displays

[1482] The analysis results and reply text are sent to the store clerk in real time via the HMD, allowing the clerk to immediately check the reply and take action if necessary.

[1483] Specific examples

[1484] When a customer emails a store with a reservation, the server parses the email and generates an appropriate reply, such as:

[1485] Thank you for your help. We have accepted your reservation for October 25, 2023 at 2:00 PM. Thank you in advance.

[1486] This reply is displayed in real time on the HMD, allowing the store clerk to check its contents and reply to the customer.

[1487] In this way, the AI ​​email assistant system of the present invention enables store staff to efficiently and effectively handle customer inquiries, which is expected to significantly reduce the burden of email processing in daily work and improve work efficiency.

[1488] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1489] Step 1:

[1490] The server receives emails received from the user's device. To analyze the content of these emails, it extracts important information such as the sender, recipient, subject, body, date, etc. The email is the input, and the extracted information is the output.

[1491] Step 2:

[1492] The server automatically generates a reply based on the analyzed content. Using a generative AI model, it generates an appropriate reply using the prompt as input. Specifically, the prompt is input into the generative AI model, which outputs a natural reply. For example, the following prompt is input:

[1493] A customer is asking: Please make an appointment for October 25, 2023 at 2 PM. Please generate a suggested response.

[1494] The generative AI model outputs an appropriate reply based on this prompt.

[1495] Step 3:

[1496] The server performs grammar and expression checks on the generated reply. It uses a grammar engine to detect grammatical errors and unnatural expressions in the reply and corrects them as necessary. The input is the generated reply, and the output is the corrected or verified reply.

[1497] Step 4:

[1498] The server automatically classifies emails based on the analysis results. It sorts emails into appropriate folders according to predefined categories (e.g., "reservations" or "inquiries"). The input is the analyzed email information, and the output is the classified email.

[1499] Step 5:

[1500] The server notifies the user's device of the analysis results, the extracted date and time, and the corresponding action. Furthermore, this information, including the reply, is displayed in real time on a head-mounted display (HMD). The input is the analysis results and the reply, and the output is the notification and reply displayed on the HMD.

[1501] Step 6:

[1502] The user can check the reply message through the HMD and make corrections if necessary. If corrections are made, the information is sent back to the server, and the final reply message is confirmed. The input is the generated and corrected reply message, and the output is the final reply message.

[1503] Step 7:

[1504] When the server receives the final sending instruction from the user, it sends a confirmed reply to the customer. The input is the final reply, and the output is the completion of the email sending.

[1505] Step 8:

[1506] The server records the analysis results and notifications as a calendar in the system, and can review them later if needed. The input is important dates and times and actions, and the output is a record of the calendar events.

[1507] 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.

[1508] This invention is an AI email assistant system that supports efficient and effective email communication and combines it with an emotion engine that recognizes the user's emotions. This system consists of three main components: a server, a terminal, and a user.

[1509] Email parsing and reply generation

[1510] When the server receives an email sent from a user's device, it automatically analyzes its contents. This analysis extracts important information (e.g., sender, recipient, subject, body, date). Based on the extracted information, the server uses a generative AI to automatically generate an appropriate reply. The generative AI generates a reply in natural-sounding sentences based on the context and content of the email.

[1511] Example: When a user receives an email proposing a new project, the server parses the email and generates a reply like this:

[1512] Thank you for your continued support.

[1513] I have reviewed your new project proposal. It is very interesting and I would like to give it a positive consideration. Could you please suggest a date for a detailed meeting?

[1514] Thank you very much.

[1515] Use of emotion engine

[1516] During analysis, the server uses an emotion engine to recognize emotions based on the content of the email received by the user and the user's response. The recognized emotions are reflected in the reply generation process, allowing for the generation of natural-looking replies that take the user's emotional state into account.

