System

Abstract

To provide a system capable of automatically recording call contents and easily confirming them.SOLUTION: The terminal automatically starts voice recording when the user starts a call, and the recorded voice is saved and uploaded to the server when the call ends. In the server, when the uploaded voice file is received, a voice recognition engine is operated, and the voice file is converted into text data and recorded. The server inputs the converted text data to the generation model and generates a summary. The server stores the summarized text data in a database and links it to the user's call history. With this link, the summary is immediately displayed when the user checks the history of incoming and outgoing calls.SELECTED DRAWING: Figure 11

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] In recent years, the number of conversations and business-related interactions conducted over the telephone has increased. However, due to the difficulty of recording the content of these calls, it is common for employees to forget what was said or important appointments or instructions. This problem can lead to reduced work efficiency and trouble due to memory errors. Furthermore, manually taking notes on phone calls is time-consuming and inaccurate, further complicating the problem. Therefore, there is a need for a method to automatically record and easily review the content of phone calls. [Means for solving the problem]

[0005] To solve this problem, the present invention provides the following means. First, it provides a means for automatically recording the contents of a call. It also provides a means for converting the recorded voice data into text data using a speech recognition engine. It then provides a means for summarizing the text data using a generative model. The summarized text data is saved in association with the call history and displayed so that the user can easily check it. This makes it possible to remember the contents of the call and instantly check important information, contributing to improved work efficiency and the prevention of problems.

[0006] "Call content" means the spoken information and communications exchanged over the telephone.

[0007] "Recording" is the act of recording audio data in digital or analog form and storing it for later playback.

[0008] "Audio data" is a digital representation of sounds such as human voices.

[0009] A "speech recognition engine" is a collection of software and algorithms used to convert speech into text.

[0010] "Text data" refers to information expressed as a string of characters.

[0011] A "generative model" is a machine learning algorithm or AI system that can summarize, translate, or generate information from input text.

[0012] A "summary" is a short text that concisely summarizes long text data and extracts only the important content.

[0013] The "call history" is a chronological record of the user's outgoing and incoming calls.

[0014] "Association" is the act of establishing a relationship between different data, allowing one piece of data to reference another piece of data.

[0015] "Display" is the act of visually providing information to a user, and is done through a screen or display. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0017] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0019] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0020] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0021] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0022] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0024] [First embodiment]

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

[0026] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0029] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0033] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0037] The present invention provides a system that automatically memos the contents of telephone conversations and allows users to check the memos with a single touch from the call history. This system is mainly composed of three entities: a terminal, a server, and a user.

[0038] Recording and uploading audio

[0039] The device automatically starts recording audio when the user starts a call. The recorded audio is saved when the call ends and uploaded to a server, allowing the call contents to be stored in digital form.

[0040] Speech-to-text conversion

[0041] When the server receives the uploaded audio file, a speech recognition engine is activated and converts the audio data into text data. For example, if a user says, "The next meeting is on Tuesday," this is recorded as text.

[0042] Text summary

[0043] The server inputs the converted text data into a generative model (e.g., GPT-3) to generate a summary. For example, the text "The next meeting is on Tuesday" is summarized as "Next meeting: Tuesday." This summary aims to simplify the information.

[0044] Save summary and link to call history

[0045] The server stores the summarized text data in a database and links it to the user's call history, so that when the user checks their call history, the summary text is immediately displayed.

[0046] Displaying notes in call history

[0047] When users check their call history, the device will display the associated summary, allowing them to easily check the content of the call. For example, simply tapping on a call history will display the summary "Next meeting: Tuesday."

[0048] Specific examples

[0049] Here's a concrete example: A user talks to a sales representative on the phone and says the following:

[0050] "Please schedule a meeting with a client for next Tuesday in the office conference room."

[0051] Calls are recorded and uploaded to a server

[0052] The server converts the speech to text, generating the following text: "Please schedule a meeting with the client for next Tuesday in a conference room in our office."

[0053] The generative model summarizes this as "Customer meeting: Tuesday, office conference room."

[0054] This summary is stored in a database and linked to the call history.

[0055] When users check their call history, a summary is displayed, allowing them to see the details of the call at a glance.

[0056] In this way, the present invention allows users to record and quickly check the contents of calls, thereby preventing users from overlooking important information and improving work efficiency.

[0057] The processing flow will be explained below.

[0058] Step 1:

[0059] The user starts a call. This causes the device to automatically prepare to start recording audio. Specifically, the device's recording function is turned on, and the device is ready to record the audio data of the call.

[0060] Step 2:

[0061] The user ends the call. When the call ends, the device stops recording and saves the recorded audio data as a file. This audio file is uploaded to the server as soon as the call ends.

[0062] Step 3:

[0063] The server receives the uploaded audio file. The server then activates a speech recognition engine and converts the audio data into text data. For example, the server passes the audio file to the speech recognition engine and obtains the text data, "The next meeting is on Tuesday."

[0064] Step 4:

[0065] The server inputs the converted text data into a generative model, which then summarizes the text data into a concise form. For example, the text data "The next meeting is on Tuesday" is summarized as "Next meeting: Tuesday."

[0066] Step 5:

[0067] The server saves the summarized text data in a database. At the same time, it associates this summary data with the user's call history, allowing the summary data to be referenced from the call history.

[0068] Step 6:

[0069] The user checks the call history. The device retrieves the summary data associated with the call history and displays it on the screen. The user can easily check the related summary data by simply tapping the call history.

[0070] Step 7:

[0071] The user checks the summary data and transcribes it into a memo or calendar as necessary. This step allows the user to smoothly move on to the next action without forgetting any important information about the call content.

[0072] Example 1

[0073] 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."

[0074] Conventional call recording systems require the manual recording of call content, which can lead to important information being overlooked. Furthermore, reviewing call content takes time, and there are few ways to improve work efficiency. Therefore, there is a need for a system that automatically records call content and can efficiently summarize and search it.

[0075] 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.

[0076] In this invention, the server includes a means for recording the contents of the call, a means for converting the recorded voice data into text data using a voice recognition system, a means for summarizing the text data using a generative AI model, a means for storing and displaying the summarized text data in association with the communication history, and a means for displaying the summary when the user checks the communication history. This enables the contents of the call to be automatically recorded and efficiently summarized and searched.

[0077] "Call content" refers to information and conversations exchanged over the telephone or other voice communications.

[0078] "Recording" refers to the process of storing audio data in digital or analog form.

[0079] "Audio Data" refers to audio information recorded in digital or analog format.

[0080] A "speech recognition system" refers to the technology that allows a computer to understand human speech and convert it into text.

[0081] "Text data" refers to digital data expressed as character information.

[0082] A "generative AI model" refers to an artificial intelligence model that learns using large amounts of data and makes predictions and generates results.

[0083] A "summary" is a short sentence that concisely summarizes the original information.

[0084] "Communication history" refers to a record of calls and messages made, including information such as the caller, recipient, and time.

[0085] "Storage" refers to keeping data or information in a fixed location.

[0086] "Display" refers to outputting information such as text or graphics to a screen or display.

[0087] "User" refers to a person who uses a system or application.

[0088] The present invention is a system that automatically records and summarizes the contents of phone calls and associates them with call history. This system is mainly composed of three entities: a terminal, a server, and a user.

[0089] Recording and uploading audio

[0090] When a user starts a call, the device automatically starts recording audio. This function uses the device's built-in microphone and a voice recording app (for example, the Voice Memos app on iOS). When the call ends, the recorded audio file is saved in a specific folder on the device. The recorded file is then uploaded to the server using a POST request using the HTTP protocol.

[0091] Speech-to-text conversion

[0092] When the server receives the uploaded audio file, a speech recognition system (e.g., Google Cloud Speech-to-Text API) is activated to convert the audio data into text data. At this time, the audio file is saved in a storage server (e.g., Amazon S3) and then analyzed by the speech recognition system. For example, if a user says, "The next meeting is on Tuesday," this content is recorded as text data.

[0093] Text summary

[0094] The server inputs the generated text data into a generative AI model (e.g., OpenAI's GPT-3) to generate a summary. As a specific example, the server receives the text "The next meeting is on Tuesday" and sends the prompt "Please summarize this text: 'The next meeting is on Tuesday.'" to the generative model. As a result, the generative model generates the summary "Next meeting: Tuesday."

[0095] Save summary and link to call history

[0096] The server stores the summary in a database (e.g., MySQL) and links it to the user's call history. This linking allows the user to instantly retrieve the summary when checking the call history. For example, by associating the call summary data with the call date and time and the other party's phone number, the call history can be managed effectively.

[0097] Displaying notes in call history

[0098] When a user checks the call history on their device, the device requests the summary stored on the server. The server then sends back the corresponding summary, which is then displayed in the device's call history app. This process allows users to instantly view summaries such as "Next meeting: Tuesday" by simply tapping on the call history.

[0099] Example operation

[0100] As a real-world use case, imagine a user is talking to a sales rep on the phone and says, "Please schedule a meeting with a customer next Tuesday in a conference room in my office." The system would automatically execute the following process:

[0101] 1. The call will be recorded and an audio file will be created using your device's built-in microphone and audio recording app.

[0102] 2. The audio file is uploaded to the server.

[0103] 3. The server converts the received voice file into text data using a voice recognition system, and the generated text is, "Please schedule a meeting with the client for next Tuesday. The location should be in the office conference room."

[0104] 4. This text data is fed into a generative AI model, which generates a summary: "Customer meeting: Tuesday, office conference room."

[0105] 5. The summary is stored in a database and linked to the call history.

[0106] 6. When the user checks their call history, a summary of "Customer Meeting: Tuesday, Office Conference Room" is instantly displayed.

[0107] In this way, the present invention allows users to record and quickly check the contents of calls, and improves work efficiency without overlooking important information.

[0108] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0109] Step 1:

[0110] The device automatically starts recording audio when the user starts a call. Specifically, the device's call management app detects the call start event and activates the audio recording module. Once recording begins, audio data is collected in real time and temporarily stored in the device's internal memory.

[0111] Input: User call start event

[0112] Output: Audio data being recorded

[0113] Step 2:

[0114] The device stops recording when the call ends and saves the collected audio data in a specific folder. The call management app detects the call end event and stops the recording module, completing the recording. The recording file is then uploaded to the server using the HTTP protocol.

[0115] Input: Call end event and recording audio data

[0116] Output: Audio file uploaded to the server

[0117] Step 3:

[0118] The server receives the HTTP POST request and saves the uploaded audio file on the storage server. Next, it calls the speech recognition system to convert the audio data into text data. This process takes the audio file and outputs it as text using the speech recognition API. For example, if the audio data says "The next meeting is on Tuesday," it will be converted into text data.

[0119] Input: Uploaded audio file

[0120] Output: Converted text data

[0121] Step 4:

[0122] The server inputs the generated text data into the generative AI model to generate a summary. Specifically, it forms a prompt sentence and sends an API request to the generative model. The text "The next meeting is on Tuesday" is input into the generative AI model, and the summary "Next meeting: Tuesday" is returned.

[0123] Input: Generated text data

[0124] Output: Summarized text data

[0125] Step 5:

[0126] The server stores the summary in a database and links it to the call history, which contains metadata such as the call date and time and the caller's phone number, and associates the summary with that information.

[0127] Input: Summarized text data

[0128] Output: A link between the summary stored in the database and the call history

[0129] Step 6:

[0130] When a user checks the call history on their device, the device sends a request to the server to retrieve the stored summary. When the user opens the call history app and taps on a specific call history, the summary linked to that call history is displayed on the device.

[0131] Input: User operation to check call history

[0132] Output: A summary of the call displayed on your device

[0133] (Application example 1)

[0134] 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."

[0135] While current call recording systems can accurately record call content, it is difficult to quickly and efficiently understand the content. Furthermore, they lack the functionality to store call records on the cloud and play them back as needed, creating a need for improved security and operational efficiency.

[0136] 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.

[0137] In this invention, the server includes means for recording the contents of the call, means for converting the recorded voice data into text data using a voice recognition engine, means for summarizing the text data using a generative model, means for saving and displaying the summarized text data in association with the call history, means for uploading and saving the voice data and the summarized text to cloud storage, and means for restoring and playing back the uploaded voice data. This enables quick and efficient understanding of the contents of the call, and improves security and business efficiency through data management on the cloud.

[0138] "Call content" refers to all words and information spoken during a call.

[0139] "Recording means" refers to any device or method that physically or digitally records the audio of a call.

[0140] "Audio Data" refers to a digital audio file that records the contents of a call.

[0141] A "voice recognition engine" refers to software or algorithms that have the function of analyzing voice data and converting it into text data.

[0142] "Text data" refers to the textual information of the speech content generated by a speech recognition engine.

[0143] A "generative model" refers to a machine learning model used to generate a summary or output in a specific format from input data.

[0144] "Means for summarizing" refers to a method or device that analyzes text data, extracts only the important points, and shortens them.

[0145] "Call history" refers to databases and log information that record the date and time of past calls, the callers, etc.

[0146] The "means for storing and displaying in association" refers to a method or device for linking the summarized text data to the call history, and storing and displaying the data so that the user can easily view it.

[0147] "Cloud storage" refers to an online storage service that allows you to store data remotely via the Internet.

[0148] "Uploading and storage means" refers to the methods and devices for transferring and securely storing audio data and summary text in cloud storage.

[0149] "Means for restoring and playing" refers to a method or device for downloading audio data stored in cloud storage and playing it back as the original audio.

[0150] An embodiment of the present invention will now be described in detail. This system is mainly composed of three elements: a terminal, a server, and a user.

[0151] Recording and uploading audio

[0152] The device automatically starts recording audio when the user starts a call. The recorded audio data is temporarily stored on the device when the call ends and then uploaded to cloud storage. For example, if you are using smart glasses, the microphone will capture the audio and the recording file will be saved in cloud storage.

[0153] Speech-to-text conversion

[0154] When the server receives the uploaded audio file, a speech recognition engine is activated and converts the audio data into text data. The speech recognition engine used in this process uses, for example, the Python speech_recognition library. For example, if a user says, "The next meeting is on Tuesday," this is accurately recorded as text.

[0155] Text summary

[0156] The server generates a summary using a generative model (e.g., an AI model such as BERT or GPT) to further simplify the converted text data. For example, the text "The next meeting is on Tuesday" is summarized as "Meeting: Tuesday." This summary is generated by inputting a prompt sentence to the generative AI model.

[0157] Save summary and link to call history

[0158] The server stores the summarized text data in a cloud database and links it to the user's call history. This allows the summary to be displayed instantly when the user checks the call history. For example, by simply tapping on a call history, the summary "Meeting: Tuesday" can be displayed.

[0159] Displaying notes in call history

[0160] When a user checks the call history, the device will display the associated summary, allowing the user to easily check the content of the call. For example, the summary "Meeting: Tuesday" will be displayed on the smart glasses display.

[0161] Data management in cloud storage

[0162] The recorded audio data and generated summary text are uploaded to cloud storage and stored securely, allowing for efficient management of large amounts of data. The stored audio data can also be restored and played back as needed.

[0163] Specific prompt examples

[0164] Conversation: "I have a conference call with a client tomorrow at 2 PM. Please send the materials by email."

[0165] Summary: "Meeting: Tomorrow at 2pm, materials to be sent."

[0166] An example of a prompt to input to a generative AI model: "Please summarize the following text: I have a conference call with a client tomorrow at 2 PM. Please also send the materials by email."

[0167] The above is a form for implementing the present invention. By using this system, users can quickly and efficiently understand the content of calls, and furthermore, data management on the cloud can improve security and business efficiency.

[0168] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0169] Step 1:

[0170] When a call is initiated, the device automatically records the call. The input for the recording is the audio from the call, and the output is audio data (e.g., a .wav file). This audio data is stored in the device's temporary memory.

[0171] Step 2:

[0172] When the call ends, the device automatically uploads the recorded audio data to cloud storage. The input is the audio data generated in step 1, and the output is the audio file stored in cloud storage.

[0173] Step 3:

[0174] The server receives the audio data uploaded from the cloud storage. The input is the audio data stored in the cloud storage, and the output is the audio data stored in the server's storage device.

[0175] Step 4:

[0176] The server uses a speech recognition engine to convert the received voice data into text data. The input is voice data (e.g., a .wav file), and the output is the text of the conversation. Specifically, the server uses the Python speech_recognition library to analyze the voice data and convert it into text information.

[0177] Step 5:

[0178] The server uses a generative AI model to summarize the converted text data. The input is the text data generated in step 4, and the output is summarized text information. A prompt sentence is used to summarize the input text for the generative AI model (e.g., GPT-3). An example of a specific prompt sentence is, "Please summarize the following text: I have a conference call with a client tomorrow at 2 p.m. Please also send the materials by email."

[0179] Step 6:

[0180] The server associates the summarized text data with the call history and stores it in a database. The input is the summarized text data and the call history information, and the output is the summary text linked to the call history. This allows the relevant summary information to be displayed immediately when the user checks the call history.

[0181] Step 7:

[0182] When a user checks the call history, the terminal displays the associated summary text. The input is the summary text obtained from the call history, and the output is the summary information displayed on the user interface, allowing the user to quickly check the key points of the call.

[0183] Step 8:

[0184] The server or device provides the functionality to restore and play back the voice data stored in the cloud storage as needed. The input is the voice file stored in the cloud storage, and the output is the voice output from the playback device (e.g., speaker or headset). This allows the user to reconfirm the contents of the call.

[0185] 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.

[0186] An embodiment of the present invention will be described in detail. The present invention is a system that combines a system that automatically takes notes on phone conversations and allows users to check the notes with a single touch from their call history, with an emotion engine that recognizes the user's emotions. This system is composed of three entities: a terminal, a server, and a user.

[0187] Recording and uploading audio

[0188] When a user starts a call, the device automatically starts recording the audio. The audio data of the call is recorded and uploaded to the server when the call ends.

[0189] Speech-to-text conversion

[0190] The server receives the uploaded audio file and activates a speech recognition engine to convert the audio data into text data. For example, if a user says, "The next meeting is on Tuesday," the audio is recorded as text: "The next meeting is on Tuesday."

[0191] Emotion recognition

[0192] The server inputs the voice data into an emotion engine to analyze the user's emotions. The emotion engine recognizes emotions from the voice and outputs the results as numerical data. For example, from the statement "The next meeting is on Tuesday," it can identify emotions such as "happiness" or "sadness."

[0193] Summarize text and add sentiment

[0194] The server inputs the converted text data into a generative model (e.g., GPT-3) to summarize the text data. Furthermore, the emotion engine adds the emotion data recognized by the text data to the summary as supplementary information. For example, the text "The next meeting is on Tuesday" becomes "Next meeting: Tuesday (emotion: happy)."