[1517] Grammar and Expression Check

[1518] The server checks the generated reply for grammar and expression. It uses a grammar engine to detect grammatical errors and unnatural expressions and corrects them as necessary. Once the reply has been corrected, it notifies the user.

[1519] Email classification and notifications

[1520] Based on the analysis results, the server automatically classifies emails into pre-defined categories. For example, categories such as "Meetings / Conferences," "Projects," and "Other" can be set. Once classification is complete, the server extracts important dates and times and matters to be addressed, and notifies the user's device of these. Using an emotion engine, the content and method of notifications can be adjusted based on the user's emotions.

[1521] Example: When an invitation email for a meeting is received, the server classifies the email into the "Meeting / Conference" category, extracts the date and time of the meeting, and notifies the user.

[1522] Subject: (Meeting Notice) Information about a new project meeting

[1523] Notification content:

[1524] A new project meeting is scheduled for the following dates:

[1525] Date and time: October 25, 2023 14:00

[1526] Location: Conference Room 1

[1527] User confirmation and correction

[1528] The user can check the notification from the server and the generated reply message through the terminal. If necessary, the user can modify the reply message and issue a final transmission instruction to the server.

[1529] Sending a reply

[1530] When the user issues a sending instruction, the terminal sends the instruction to the server, which then sends the final reply to the recipient.

[1531] In this way, the AI ​​email assistant system of the present invention can efficiently and appropriately reply to emails received by the user. Furthermore, by utilizing the emotion engine, natural replies that reflect the user's emotions are possible, improving the efficiency and quality of the entire email process.

[1532] The processing flow will be explained below.

[1533] Step 1:

[1534] The user receives an email on the terminal, which then sends the received email to the server.

[1535] Step 2:

[1536] The server analyzes the content of the email received, extracting information such as the email text, sender, recipient, subject, and date.

[1537] Step 3:

[1538] The server uses an emotion engine to recognize the user's emotions, which are determined based on the content of the email and the user's past email history.

[1539] Step 4:

[1540] The server automatically generates a reply using generative AI based on the analysis results and the recognized emotions. The generative AI generates natural-sounding sentences taking into account the context and the user's emotions.

[1541] Example: If a user receives an email proposing a new project and the emotion engine recognizes that the user is "interested," the server generates a reply like this:

[1542] Thank you for your continued support.

[1543] I have reviewed your new project proposal. It is very interesting and I would like to give it a positive consideration. Could you please suggest a date for a detailed meeting?

[1544] Thank you very much.

[1545] Step 5:

[1546] The server performs a grammar and expression check on the generated reply, using a grammar engine to detect grammatical errors and unnatural expressions and correct them as necessary.

[1547] Step 6:

[1548] The server automatically classifies emails based on the analysis results, for example, into categories such as "Meetings / Conferences," "Projects," and "Other."

[1549] Step 7:

[1550] The server notifies the user's device of the date and time and action items extracted from the email. Notifications include meeting schedules and important tasks. The emotion engine adjusts the content and method of notifications based on the user's emotions.

[1551] Example: When a user receives an invitation email for a meeting and the emotion engine recognizes "tension," the server notifies the user of the meeting schedule in detail.

[1552] Subject: (Meeting Notice) Information about a new project meeting

[1553] Notification content:

[1554] A new project meeting is scheduled for the following dates:

[1555] Date and time: October 25, 2023 14:00

[1556] Location: Conference Room 1

[1557] Thank you for your preparation.

[1558] Step 8:

[1559] The user checks the notification content from the server and the generated reply text on the terminal, and can modify the reply text if necessary.

[1560] Step 9:

[1561] The user issues a transmission instruction for the reply message after editing, and the terminal sends the instruction to the server.

[1562] Step 10:

[1563] The server then sends the final reply to the recipient, allowing users to communicate efficiently via email.

[1564] Example 2

[1565] 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."

[1566] In conventional email processing systems, the content analysis of received emails, automatic generation of reply messages, grammar and expression checks, and email classification and notification are all performed separately, resulting in problems with consistency and accuracy. Furthermore, reply messages do not reflect the user's feelings, making it difficult to communicate naturally and appropriately.

[1567] 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 analyzing received emails, means for automatically generating reply messages based on the analysis results, means for recognizing the sentiment of the generated reply messages, means for checking grammar and expressions, means for automatically classifying emails based on the analyzed content, and means for notifying the analysis results, extracted date and time, and corresponding matters. This improves the consistency and accuracy of email processing and enables natural replies that reflect the user's sentiment.

[1568] The "means for analyzing received e-mail" is a means for automatically analyzing the contents of received e-mail and extracting important information.

[1569] The "means for automatically generating a reply message based on the analysis results" is a means for automatically generating an appropriate reply message based on the analyzed information.

[1570] The "means for recognizing the emotion of the generated reply message" is a means for analyzing and recognizing the emotion contained in the content of the generated reply message.

[1571] The "means for checking grammar and expressions" refers to a means for checking and correcting grammatical errors and unnatural expressions in the generated reply text.

[1572] The "means for automatically classifying e-mails based on the analyzed content" is a means for automatically classifying e-mails into pre-defined categories based on the analyzed content.

[1573] The "means for notifying the analysis results, extracted dates and times, and corresponding matters" refers to a means for notifying the user of the analysis results and important dates and times and corresponding matters extracted from them.

[1574] This invention is an AI email assistant system that supports efficient and effective email communication and combines it with an emotion engine that recognizes the user's emotions. This system consists of three main components: a server, a terminal, and a user.

[1575] Email parsing and reply generation

[1576] When the server receives an email sent from a user's device, it automatically analyzes the content using NLP (Natural Language Processing) algorithms, particularly tools such as Google Cloud Natural Language API, to extract important information such as the sender, recipient, subject, body, and date.

[1577] Based on the extracted information, the server automatically generates an appropriate reply using generative AI (e.g., OpenAI's GPT-3), which generates a natural-sounding reply based on the context and content of the email.

[1578] As a concrete example, when a user receives an email proposing a new project, the following prompt sentence is input to the generative AI model:

[1579] Generate a sample project proposal email response using the information below.

[1580] Sender: Taro Tanaka

[1581] Recipient: Ichiro Suzuki

[1582] Subject: New Project Proposal

[1583] Body: I ​​would like to propose a new project.

[1584] Date: October 10, 2023

[1585] Based on this prompt, the following reply will be generated:

[1586] Thank you for your continued support.

[1587] I have reviewed your new project proposal. It is very interesting and I would like to give it a positive consideration. Could you please suggest a date for a detailed meeting?

[1588] Thank you very much.

[1589] Use of emotion engine

[1590] The server uses an emotion engine for analysis. This emotion engine uses, for example, the emotion analysis function of Microsoft Azure's Text Analytics API. The emotion engine recognizes emotions based on the content of the email received by the user and the user's response. This recognized emotion is reflected in the reply generation process. This allows for the generation of natural-looking replies that take the user's emotional state into account.

[1591] Grammar and Expression Check

[1592] The server then checks the generated reply for grammar and expression. Specifically, it uses a grammar engine such as the Grammarly API to detect grammatical errors and unnatural expressions and corrects them as necessary. Once the corrections are complete, the user is notified.

[1593] Email classification and notifications

[1594] The server automatically classifies emails into predefined categories based on the analysis results. For example, a machine learning model (trained with scikit-learn) can be used. Once classification is complete, the server extracts important dates and times and actions to take and notifies the user's device. An emotion engine can be used to adjust the content and method of notifications based on the user's emotions.

[1595] For example, when a user receives an invitation email for a meeting, the server classifies the email into the "Meeting / Conference" category, extracts the date and time of the meeting, and notifies the user:

[1596] Subject: (Meeting Notice) Information about a new project meeting

[1597] Notification content:

[1598] A new project meeting is scheduled for the following dates:

[1599] Date and time: October 25, 2023 14:00

[1600] Location: Conference Room 1

[1601] User confirmation and correction

[1602] The user can check the notification from the server and the generated reply message through the terminal. If the reply message is not appropriate, the user can correct it and issue a final transmission instruction to the server.