[0195] Save summary and link to call history

[0196] The server stores the summarized text data in a database and associates it with the user's call history, so that when the user checks their call history, the summary text and emotion data are instantly displayed.

[0197] Displaying notes in call history

[0198] When a user checks their call history, the device retrieves summary data and emotion data associated with the call history and displays them on the screen. For example, it displays "Next meeting: Tuesday (Emotion: Happiness)." This allows the user to easily understand the content of the call and the emotion expressed at the time.

[0199] Specific examples

[0200] Here's a concrete example: A user has a phone conversation with a colleague, with the following content and emotions:

[0201] "Schedule a meeting with the client next Tuesday in a conference room at our office." (Emotion: Happiness)

[0202] The call is recorded and the audio is uploaded to a server.

[0203] The server converts the speech to text, resulting in the text "Please schedule a meeting with the client for next Tuesday in a conference room in our office."

[0204] The emotion engine recognizes "happiness" from this statement

[0205] The generative model summarizes the text as "Customer meeting: Tuesday, office conference room."

[0206] The server adds the emotion data "happiness" to the summary text, resulting in "Customer meeting: Tuesday, office conference room (emotion: happiness)."

[0207] This summary and emotion data are stored in a database and linked to call history.

[0208] When a user checks their call history, they see a summary of the call, "Customer Meeting: Tuesday, Office Conference Room (Emotion: Happiness)," along with emotion data.

[0209] This system allows users to manage not only the content of a call, but also their emotions during the call. By checking emotional data, users can more accurately grasp the nuances and importance of the conversation, contributing to improved work efficiency and communication quality.

[0210] The processing flow will be explained below.

[0211] Step 1:

[0212] The user starts a call. This causes the device to automatically prepare to start recording audio. Specifically, the device's recording function is turned on, and the device is ready to record the audio data of the call.

[0213] Step 2:

[0214] The user ends the call. When the call ends, the device stops recording and saves the recorded audio data as a file. This audio file is uploaded to the server as soon as the call ends.

[0215] Step 3:

[0216] The server receives the uploaded audio file. The server then activates a speech recognition engine and converts the audio data into text data. For example, the server passes the audio file to the speech recognition engine and obtains the text data, "The next meeting is on Tuesday."

[0217] Step 4:

[0218] The server inputs the voice data into an emotion engine to analyze the user's emotions. The emotion engine recognizes emotions from the voice and outputs the results as numerical data. For example, from the statement "The next meeting is on Tuesday," it can identify emotions such as "happiness" or "sadness."

[0219] Step 5:

[0220] The server inputs the converted text data into a generative model, which then summarizes the text data into a concise form. For example, the text data "The next meeting is on Tuesday" is summarized as "Next meeting: Tuesday."

[0221] Step 6:

[0222] The server adds the emotion data recognized by the emotion engine to the summarized text data. This adds the emotion data as supplementary information to the summary text. For example, "Next meeting: Tuesday (emotion: happy)".

[0223] Step 7:

[0224] The server stores the summary and emotion data in a database. At the same time, it associates the summary data with the user's call history. This allows the summary and emotion data to be referenced from the call history.

[0225] Step 8:

[0226] The user checks the call history. The device retrieves the summary data and emotion data associated with the call history and displays them on the screen. The user can easily check the related summary data and emotion data by simply tapping the call history.

[0227] Step 9:

[0228] The user checks the summary data and emotion data, and transcribes it into a memo or calendar as needed. This step allows the user to smoothly move on to the next action without forgetting important information about the call content.

[0229] Example 2

[0230] 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."

[0231] Conventional call management systems only record call content and convert it into text, but do not associate it with the user's call history or recognize emotions. This makes it difficult for users to easily understand detailed call records, especially important call content and emotional nuances. Furthermore, summarizing conversations and extracting key points is time-consuming and laborious. The objective of this invention is to solve these problems and provide a system that enables efficient call content management and immediate understanding of important information.

[0232] 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 a means for recording the contents of the call, a means for converting the recorded voice data into text data using a voice recognition engine, a means for analyzing the user's emotions from the text data using an emotion analysis engine, a means for summarizing the converted text data using a generative model, a means for adding emotion data to the summarized text data, and a means for saving and displaying the summarized text data and emotion data in association with the call history. This not only allows the user to easily understand the contents of the call, but also allows the user to simultaneously manage emotions during the call, making it possible to instantly check the nuances and important points of the conversation.

[0233] "Means for recording phone calls" refers to the ability to capture audio when a call begins and save that audio as a file when the call ends.

[0234] "Speech recognition engine" refers to software or hardware technology that has the function of analyzing voice data and converting it into corresponding text data.

[0235] "Means for analyzing user emotions from text data using an emotion analysis engine" refers to a function that receives text data as input, identifies the user's emotions from the content, and outputs them as a number or category.

[0236] "Means for summarizing converted text data using a generative model" refers to the function of using a generative AI model to summarize long text data and extract only the important points.

[0237] "Means for adding emotion data to summarized text data" refers to a function that adds emotion analysis results to summarized text and provides them as supplementary information.

[0238] "Means for storing and displaying summarized text data and emotional data in association with call history" refers to a function that stores summarized text data and emotional data in a database and links them to call history, thereby displaying this data simultaneously when the user checks the history.

[0239] According to an embodiment of the present invention, a system is provided that automates the entire process of recording, converting to text, analyzing emotions, and summarizing call content. This system records call content and associates it with a user's call history, enabling efficient management of call content and emotion information.

[0240] Recording and uploading audio

[0241] When a user starts a call, the device automatically starts recording audio. This is achieved using the device's recording functionality (e.g., the MediaRecorder class in Android or the AVAudioRecorder class in iOS). The recorded audio data is saved as a temporary file. When the call ends, the device uploads this audio data to the server using an HTTP POST request.

[0242] Speech-to-text conversion

[0243] The server receives the uploaded audio file. It invokes a speech recognition engine (e.g., Google Speech-to-Text API) to convert the audio data into text data. Specifically, the audio file is sent as an API request, and the server receives the text data as a response. For example, if a user says, "The next meeting is on Tuesday," this speech is converted into the text data, "The next meeting is on Tuesday."

[0244] Emotion recognition

[0245] The server inputs the voice data into a sentiment analysis engine (e.g., IBM Watson's Tone Analyzer). The sentiment analysis engine analyzes the user's emotions from the voice and text data and outputs the results as numerical or categorical data. For example, the server sends the text data "The next meeting is on Tuesday" to the sentiment analysis engine and receives the sentiment analysis result of "Happy."

[0246] Summarize text and add sentiment

[0247] The server then uses a generative AI model (e.g., OpenAI GPT-3) to summarize the converted text data. The server sends the generative AI model a prompt like this:

[0248] "Please summarize the following conversation: The next meeting is on Tuesday."

[0249] The generative AI model responds to this prompt sentence and generates a summary such as "Next meeting: Tuesday." Furthermore, the sentiment analysis engine adds emotional data recognized by the summary text to the summary text. For example, emotional data is added to the summary sentence, such as "Next meeting: Tuesday (Emotion: Happy)."

[0250] Save summary data and link to call history

[0251] The server stores the summarized text data and emotion data in a database and associates them with the user's call history. Specifically, the server links the summarized data to specific records in the call history database, enabling efficient management of call content and emotion information.

[0252] Displaying notes in call history

[0253] When a user checks their call history, the device retrieves the relevant summary data and emotion data from the server and displays it on the screen. For example, it might display "Next meeting: Tuesday (Emotion: Happiness)," allowing the user to understand the content of the call and the emotion expressed at the time at a glance. This allows users to quickly check detailed information about particularly important calls.

[0254] As described above, the present invention provides a system that effectively integrates all processes from recording call content to emotion recognition, summarization, storage, and display, which will greatly contribute to improving work efficiency and the quality of communication.

[0255] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0256] Step 1:

[0257] When a user starts a call, the device automatically starts recording audio. This is achieved by using the call start event as a trigger to launch the device's recording function (e.g., the MediaRecorder class in Android or the AVAudioRecorder class in iOS). The input is the call start event, and the output is the audio data being recorded. The recorded audio data is saved as a temporary file on the device.

[0258] Step 2:

[0259] When the call ends, the device uploads the audio data stored in the temporary file to the server. Specifically, it uses an HTTP POST request to send the audio file to a specific endpoint on the server. The input is the recorded audio file, and the output is the audio data uploaded to the server. The audio file is then sent to the server and ready for the next processing step.

[0260] Step 3:

[0261] The server receives the uploaded audio file. The server invokes a speech recognition engine (e.g., Google Speech-to-Text API) to convert the audio data into text data. The input is the audio file uploaded to the server, and the output is the converted text data. Specifically, the server sends the audio file as an API request and receives text data as a response. For example, the generated text is "The next meeting is on Tuesday."

[0262] Step 4:

[0263] The server inputs the voice data into a sentiment analysis engine (e.g., IBM Watson's Tone Analyzer). The input is converted text data, and the output is analyzed emotional data. The sentiment analysis engine analyzes the user's emotions from the voice and text data and outputs the results as numerical or categorical data. The server sends the text data to the API endpoint of the sentiment analysis engine and receives the sentiment analysis result, for example, "happy."

[0264] Step 5:

[0265] The server inputs the converted text data into a generative AI model (e.g., OpenAI GPT-3) and summarizes the text data. The input is the converted text data and a prompt, and the output is the summary text. The server sends the following prompt to the generative AI model:

[0266] "Please summarize the following conversation: The next meeting is on Tuesday."

[0267] The generative AI model responds to this prompt and generates a summary such as "Next meeting: Tuesday."

[0268] Step 6:

[0269] The server adds emotional data to the summarized text data. The input is the summary text and emotional data, and the output is the summary text with the added emotional data. For example, emotional data is added to the summary sentence, such as "Next meeting: Tuesday (emotion: happy)." As a result, the content of the call and the emotions felt at the time can be understood at a glance.

[0270] Step 7:

[0271] The server stores the summarized text data and emotion data in a database and associates it with the user's call history. The input is the summarized text and emotion data, and the output is data associated with the call history. Specifically, the summary data and emotion data are linked to each call history record. This process allows the necessary information to be managed quickly and efficiently.

[0272] Step 8:

[0273] When a user checks their call history, the device retrieves the relevant summary data and emotion data from the server and displays them on the screen. The input is a request for call history, and the output is a screen displaying the summary data and emotion data. For example, information is displayed in the format "Next meeting: Tuesday (emotion: happy)." As a result, the user can easily check the content of the call and the emotion at the time.

[0274] (Application example 2)

[0275] 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."

[0276] Autonomous vehicles require a means to efficiently record conversations and instructions between passengers and quickly obtain necessary information by summarizing it. However, conventional technology only records voice data and does not analyze or summarize emotional data, making it difficult to grasp the nuances and importance of the information. This makes it easy to overlook passenger intentions and important information, increasing the risk of problems occurring.

[0277] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0278] In this invention, the server includes means for recording the contents of the call, means for converting the recorded voice data into text data using a voice recognition engine, means for summarizing the text data using a generative model, means for saving and displaying the summarized text data together with emotion data added by an emotion recognition engine, and means for saving and displaying the data in association with the call history. This not only records the voice but also performs summarization and emotion analysis, making it possible to prevent important information from being overlooked and to quickly and accurately grasp the intentions and importance of passengers.

[0279] A "call recording device" is any device or software that is capable of adequately capturing and digitally recording the audio of a call.

[0280] "Means for converting recorded voice data into text using a voice recognition engine" means software or algorithms that use voice recognition technology to convert recorded voice files into text format.

[0281] "Means for summarizing text data using a generative model" refers to an algorithm that uses a generative AI model to extract key information from long text data and output it as a short summary.

[0282] The "means for saving and displaying summarized text data together with emotion data added by an emotion recognition engine" refers to a system for adding emotion information analyzed by an emotion recognition system to summarized text data, saving the data, and displaying it so that the user can easily access it.

[0283] "Means for storing and displaying data in association with call history" refers to technology for linking and associating data containing summaries and emotional information with call history, thereby picking up and displaying the content of past calls and emotions.

[0284] MODE FOR CARRYING OUT THE INVENTION

[0285] The system for carrying out the present invention mainly has the following configuration.

[0286] Recording and uploading audio

[0287] When a user starts a call, the device automatically starts recording audio. It uses the microphone to capture audio data and continues recording for the duration of the call. When the call ends, the recorded audio data is uploaded to the server.

[0288] Speech-to-text conversion

[0289] When the server receives the recorded voice data, it activates a speech recognition engine and converts the voice data into text data. In this case, the speech_recognition library is used. For example, if a user says, "Please make the next stop Shibuya Station," the speech is recorded as text saying, "Please make the next stop Shibuya Station."

[0290] Emotion recognition

[0291] The server inputs the converted text data into an emotion recognition engine to analyze the user's emotions. A custom module called EmotionRecognition is used for this emotion recognition. For example, the same utterance "Please make the next stop Shibuya Station" can identify the emotion "neutral."

[0292] Summarize text and add sentiment

[0293] The server then inputs the converted text data into a generative AI model to summarize it. This summarization is performed using a generative model such as T5 from the Transformers library. Emotion data recognized by an emotion recognition engine is then added to the summary. For example, the text "Please make the next stop Shibuya Station" is summarized as "Stops at Shibuya Station," and the emotion data "neutral" is added to make it "Stops at Shibuya Station (Emotion: Neutral)."

[0294] Save to database and link to call history

[0295] Once the summary and emotion data are created, the server stores them in a database and associates them with the call history, so that when a user checks their call history, the summary and emotion data are instantly displayed.

[0296] Displaying notes in call history

[0297] When a user checks their call history, the device retrieves summary data and emotion data associated with the call history and displays them on the screen. Specifically, it displays the data in the form of "Shibuya Station Stop (Emotion: Neutral)." This system allows users to easily understand the content of the call and the emotion expressed at the time.

[0298] Specific examples

[0299] Scenario: A passenger gives instructions on board, such as "Next stop: Shibuya Station."

[0300] Example prompt sentence:

[0301] "Next stop please be Shibuya Station."

[0302] Example output:

[0303] Shibuya Station (Emotion: Neutral)

[0304] This invention converts voice data into text and automatically displays summaries and memos with emotion recognition, enabling quick and accurate understanding of information. This is expected to reduce troubles in self-driving vehicles and improve the passenger experience.

[0305] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0306] Step 1:

[0307] When the device detects the start of a call, it automatically starts recording audio. Specifically, the device's microphone captures the audio data and continues recording until the call ends. The recorded audio data is saved in raw audio format. The input is the audio during the call, and the output is the recorded audio file.

[0308] Step 2:

[0309] When the call ends, the device uploads the recorded audio data to the server. At this time, the audio data is sent to the server via the network. The input is the recorded audio file, and the output is the audio data stored on the server.

[0310] Step 3:

[0311] The server inputs the received voice data into a voice recognition engine and converts it into text data. This process uses the speech_recognition library to convert the voice file into text format. The input is voice data and the output is text data.

[0312] Step 4:

[0313] The server inputs the converted text data into an emotion recognition engine to analyze the user's emotions. The emotion recognition engine uses the EmotionRecognition module to extract emotional information from the text data. The input is text data, and the output is emotional data.

[0314] Step 5:

[0315] The server inputs the text data into a generative AI model to generate a summary. In this case, it uses the T5 model from the Transformers library to extract key information from long text data and generate a short summary. The input is text data, and the output is a summary.

[0316] Step 6:

[0317] The server adds emotional data to the generated summary to create the final summary. The server then integrates and formats the generated summary and emotional data. The input is the summary and emotional data, and the output is summary data that includes emotional information.

[0318] Step 7:

[0319] The server stores this summary data in a database and associates it with the call history. A new entry is added to the database and linked to the call history. The input is the summary data including emotion information, and the output is a database entry associated with the call history.

[0320] Step 8:

[0321] When a user checks the call history, the device retrieves the associated summary data from the database and displays it on the screen. The device sends a query to the database and displays the retrieved summary data and emotion information on the screen. The input is the operation to reference the call history, and the output is the summary data displayed on the screen.

[0322] 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.

[0323] 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.

[0324] 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.

[0325] [Second embodiment]

[0326] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0327] 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.

[0328] 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).

[0329] 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.

[0330] 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.

[0331] 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).

[0332] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0333] 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.

[0334] 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.

[0335] 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.

[0336] 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.

[0337] 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."

[0338] The present invention provides a system that automatically memos the contents of telephone conversations and allows users to check the memos with a single touch from the call history. This system is mainly composed of three entities: a terminal, a server, and a user.

[0339] Recording and uploading audio

[0340] The device automatically starts recording audio when the user starts a call. The recorded audio is saved when the call ends and uploaded to a server, allowing the call contents to be stored in digital form.

[0341] Speech-to-text conversion

[0342] When the server receives the uploaded audio file, a speech recognition engine is activated and converts the audio data into text data. For example, if a user says, "The next meeting is on Tuesday," this is recorded as text.

[0343] Text summary

[0344] The server inputs the converted text data into a generative model (e.g., GPT-3) to generate a summary. For example, the text "The next meeting is on Tuesday" is summarized as "Next meeting: Tuesday." This summary aims to simplify the information.

[0345] Save summary and link to call history

[0346] The server stores the summarized text data in a database and links it to the user's call history, so that when the user checks their call history, the summary text is immediately displayed.

[0347] Displaying notes in call history

[0348] When users check their call history, the device will display the associated summary, allowing them to easily check the content of the call. For example, simply tapping on a call history will display the summary "Next meeting: Tuesday."

[0349] Specific examples

[0350] Here's a concrete example: A user talks to a sales representative on the phone and says the following:

[0351] "Please schedule a meeting with a client for next Tuesday in the office conference room."

[0352] Calls are recorded and uploaded to a server

[0353] The server converts the speech to text, generating the following text: "Please schedule a meeting with the client for next Tuesday in a conference room in our office."

[0354] The generative model summarizes this as "Customer meeting: Tuesday, office conference room."

[0355] This summary is stored in a database and linked to the call history.

[0356] When users check their call history, a summary is displayed, allowing them to see the details of the call at a glance.

[0357] In this way, the present invention allows users to record and quickly check the contents of calls, thereby preventing users from overlooking important information and improving work efficiency.

[0358] The processing flow will be explained below.

[0359] Step 1:

[0360] The user starts a call. This causes the device to automatically prepare to start recording audio. Specifically, the device's recording function is turned on, and the device is ready to record the audio data of the call.