[1603] Sending a reply

[1604] When a user issues a sending instruction, the device sends the instruction to the server, which then sends the final reply to the recipient. Through this series of processes, the AI ​​email assistant system efficiently and effectively supports users' email communication.

[1605] The above is a specific embodiment for carrying out the present invention. The combination of the generative AI model and emotion engine, which is a feature of this invention, significantly improves the efficiency and quality of email processing.

[1606] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1607] Step 1:

[1608] The server receives emails sent from the user's terminal. The input is the email information received from the terminal, and the output is the received email data. The server performs the specific operation of retrieving new emails from the email server using the IMAP or POP3 protocol.

[1609] Step 2:

[1610] The server analyzes the content of received emails. The input is the received email data, and the output is the analysis results (e.g., sender, recipient, subject, body, date). NLP algorithms are used for the analysis, and specific operations are performed using, for example, the Google Cloud Natural Language API.

[1611] Step 3:

[1612] The server generates a reply using a generative AI model (e.g., OpenAI's GPT-3) based on the analyzed email content. The input is the analysis result, and the output is the generated reply. Specifically, the following prompt is input to the generative AI model:

[1613] Generate a sample project proposal email response using the information below.

[1614] Sender: Taro Tanaka

[1615] Recipient: Ichiro Suzuki

[1616] Subject: New Project Proposal

[1617] Body: I ​​would like to propose a new project.

[1618] Date: October 10, 2023

[1619] Step 4:

[1620] The server uses an emotion engine to recognize emotions from the generated reply text. The input is the generated reply text, and the output is emotion data. Specifically, the server uses Microsoft Azure's Text Analytics API.

[1621] Step 5:

[1622] The server adjusts the naturalness and appropriateness of the reply based on the emotional data. The input is the emotional data, and the output is a modified reply that reflects the emotion. In this step, the tone and expression of the document generated by the generative AI model are changed to match the emotion.

[1623] Step 6:

[1624] The server checks the generated reply for grammar and expression. The input is the generated reply, and the output is the grammar- and expression-checked reply. Specifically, it uses the Grammarly API to detect and correct grammatical errors and unnatural expressions.

[1625] Step 7:

[1626] The server automatically classifies emails based on the analysis results. The input is the analysis results, and the output is the classified emails. Specifically, it uses a machine learning model (trained using scikit-learn) to classify emails into pre-defined categories.

[1627] Step 8:

[1628] The server notifies the user's device of the analysis results, the extracted date and time, and the necessary actions. The input is the analysis results and extracted important information, and the output is notification information. Using an emotion engine, the content and method of notification are adjusted based on the user's emotions.

[1629] Step 9:

[1630] The user checks the notification from the server and the generated reply through the terminal. The input is the notification from the server and the generated reply, and the output is the reply that has been checked and corrected by the user. The user corrects the reply as necessary and performs the specific action of issuing a final transmission instruction to the server.

[1631] Step 10:

[1632] The terminal transmits the user's transmission instruction to the server. The input is the user's transmission instruction, and the output is the transmission instruction data to the server.

[1633] Step 11:

[1634] The server sends the final reply to the recipient. The input is the revised final reply, and the output is the result sent to the recipient. Through this series of processes, the AI ​​email assistant system efficiently and effectively supports users' email communication.

[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] There is a need for systems that can reduce the burden of responding to large volumes of inquiries via e-mail and other electronic communications, while also providing appropriate replies based on emotions. However, conventional systems have difficulty taking emotions into account, making it difficult to respond in a way that is sensitive to the user's feelings. It is also difficult to respond quickly while ensuring the appropriateness of grammar and expression in replies.

[1638] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing received electronic communications, means for automatically generating a reply message based on the analysis results, means for checking the grammar and expressions of the generated reply message, means for automatically classifying the electronic communications based on the analyzed content, means for notifying the analysis results and the extracted date and time and response matters, means for using an emotion engine for emotion analysis, and means for generating a reply message based on the emotion. This enables natural replies that are in line with the user's emotions and improves the efficiency and quality of inquiry responses.

[1639] "Means for analyzing received electronic communications" refers to the function of devices such as servers and terminals to analyze the contents of received emails and messages and extract important information.

[1640] "Means for automatically generating a reply based on the analysis results" refers to technology for automatically generating an appropriate reply based on the analyzed data.

[1641] "Means for checking the grammar and expressions of generated replies" refers to a function for detecting and correcting grammatical errors and unnatural expressions in automatically generated replies.

[1642] "Means for automatically classifying electronic communications based on analyzed content" refers to a function that automatically sorts emails and messages into specific categories based on the results of the analysis.

[1643] "Means for notifying the analysis results, extracted dates and times, and action items" refers to a notification function for informing the user of the analysis results, important dates, and action items.

[1644] "Means of using an emotion engine to perform emotion analysis" is a technology for recognizing and analyzing emotions from the text of emails and messages.

[1645] The "means for generating a reply based on emotion" is a technology for automatically generating a reply using expressions and writing styles that correspond to the recognized emotion.

[1646] MODE FOR CARRYING OUT THE INVENTION

[1647] The system for implementing the present invention is composed of three main components: a server, a terminal, and a user. A specific embodiment of the system is shown below.

[1648] Email parsing and reply generation

[1649] When the server receives an electronic communication sent from a user's device, it analyzes the content of the communication using natural language processing (NLP) technology. The analysis is performed with the aim of extracting important information from the electronic communication (e.g., sender, recipient, subject, body, date). Based on the extracted information, the server uses generative AI (e.g., GPT-4) to automatically generate an appropriate reply.

[1650] Example: When a user receives an inquiry email stating that "the product has not arrived," the server analyzes the email and generates a reply message like the following:

[1651] We are very sorry. We checked the tracking information for your order and it was shipped on XX date, but we will reconfirm the delivery status and respond accordingly.

[1652] Use of emotion engine

[1653] During analysis, the server uses an emotion engine (e.g., Google Cloud Natural Language API) to recognize emotions based on the content of the email received by the user and the user's response. The recognized emotions are reflected in the reply generation process. This allows for the generation of natural-looking replies that take the user's emotional state into account.

[1654] Example: For an angry customer, the server uses more polite and careful language. An example of a prompt for the AI ​​generation is, "Read the email about the customer's angry delivery of the item and generate a reply that includes an apology and reconfirmation."

[1655] Grammar and Expression Check

[1656] The server uses a grammar checker (e.g., Grammarly API) to detect grammatical errors and unnatural expressions in the generated reply, correcting them as necessary, and notifying the user of the completed reply.

[1657] Email classification and notifications

[1658] Based on the analysis results, the server automatically classifies electronic communications into pre-defined categories. For example, categories such as "Meetings / Consultations," "Projects," and "Other" may be set. Once classification is complete, the server extracts important dates and times and matters to be addressed, and notifies the user of these on their device. Using an emotion engine, the content and method of notifications can be adjusted based on the user's emotions.

[1659] Example: When an invitation email for a meeting is received, the server classifies the email into the "Meeting / Conference" category, extracts the date and time of the meeting, and notifies the user.

[1660] Subject: (Meeting Notice) Information about a new project meeting

[1661] Notification content:

[1662] A new project meeting is scheduled for the following dates:

[1663] Date and time: October 25, 2023 14:00

[1664] Location: Conference Room 1

[1665] User confirmation and correction

[1666] The user can check the notification from the server and the generated reply message through the terminal. If necessary, the user can modify the reply message and issue a final transmission instruction to the server.