[0361] Step 2:

[0362] The user ends the call. When the call ends, the device stops recording and saves the recorded audio data as a file. This audio file is uploaded to the server as soon as the call ends.

[0363] Step 3:

[0364] The server receives the uploaded audio file. The server then activates a speech recognition engine and converts the audio data into text data. For example, the server passes the audio file to the speech recognition engine and obtains the text data, "The next meeting is on Tuesday."

[0365] Step 4:

[0366] The server inputs the converted text data into a generative model, which then summarizes the text data into a concise form. For example, the text data "The next meeting is on Tuesday" is summarized as "Next meeting: Tuesday."

[0367] Step 5:

[0368] The server saves the summarized text data in a database. At the same time, it associates this summary data with the user's call history, allowing the summary data to be referenced from the call history.

[0369] Step 6:

[0370] The user checks the call history. The device retrieves the summary data associated with the call history and displays it on the screen. The user can easily check the related summary data by simply tapping the call history.

[0371] Step 7:

[0372] The user checks the summary data and transcribes it into a memo or calendar as necessary. This step allows the user to smoothly move on to the next action without forgetting any important information about the call content.

[0373] Example 1

[0374] 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."

[0375] Conventional call recording systems require the manual recording of call content, which can lead to important information being overlooked. Furthermore, reviewing call content takes time, and there are few ways to improve work efficiency. Therefore, there is a need for a system that automatically records call content and can efficiently summarize and search it.

[0376] 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.

[0377] In this invention, the server includes a means for recording the contents of the call, a means for converting the recorded voice data into text data using a voice recognition system, a means for summarizing the text data using a generative AI model, a means for storing and displaying the summarized text data in association with the communication history, and a means for displaying the summary when the user checks the communication history. This enables the contents of the call to be automatically recorded and efficiently summarized and searched.

[0378] "Call content" refers to information and conversations exchanged over the telephone or other voice communications.

[0379] "Recording" refers to the process of storing audio data in digital or analog form.

[0380] "Audio Data" refers to audio information recorded in digital or analog format.

[0381] A "speech recognition system" refers to the technology that allows a computer to understand human speech and convert it into text.

[0382] "Text data" refers to digital data expressed as character information.

[0383] A "generative AI model" refers to an artificial intelligence model that learns using large amounts of data and makes predictions and generates results.

[0384] A "summary" is a short sentence that concisely summarizes the original information.

[0385] "Communication history" refers to a record of calls and messages made, including information such as the caller, recipient, and time.

[0386] "Storage" refers to keeping data or information in a fixed location.

[0387] "Display" refers to outputting information such as text or graphics to a screen or display.

[0388] "User" refers to a person who uses a system or application.

[0389] The present invention is a system that automatically records and summarizes the contents of phone calls and associates them with call history. This system is mainly composed of three entities: a terminal, a server, and a user.

[0390] Recording and uploading audio

[0391] When a user starts a call, the device automatically starts recording audio. This function uses the device's built-in microphone and a voice recording app (for example, the Voice Memos app on iOS). When the call ends, the recorded audio file is saved in a specific folder on the device. The recorded file is then uploaded to the server using a POST request using the HTTP protocol.

[0392] Speech-to-text conversion

[0393] When the server receives the uploaded audio file, a speech recognition system (e.g., Google Cloud Speech-to-Text API) is activated to convert the audio data into text data. At this time, the audio file is saved in a storage server (e.g., Amazon S3) and then analyzed by the speech recognition system. For example, if a user says, "The next meeting is on Tuesday," this content is recorded as text data.

[0394] Text summary

[0395] The server inputs the generated text data into a generative AI model (e.g., OpenAI's GPT-3) to generate a summary. As a specific example, the server receives the text "The next meeting is on Tuesday" and sends the prompt "Please summarize this text: 'The next meeting is on Tuesday.'" to the generative model. As a result, the generative model generates the summary "Next meeting: Tuesday."

[0396] Save summary and link to call history

[0397] The server stores the summary in a database (e.g., MySQL) and links it to the user's call history. This linking allows the user to instantly retrieve the summary when checking the call history. For example, by associating the call summary data with the call date and time and the other party's phone number, the call history can be managed effectively.

[0398] Displaying notes in call history

[0399] When a user checks the call history on their device, the device requests the summary stored on the server. The server then sends back the corresponding summary, which is then displayed in the device's call history app. This process allows users to instantly view summaries such as "Next meeting: Tuesday" by simply tapping on the call history.

[0400] Example operation

[0401] As a real-world use case, imagine a user is talking to a sales rep on the phone and says, "Please schedule a meeting with a customer next Tuesday in a conference room in my office." The system would automatically execute the following process:

[0402] 1. The call will be recorded and an audio file will be created using your device's built-in microphone and audio recording app.

[0403] 2. The audio file is uploaded to the server.

[0404] 3. The server converts the received voice file into text data using a voice recognition system, and the generated text is, "Please schedule a meeting with the client for next Tuesday. The location should be in the office conference room."

[0405] 4. This text data is fed into a generative AI model, which generates a summary: "Customer meeting: Tuesday, office conference room."

[0406] 5. The summary is stored in a database and linked to the call history.

[0407] 6. When the user checks their call history, a summary of "Customer Meeting: Tuesday, Office Conference Room" is instantly displayed.

[0408] In this way, the present invention allows users to record and quickly check the contents of calls, and improves work efficiency without overlooking important information.

[0409] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0410] Step 1:

[0411] The device automatically starts recording audio when the user starts a call. Specifically, the device's call management app detects the call start event and activates the audio recording module. Once recording begins, audio data is collected in real time and temporarily stored in the device's internal memory.

[0412] Input: User call start event

[0413] Output: Audio data being recorded

[0414] Step 2:

[0415] The device stops recording when the call ends and saves the collected audio data in a specific folder. The call management app detects the call end event and stops the recording module, completing the recording. The recording file is then uploaded to the server using the HTTP protocol.

[0416] Input: Call end event and recording audio data

[0417] Output: Audio file uploaded to the server

[0418] Step 3:

[0419] The server receives the HTTP POST request and saves the uploaded audio file on the storage server. Next, it calls the speech recognition system to convert the audio data into text data. This process takes the audio file and outputs it as text using the speech recognition API. For example, if the audio data says "The next meeting is on Tuesday," it will be converted into text data.

[0420] Input: Uploaded audio file

[0421] Output: Converted text data

[0422] Step 4:

[0423] The server inputs the generated text data into the generative AI model to generate a summary. Specifically, it forms a prompt sentence and sends an API request to the generative model. The text "The next meeting is on Tuesday" is input into the generative AI model, and the summary "Next meeting: Tuesday" is returned.

[0424] Input: Generated text data

[0425] Output: Summarized text data

[0426] Step 5:

[0427] The server stores the summary in a database and links it to the call history, which contains metadata such as the call date and time and the caller's phone number, and associates the summary with that information.

[0428] Input: Summarized text data

[0429] Output: A link between the summary stored in the database and the call history

[0430] Step 6:

[0431] When a user checks the call history on their device, the device sends a request to the server to retrieve the stored summary. When the user opens the call history app and taps on a specific call history, the summary linked to that call history is displayed on the device.

[0432] Input: User operation to check call history

[0433] Output: A summary of the call displayed on your device

[0434] (Application example 1)

[0435] 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."

[0436] While current call recording systems can accurately record call content, it is difficult to quickly and efficiently understand the content. Furthermore, they lack the functionality to store call records on the cloud and play them back as needed, creating a need for improved security and operational efficiency.

[0437] 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.

[0438] In this invention, the server includes means for recording the contents of the call, means for converting the recorded voice data into text data using a voice recognition engine, means for summarizing the text data using a generative model, means for saving and displaying the summarized text data in association with the call history, means for uploading and saving the voice data and the summarized text to cloud storage, and means for restoring and playing back the uploaded voice data. This enables quick and efficient understanding of the contents of the call, and improves security and business efficiency through data management on the cloud.

[0439] "Call content" refers to all words and information spoken during a call.

[0440] "Recording means" refers to any device or method that physically or digitally records the audio of a call.

[0441] "Audio Data" refers to a digital audio file that records the contents of a call.

[0442] A "voice recognition engine" refers to software or algorithms that have the function of analyzing voice data and converting it into text data.

[0443] "Text data" refers to the textual information of the speech content generated by a speech recognition engine.

[0444] A "generative model" refers to a machine learning model used to generate a summary or output in a specific format from input data.

[0445] "Means for summarizing" refers to a method or device that analyzes text data, extracts only the important points, and shortens them.

[0446] "Call history" refers to databases and log information that record the date and time of past calls, the callers, etc.

[0447] The "means for storing and displaying in association" refers to a method or device for linking the summarized text data to the call history, and storing and displaying the data so that the user can easily view it.

[0448] "Cloud storage" refers to an online storage service that allows you to store data remotely via the Internet.

[0449] "Uploading and storage means" refers to the methods and devices for transferring and securely storing audio data and summary text in cloud storage.

[0450] "Means for restoring and playing" refers to a method or device for downloading audio data stored in cloud storage and playing it back as the original audio.

[0451] An embodiment of the present invention will now be described in detail. This system is mainly composed of three elements: a terminal, a server, and a user.

[0452] Recording and uploading audio

[0453] The device automatically starts recording audio when the user starts a call. The recorded audio data is temporarily stored on the device when the call ends and then uploaded to cloud storage. For example, if you are using smart glasses, the microphone will capture the audio and the recording file will be saved in cloud storage.