[1667] Sending a reply

[1668] When the user issues a sending instruction, the terminal sends the instruction to the server, which then sends the final reply to the recipient.

[1669] In this way, the system of the present invention allows users to efficiently and appropriately reply to electronic communications received. Furthermore, by utilizing the emotion engine, natural replies that reflect the user's emotions are possible, thereby improving the efficiency and quality of the entire email process.

[1670] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1671] Step 1:

[1672] The server receives the incoming electronic communication. The input is an email or message sent from a user terminal. The server converts the received communication into a format suitable for analysis. The output of this step is text data that is used for analysis.

[1673] Step 2:

[1674] The server analyzes the content of the received electronic communication. The input is the text data generated in step 1. The server uses natural language processing (NLP) techniques to extract important information (sender, recipient, subject, body, date, etc.). Specifically, it uses a text analysis library (e.g., spaCy) for parsing and information extraction. The output of this step is a set of extracted important information.

[1675] Step 3:

[1676] The server uses an emotion engine to recognize emotions contained in the extracted information. The input is the information extracted in step 2. The emotion engine (e.g., Google Cloud Natural Language API) is used to parse emotions from the email body. The output of this step is the recognized emotion data.

[1677] Step 4:

[1678] The server automatically generates a reply based on the analysis results. The input is the set of important information extracted in step 2 and the emotion data obtained in step 3. The server uses a generative AI model (e.g., GPT-4) to set the prompt as follows:

[1679] A customer receives an email saying they didn't receive their item. You recognize their anger. Generate a response based on this information, including an apology and reassurance.

[1680] The output of this step is the generated reply.

[1681] Step 5:

[1682] The server checks the grammar and language of the generated reply. The input is the reply generated in step 4. A grammar checker (e.g., Grammarly API) is used to detect grammatical errors and unnatural language and correct them if necessary. The output of this step is the reply whose grammar and language have been checked and corrected.

[1683] Step 6:

[1684] The server automatically classifies the electronic communications based on the analysis results. The input is the information extracted in step 2. The server performs classification based on pre-defined categories (e.g., "meetings / discussions," "projects," "others"). The output of this step is a set of classified electronic communications.

[1685] Step 7:

[1686] The server notifies the analysis results, extracted dates and times, and actions to be taken. The input is the information extracted in step 2 and the classification results in step 6. The server extracts important dates and times and actions to be taken, and sends a notification to the user device. The output of this step is the notification sent to the user device.

[1687] Step 8:

[1688] The user checks the notification from the server and the generated reply through the terminal. The input is the notification sent in step 7 and the reply corrected in step 5. The user checks the reply and corrects it if necessary. The output of this step is the checked and corrected reply.

[1689] Step 9:

[1690] The user issues a sending instruction, and the terminal sends that instruction to the server. The input is the reply message that was checked and corrected in step 8. The server sends the final reply message to the recipient. The output of this step is the sent reply message.

[1691] 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.

[1692] 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.

[1693] 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 robot 414.

[1694] 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.

[1695] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1696] 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.

[1697] 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).

[1698] 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.

[1699] 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."

[1700] 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.

[1701] 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).

[1702] 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.

[1703] 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.

[1704] 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.

[1705] 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.

[1706] 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.

[1707] 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.

[1708] 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.

[1709] 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.

[1710] 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.

[1711] 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.

[1712] The following is further disclosed regarding the above embodiment.

[1713] (Claim 1)

[1714] means for analyzing received emails;

[1715] A means for automatically generating a reply based on the analysis results;

[1716] means for checking the grammar and expression of the generated reply;

[1717] means for automatically classifying emails based on the analyzed content;

[1718] A means of notifying the analysis results, the extracted date and time, and the corresponding action items;

[1719] A system including:

[1720] (Claim 2)

[1721] 10. The system of claim 1, further comprising means for correcting grammatical and phrasing errors in the generated reply.