[0454] Speech-to-text conversion

[0455] When the server receives the uploaded audio file, a speech recognition engine is activated and converts the audio data into text data. The speech recognition engine used in this process uses, for example, the Python speech_recognition library. For example, if a user says, "The next meeting is on Tuesday," this is accurately recorded as text.

[0456] Text summary

[0457] The server generates a summary using a generative model (e.g., an AI model such as BERT or GPT) to further simplify the converted text data. For example, the text "The next meeting is on Tuesday" is summarized as "Meeting: Tuesday." This summary is generated by inputting a prompt sentence to the generative AI model.

[0458] Save summary and link to call history

[0459] The server stores the summarized text data in a cloud database and links it to the user's call history. This allows the summary to be displayed instantly when the user checks the call history. For example, by simply tapping on a call history, the summary "Meeting: Tuesday" can be displayed.

[0460] Displaying notes in call history

[0461] When a user checks the call history, the device will display the associated summary, allowing the user to easily check the content of the call. For example, the summary "Meeting: Tuesday" will be displayed on the smart glasses display.

[0462] Data management in cloud storage

[0463] The recorded audio data and generated summary text are uploaded to cloud storage and stored securely, allowing for efficient management of large amounts of data. The stored audio data can also be restored and played back as needed.

[0464] Specific prompt examples

[0465] Conversation: "I have a conference call with a client tomorrow at 2 PM. Please send the materials by email."

[0466] Summary: "Meeting: Tomorrow at 2pm, materials to be sent."

[0467] An example of a prompt to input to a generative AI model: "Please summarize the following text: I have a conference call with a client tomorrow at 2 PM. Please also send the materials by email."

[0468] The above is a form for implementing the present invention. By using this system, users can quickly and efficiently understand the content of calls, and furthermore, data management on the cloud can improve security and business efficiency.

[0469] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0470] Step 1:

[0471] When a call is initiated, the device automatically records the call. The input for the recording is the audio from the call, and the output is audio data (e.g., a .wav file). This audio data is stored in the device's temporary memory.

[0472] Step 2:

[0473] When the call ends, the device automatically uploads the recorded audio data to cloud storage. The input is the audio data generated in step 1, and the output is the audio file stored in cloud storage.

[0474] Step 3:

[0475] The server receives the audio data uploaded from the cloud storage. The input is the audio data stored in the cloud storage, and the output is the audio data stored in the server's storage device.

[0476] Step 4:

[0477] The server uses a speech recognition engine to convert the received voice data into text data. The input is voice data (e.g., a .wav file), and the output is the text of the conversation. Specifically, the server uses the Python speech_recognition library to analyze the voice data and convert it into text information.

[0478] Step 5:

[0479] The server uses a generative AI model to summarize the converted text data. The input is the text data generated in step 4, and the output is summarized text information. A prompt sentence is used to summarize the input text for the generative AI model (e.g., GPT-3). An example of a specific prompt sentence is, "Please summarize the following text: I have a conference call with a client tomorrow at 2 p.m. Please also send the materials by email."

[0480] Step 6:

[0481] The server associates the summarized text data with the call history and stores it in a database. The input is the summarized text data and the call history information, and the output is the summary text linked to the call history. This allows the relevant summary information to be displayed immediately when the user checks the call history.

[0482] Step 7:

[0483] When a user checks the call history, the terminal displays the associated summary text. The input is the summary text obtained from the call history, and the output is the summary information displayed on the user interface, allowing the user to quickly check the key points of the call.

[0484] Step 8:

[0485] The server or device provides the functionality to restore and play back the voice data stored in the cloud storage as needed. The input is the voice file stored in the cloud storage, and the output is the voice output from the playback device (e.g., speaker or headset). This allows the user to reconfirm the contents of the call.

[0486] 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.

[0487] An embodiment of the present invention will be described in detail. The present invention is a system that combines a system that automatically takes notes on phone conversations and allows users to check the notes with a single touch from their call history, with an emotion engine that recognizes the user's emotions. This system is composed of three entities: a terminal, a server, and a user.

[0488] Recording and uploading audio

[0489] When a user starts a call, the device automatically starts recording the audio. The audio data of the call is recorded and uploaded to the server when the call ends.

[0490] Speech-to-text conversion

[0491] The server receives the uploaded audio file and activates a speech recognition engine to convert the audio data into text data. For example, if a user says, "The next meeting is on Tuesday," the audio is recorded as text: "The next meeting is on Tuesday."

[0492] Emotion recognition

[0493] The server inputs the voice data into an emotion engine to analyze the user's emotions. The emotion engine recognizes emotions from the voice and outputs the results as numerical data. For example, from the statement "The next meeting is on Tuesday," it can identify emotions such as "happiness" or "sadness."

[0494] Summarize text and add sentiment

[0495] The server inputs the converted text data into a generative model (e.g., GPT-3) to summarize the text data. Furthermore, the emotion engine adds the emotion data recognized by the text data to the summary as supplementary information. For example, the text "The next meeting is on Tuesday" becomes "Next meeting: Tuesday (emotion: happy)."

[0496] Save summary and link to call history

[0497] The server stores the summarized text data in a database and associates it with the user's call history, so that when the user checks their call history, the summary text and emotion data are instantly displayed.

[0498] Displaying notes in call history

[0499] When a user checks their call history, the device retrieves summary data and emotion data associated with the call history and displays them on the screen. For example, it displays "Next meeting: Tuesday (Emotion: Happiness)." This allows the user to easily understand the content of the call and the emotion expressed at the time.

[0500] Specific examples

[0501] Here's a concrete example: A user has a phone conversation with a colleague, with the following content and emotions:

[0502] "Schedule a meeting with the client next Tuesday in a conference room at our office." (Emotion: Happiness)

[0503] The call is recorded and the audio is uploaded to a server.

[0504] The server converts the speech to text, resulting in the text "Please schedule a meeting with the client for next Tuesday in a conference room in our office."

[0505] The emotion engine recognizes "happiness" from this statement

[0506] The generative model summarizes the text as "Customer meeting: Tuesday, office conference room."

[0507] The server adds the emotion data "happiness" to the summary text, resulting in "Customer meeting: Tuesday, office conference room (emotion: happiness)."

[0508] This summary and emotion data are stored in a database and linked to call history.

[0509] When a user checks their call history, they see a summary of the call, "Customer Meeting: Tuesday, Office Conference Room (Emotion: Happiness)," along with emotion data.

[0510] This system allows users to manage not only the content of a call, but also their emotions during the call. By checking emotional data, users can more accurately grasp the nuances and importance of the conversation, contributing to improved work efficiency and communication quality.

[0511] The processing flow will be explained below.

[0512] Step 1:

[0513] The user starts a call. This causes the device to automatically prepare to start recording audio. Specifically, the device's recording function is turned on, and the device is ready to record the audio data of the call.

[0514] Step 2:

[0515] The user ends the call. When the call ends, the device stops recording and saves the recorded audio data as a file. This audio file is uploaded to the server as soon as the call ends.

[0516] Step 3:

[0517] The server receives the uploaded audio file. The server then activates a speech recognition engine and converts the audio data into text data. For example, the server passes the audio file to the speech recognition engine and obtains the text data, "The next meeting is on Tuesday."

[0518] Step 4:

[0519] The server inputs the voice data into an emotion engine to analyze the user's emotions. The emotion engine recognizes emotions from the voice and outputs the results as numerical data. For example, from the statement "The next meeting is on Tuesday," it can identify emotions such as "happiness" or "sadness."

[0520] Step 5:

[0521] The server inputs the converted text data into a generative model, which then summarizes the text data into a concise form. For example, the text data "The next meeting is on Tuesday" is summarized as "Next meeting: Tuesday."

[0522] Step 6:

[0523] The server adds the emotion data recognized by the emotion engine to the summarized text data. This adds the emotion data as supplementary information to the summary text. For example, "Next meeting: Tuesday (emotion: happy)".

[0524] Step 7:

[0525] The server stores the summary and emotion data in a database. At the same time, it associates the summary data with the user's call history. This allows the summary and emotion data to be referenced from the call history.

[0526] Step 8:

[0527] The user checks the call history. The device retrieves the summary data and emotion data associated with the call history and displays them on the screen. The user can easily check the related summary data and emotion data by simply tapping the call history.

[0528] Step 9:

[0529] The user checks the summary data and emotion data, and transcribes it into a memo or calendar as needed. This step allows the user to smoothly move on to the next action without forgetting important information about the call content.

[0530] Example 2

[0531] 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."

[0532] Conventional call management systems only record call content and convert it into text, but do not associate it with the user's call history or recognize emotions. This makes it difficult for users to easily understand detailed call records, especially important call content and emotional nuances. Furthermore, summarizing conversations and extracting key points is time-consuming and laborious. The objective of this invention is to solve these problems and provide a system that enables efficient call content management and immediate understanding of important information.

[0533] 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 a means for recording the contents of the call, a means for converting the recorded voice data into text data using a voice recognition engine, a means for analyzing the user's emotions from the text data using an emotion analysis engine, a means for summarizing the converted text data using a generative model, a means for adding emotion data to the summarized text data, and a means for saving and displaying the summarized text data and emotion data in association with the call history. This not only allows the user to easily understand the contents of the call, but also allows the user to simultaneously manage emotions during the call, making it possible to instantly check the nuances and important points of the conversation.

[0534] "Means for recording phone calls" refers to the ability to capture audio when a call begins and save that audio as a file when the call ends.

[0535] "Speech recognition engine" refers to software or hardware technology that has the function of analyzing voice data and converting it into corresponding text data.

[0536] "Means for analyzing user emotions from text data using an emotion analysis engine" refers to a function that receives text data as input, identifies the user's emotions from the content, and outputs them as a number or category.

[0537] "Means for summarizing converted text data using a generative model" refers to the function of using a generative AI model to summarize long text data and extract only the important points.

[0538] "Means for adding emotion data to summarized text data" refers to a function that adds emotion analysis results to summarized text and provides them as supplementary information.

[0539] "Means for storing and displaying summarized text data and emotional data in association with call history" refers to a function that stores summarized text data and emotional data in a database and links them to call history, thereby displaying this data simultaneously when the user checks the history.

[0540] According to an embodiment of the present invention, a system is provided that automates the entire process of recording, converting to text, analyzing emotions, and summarizing call content. This system records call content and associates it with a user's call history, enabling efficient management of call content and emotion information.

[0541] Recording and uploading audio

[0542] When a user starts a call, the device automatically starts recording audio. This is achieved using the device's recording functionality (e.g., the MediaRecorder class in Android or the AVAudioRecorder class in iOS). The recorded audio data is saved as a temporary file. When the call ends, the device uploads this audio data to the server using an HTTP POST request.

[0543] Speech-to-text conversion

[0544] The server receives the uploaded audio file. It invokes a speech recognition engine (e.g., Google Speech-to-Text API) to convert the audio data into text data. Specifically, the audio file is sent as an API request, and the server receives the text data as a response. For example, if a user says, "The next meeting is on Tuesday," this speech is converted into the text data, "The next meeting is on Tuesday."

[0545] Emotion recognition

[0546] The server inputs the voice data into a sentiment analysis engine (e.g., IBM Watson's Tone Analyzer). The sentiment analysis engine analyzes the user's emotions from the voice and text data and outputs the results as numerical or categorical data. For example, the server sends the text data "The next meeting is on Tuesday" to the sentiment analysis engine and receives the sentiment analysis result of "Happy."

[0547] Summarize text and add sentiment

[0548] The server then uses a generative AI model (e.g., OpenAI GPT-3) to summarize the converted text data. The server sends the generative AI model a prompt like this:

[0549] "Please summarize the following conversation: The next meeting is on Tuesday."

[0550] The generative AI model responds to this prompt sentence and generates a summary such as "Next meeting: Tuesday." Furthermore, the sentiment analysis engine adds emotional data recognized by the summary text to the summary text. For example, emotional data is added to the summary sentence, such as "Next meeting: Tuesday (Emotion: Happy)."

[0551] Save summary data and link to call history

[0552] The server stores the summarized text data and emotion data in a database and associates them with the user's call history. Specifically, the server links the summarized data to specific records in the call history database, enabling efficient management of call content and emotion information.

[0553] Displaying notes in call history

[0554] When a user checks their call history, the device retrieves the relevant summary data and emotion data from the server and displays it on the screen. For example, it might display "Next meeting: Tuesday (Emotion: Happiness)," allowing the user to understand the content of the call and the emotion expressed at the time at a glance. This allows users to quickly check detailed information about particularly important calls.

[0555] As described above, the present invention provides a system that effectively integrates all processes from recording call content to emotion recognition, summarization, storage, and display, which will greatly contribute to improving work efficiency and the quality of communication.

[0556] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0557] Step 1:

[0558] When a user starts a call, the device automatically starts recording audio. This is achieved by using the call start event as a trigger to launch the device's recording function (e.g., the MediaRecorder class in Android or the AVAudioRecorder class in iOS). The input is the call start event, and the output is the audio data being recorded. The recorded audio data is saved as a temporary file on the device.

[0559] Step 2:

[0560] When the call ends, the device uploads the audio data stored in the temporary file to the server. Specifically, it uses an HTTP POST request to send the audio file to a specific endpoint on the server. The input is the recorded audio file, and the output is the audio data uploaded to the server. The audio file is then sent to the server and ready for the next processing step.

[0561] Step 3:

[0562] The server receives the uploaded audio file. The server invokes a speech recognition engine (e.g., Google Speech-to-Text API) to convert the audio data into text data. The input is the audio file uploaded to the server, and the output is the converted text data. Specifically, the server sends the audio file as an API request and receives text data as a response. For example, the generated text is "The next meeting is on Tuesday."

[0563] Step 4:

[0564] The server inputs the voice data into a sentiment analysis engine (e.g., IBM Watson's Tone Analyzer). The input is converted text data, and the output is analyzed emotional data. The sentiment analysis engine analyzes the user's emotions from the voice and text data and outputs the results as numerical or categorical data. The server sends the text data to the API endpoint of the sentiment analysis engine and receives the sentiment analysis result, for example, "happy."

[0565] Step 5:

[0566] The server inputs the converted text data into a generative AI model (e.g., OpenAI GPT-3) and summarizes the text data. The input is the converted text data and a prompt, and the output is the summary text. The server sends the following prompt to the generative AI model:

[0567] "Please summarize the following conversation: The next meeting is on Tuesday."

[0568] The generative AI model responds to this prompt and generates a summary such as "Next meeting: Tuesday."

[0569] Step 6:

[0570] The server adds emotional data to the summarized text data. The input is the summary text and emotional data, and the output is the summary text with the added emotional data. For example, emotional data is added to the summary sentence, such as "Next meeting: Tuesday (emotion: happy)." As a result, the content of the call and the emotions felt at the time can be understood at a glance.

[0571] Step 7:

[0572] The server stores the summarized text data and emotion data in a database and associates it with the user's call history. The input is the summarized text and emotion data, and the output is data associated with the call history. Specifically, the summary data and emotion data are linked to each call history record. This process allows the necessary information to be managed quickly and efficiently.

[0573] Step 8:

[0574] When a user checks their call history, the device retrieves the relevant summary data and emotion data from the server and displays them on the screen. The input is a request for call history, and the output is a screen displaying the summary data and emotion data. For example, information is displayed in the format "Next meeting: Tuesday (emotion: happy)." As a result, the user can easily check the content of the call and the emotion at the time.

[0575] (Application example 2)

[0576] 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."

[0577] Autonomous vehicles require a means to efficiently record conversations and instructions between passengers and quickly obtain necessary information by summarizing it. However, conventional technology only records voice data and does not analyze or summarize emotional data, making it difficult to grasp the nuances and importance of the information. This makes it easy to overlook passenger intentions and important information, increasing the risk of problems occurring.

[0578] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0579] In this invention, the server includes means for recording the contents of the call, means for converting the recorded voice data into text data using a voice recognition engine, means for summarizing the text data using a generative model, means for saving and displaying the summarized text data together with emotion data added by an emotion recognition engine, and means for saving and displaying the data in association with the call history. This not only records the voice but also performs summarization and emotion analysis, making it possible to prevent important information from being overlooked and to quickly and accurately grasp the intentions and importance of passengers.

[0580] A "call recording device" is any device or software that is capable of adequately capturing and digitally recording the audio of a call.

[0581] "Means for converting recorded voice data into text using a voice recognition engine" means software or algorithms that use voice recognition technology to convert recorded voice files into text format.

[0582] "Means for summarizing text data using a generative model" refers to an algorithm that uses a generative AI model to extract key information from long text data and output it as a short summary.

[0583] The "means for saving and displaying summarized text data together with emotion data added by an emotion recognition engine" refers to a system for adding emotion information analyzed by an emotion recognition system to summarized text data, saving the data, and displaying it so that the user can easily access it.

[0584] "Means for storing and displaying data in association with call history" refers to technology for linking and associating data containing summaries and emotional information with call history, thereby picking up and displaying the content of past calls and emotions.

[0585] MODE FOR CARRYING OUT THE INVENTION

[0586] The system for carrying out the present invention mainly has the following configuration.

[0587] Recording and uploading audio

[0588] When a user starts a call, the device automatically starts recording audio. It uses the microphone to capture audio data and continues recording for the duration of the call. When the call ends, the recorded audio data is uploaded to the server.

[0589] Speech-to-text conversion

[0590] When the server receives the recorded voice data, it activates a speech recognition engine and converts the voice data into text data. In this case, the speech_recognition library is used. For example, if a user says, "Please make the next stop Shibuya Station," the speech is recorded as text saying, "Please make the next stop Shibuya Station."

[0591] Emotion recognition

[0592] The server inputs the converted text data into an emotion recognition engine to analyze the user's emotions. A custom module called EmotionRecognition is used for this emotion recognition. For example, the same utterance "Please make the next stop Shibuya Station" can identify the emotion "neutral."

[0593] Summarize text and add sentiment

[0594] The server then inputs the converted text data into a generative AI model to summarize it. This summarization is performed using a generative model such as T5 from the Transformers library. Emotion data recognized by an emotion recognition engine is then added to the summary. For example, the text "Please make the next stop Shibuya Station" is summarized as "Stops at Shibuya Station," and the emotion data "neutral" is added to make it "Stops at Shibuya Station (Emotion: Neutral)."

[0595] Save to database and link to call history

[0596] Once the summary and emotion data are created, the server stores them in a database and associates them with the call history, so that when a user checks their call history, the summary and emotion data are instantly displayed.