[1722] (Claim 3)

[1723] 10. The system of claim 1, further comprising means for recording the date, time and corresponding event as a calendar event.

[1724] "Example 1"

[1725] (Claim 1)

[1726] means for analyzing received messages;

[1727] means for automatically generating a response sentence based on the analysis result;

[1728] means for checking the grammar and expression of the generated response sentence;

[1729] means for automatically classifying messages based on the analyzed content;

[1730] A means of notifying the analysis results, the extracted date and time, and the corresponding action items;

[1731] A system including:

[1732] (Claim 2)

[1733] 10. The system of claim 1, further comprising means for correcting grammatical and phrasal errors in the generated response sentence.

[1734] (Claim 3)

[1735] 10. The system of claim 1, further comprising means for recording the date, time and corresponding event as a calendar event.

[1736] "Application Example 1"

[1737] (Claim 1)

[1738] means for analyzing received emails;

[1739] A means for automatically generating a reply based on the analysis results;

[1740] means for checking the grammar and expression of the generated reply;

[1741] means for automatically classifying emails based on the analyzed content;

[1742] A means of notifying the analysis results, the extracted date and time, and the corresponding action items;

[1743] means for displaying the reply message in real time via a display device;

[1744] A system including:

[1745] (Claim 2)

[1746] 10. The system of claim 1, further comprising means for correcting grammatical and phrasing errors in the generated reply.

[1747] (Claim 3)

[1748] 10. The system of claim 1, further comprising means for recording the date, time and corresponding event as a calendar event.

[1749] (Claim 4)

[1750] 10. The system according to claim 1, further comprising means for displaying the analysis results and notification contents on a display device in real time.

[1751] (Claim 5)

[1752] 2. The system of claim 1, further comprising means for generating a reply sentence based on the analysis result using the prompt sentence using a generative AI model.

[1753] "Example 2: Combining Emotion Engines"

[1754] (Claim 1)

[1755] means for analyzing received emails;

[1756] A means for automatically generating a reply based on the analysis results;

[1757] a means for recognizing the sentiment of the generated reply;

[1758] a means of checking grammar and expression;

[1759] means for automatically classifying emails based on the analyzed content;

[1760] A means of notifying the analysis results, the extracted date and time, and the corresponding action items;

[1761] A system including:

[1762] (Claim 2)

[1763] 10. The system of claim 1, further comprising means for correcting grammatical and phrasing errors in the generated reply.

[1764] (Claim 3)

[1765] 10. The system of claim 1, further comprising means for recording the date, time and corresponding event as a calendar event.

[1766] "Application example 2 when combining emotion engines"

[1767] (Claim 1)

[1768] means for analyzing received electronic communications;

[1769] A means for automatically generating a reply based on the analysis results;

[1770] means for checking the grammar and expression of the generated reply;

[1771] means for automatically classifying electronic communications based on the analyzed content;

[1772] A means of notifying the analysis results, the extracted date and time, and the corresponding action items;

[1773] a means for utilizing a sentiment engine for performing sentiment analysis;

[1774] a means for generating a response based on the sentiment;

[1775] A system including:

[1776] (Claim 2)

[1777] 10. The system of claim 1, further comprising means for correcting grammatical and phrasing errors in the generated reply.

[1778] (Claim 3)

[1779] 10. The system of claim 1, further comprising means for recording the date, time and corresponding event as a calendar event. [Explanation of symbols]

[1780] 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 analyzing received emails; A means for automatically generating a reply based on the analysis results; means for checking the grammar and expression of the generated reply; means for automatically classifying emails based on the analyzed content; A means of notifying the analysis results, the extracted date and time, and the corresponding action items; A system including:

2. 10. The system of claim 1, further comprising means for correcting grammatical and phrasing errors in the generated reply.

3. 10. The system of claim 1, further comprising means for recording the date, time and corresponding event as a calendar event.

Images