[0597] Displaying notes in call history

[0598] When a user checks their call history, the device retrieves summary data and emotion data associated with the call history and displays them on the screen. Specifically, it displays the data in the form of "Shibuya Station Stop (Emotion: Neutral)." This system allows users to easily understand the content of the call and the emotion expressed at the time.

[0599] Specific examples

[0600] Scenario: A passenger gives instructions on board, such as "Next stop: Shibuya Station."

[0601] Example prompt sentence:

[0602] "Next stop please be Shibuya Station."

[0603] Example output:

[0604] Shibuya Station (Emotion: Neutral)

[0605] This invention converts voice data into text and automatically displays summaries and memos with emotion recognition, enabling quick and accurate understanding of information. This is expected to reduce troubles in self-driving vehicles and improve the passenger experience.

[0606] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0607] Step 1:

[0608] When the device detects the start of a call, it automatically starts recording audio. Specifically, the device's microphone captures the audio data and continues recording until the call ends. The recorded audio data is saved in raw audio format. The input is the audio during the call, and the output is the recorded audio file.

[0609] Step 2:

[0610] When the call ends, the device uploads the recorded audio data to the server. At this time, the audio data is sent to the server via the network. The input is the recorded audio file, and the output is the audio data stored on the server.

[0611] Step 3:

[0612] The server inputs the received voice data into a voice recognition engine and converts it into text data. This process uses the speech_recognition library to convert the voice file into text format. The input is voice data and the output is text data.

[0613] Step 4:

[0614] The server inputs the converted text data into an emotion recognition engine to analyze the user's emotions. The emotion recognition engine uses the EmotionRecognition module to extract emotional information from the text data. The input is text data, and the output is emotional data.

[0615] Step 5:

[0616] The server inputs the text data into a generative AI model to generate a summary. In this case, it uses the T5 model from the Transformers library to extract key information from long text data and generate a short summary. The input is text data, and the output is a summary.

[0617] Step 6:

[0618] The server adds emotional data to the generated summary to create the final summary. The server then integrates and formats the generated summary and emotional data. The input is the summary and emotional data, and the output is summary data that includes emotional information.

[0619] Step 7:

[0620] The server stores this summary data in a database and associates it with the call history. A new entry is added to the database and linked to the call history. The input is the summary data including emotion information, and the output is a database entry associated with the call history.

[0621] Step 8:

[0622] When a user checks the call history, the device retrieves the associated summary data from the database and displays it on the screen. The device sends a query to the database and displays the retrieved summary data and emotion information on the screen. The input is the operation to reference the call history, and the output is the summary data displayed on the screen.

[0623] 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.

[0624] 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.

[0625] 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.

[0626] [Third embodiment]

[0627] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0628] 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.

[0629] 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).

[0630] 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.

[0631] 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.

[0632] 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).

[0633] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0634] 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.

[0635] 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.

[0636] 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.

[0637] 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.

[0638] 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."

[0639] The present invention provides a system that automatically memos the contents of telephone conversations and allows users to check the memos with a single touch from the call history. This system is mainly composed of three entities: a terminal, a server, and a user.

[0640] Recording and uploading audio

[0641] The device automatically starts recording audio when the user starts a call. The recorded audio is saved when the call ends and uploaded to a server, allowing the call contents to be stored in digital form.

[0642] Speech-to-text conversion

[0643] When the server receives the uploaded audio file, a speech recognition engine is activated and converts the audio data into text data. For example, if a user says, "The next meeting is on Tuesday," this is recorded as text.

[0644] Text summary

[0645] The server inputs the converted text data into a generative model (e.g., GPT-3) to generate a summary. For example, the text "The next meeting is on Tuesday" is summarized as "Next meeting: Tuesday." This summary aims to simplify the information.

[0646] Save summary and link to call history

[0647] The server stores the summarized text data in a database and links it to the user's call history, so that when the user checks their call history, the summary text is immediately displayed.

[0648] Displaying notes in call history

[0649] When users check their call history, the device will display the associated summary, allowing them to easily check the content of the call. For example, simply tapping on a call history will display the summary "Next meeting: Tuesday."

[0650] Specific examples

[0651] Here's a concrete example: A user talks to a sales representative on the phone and says the following:

[0652] "Please schedule a meeting with a client for next Tuesday in the office conference room."

[0653] Calls are recorded and uploaded to a server

[0654] The server converts the speech to text, generating the following text: "Please schedule a meeting with the client for next Tuesday in a conference room in our office."

[0655] The generative model summarizes this as "Customer meeting: Tuesday, office conference room."

[0656] This summary is stored in a database and linked to the call history.

[0657] When users check their call history, a summary is displayed, allowing them to see the details of the call at a glance.

[0658] In this way, the present invention allows users to record and quickly check the contents of calls, thereby preventing users from overlooking important information and improving work efficiency.

[0659] The processing flow will be explained below.

[0660] Step 1:

[0661] The user starts a call. This causes the device to automatically prepare to start recording audio. Specifically, the device's recording function is turned on, and the device is ready to record the audio data of the call.

[0662] Step 2:

[0663] The user ends the call. When the call ends, the device stops recording and saves the recorded audio data as a file. This audio file is uploaded to the server as soon as the call ends.

[0664] Step 3:

[0665] The server receives the uploaded audio file. The server then activates a speech recognition engine and converts the audio data into text data. For example, the server passes the audio file to the speech recognition engine and obtains the text data, "The next meeting is on Tuesday."

[0666] Step 4:

[0667] The server inputs the converted text data into a generative model, which then summarizes the text data into a concise form. For example, the text data "The next meeting is on Tuesday" is summarized as "Next meeting: Tuesday."

[0668] Step 5:

[0669] The server saves the summarized text data in a database. At the same time, it associates this summary data with the user's call history, allowing the summary data to be referenced from the call history.

[0670] Step 6:

[0671] The user checks the call history. The device retrieves the summary data associated with the call history and displays it on the screen. The user can easily check the related summary data by simply tapping the call history.

[0672] Step 7:

[0673] The user checks the summary data and transcribes it into a memo or calendar as necessary. This step allows the user to smoothly move on to the next action without forgetting any important information about the call content.

[0674] Example 1

[0675] 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."

[0676] Conventional call recording systems require the manual recording of call content, which can lead to important information being overlooked. Furthermore, reviewing call content takes time, and there are few ways to improve work efficiency. Therefore, there is a need for a system that automatically records call content and can efficiently summarize and search it.

[0677] 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.

[0678] In this invention, the server includes a means for recording the contents of the call, a means for converting the recorded voice data into text data using a voice recognition system, a means for summarizing the text data using a generative AI model, a means for storing and displaying the summarized text data in association with the communication history, and a means for displaying the summary when the user checks the communication history. This enables the contents of the call to be automatically recorded and efficiently summarized and searched.

[0679] "Call content" refers to information and conversations exchanged over the telephone or other voice communications.

[0680] "Recording" refers to the process of storing audio data in digital or analog form.

[0681] "Audio Data" refers to audio information recorded in digital or analog format.

[0682] A "speech recognition system" refers to the technology that allows a computer to understand human speech and convert it into text.

[0683] "Text data" refers to digital data expressed as character information.

[0684] A "generative AI model" refers to an artificial intelligence model that learns using large amounts of data and makes predictions and generates results.

[0685] A "summary" is a short sentence that concisely summarizes the original information.

[0686] "Communication history" refers to a record of calls and messages made, including information such as the caller, recipient, and time.

[0687] "Storage" refers to keeping data or information in a fixed location.

[0688] "Display" refers to outputting information such as text or graphics to a screen or display.

[0689] "User" refers to a person who uses a system or application.

[0690] The present invention is a system that automatically records and summarizes the contents of phone calls and associates them with call history. This system is mainly composed of three entities: a terminal, a server, and a user.

[0691] Recording and uploading audio

[0692] When a user starts a call, the device automatically starts recording audio. This function uses the device's built-in microphone and a voice recording app (for example, the Voice Memos app on iOS). When the call ends, the recorded audio file is saved in a specific folder on the device. The recorded file is then uploaded to the server using a POST request using the HTTP protocol.

[0693] Speech-to-text conversion

[0694] When the server receives the uploaded audio file, a speech recognition system (e.g., Google Cloud Speech-to-Text API) is activated to convert the audio data into text data. At this time, the audio file is saved in a storage server (e.g., Amazon S3) and then analyzed by the speech recognition system. For example, if a user says, "The next meeting is on Tuesday," this content is recorded as text data.

[0695] Text summary

[0696] The server inputs the generated text data into a generative AI model (e.g., OpenAI's GPT-3) to generate a summary. As a specific example, the server receives the text "The next meeting is on Tuesday" and sends the prompt "Please summarize this text: 'The next meeting is on Tuesday.'" to the generative model. As a result, the generative model generates the summary "Next meeting: Tuesday."

[0697] Save summary and link to call history

[0698] The server stores the summary in a database (e.g., MySQL) and links it to the user's call history. This linking allows the user to instantly retrieve the summary when checking the call history. For example, by associating the call summary data with the call date and time and the other party's phone number, the call history can be managed effectively.

[0699] Displaying notes in call history

[0700] When a user checks the call history on their device, the device requests the summary stored on the server. The server then sends back the corresponding summary, which is then displayed in the device's call history app. This process allows users to instantly view summaries such as "Next meeting: Tuesday" by simply tapping on the call history.

[0701] Example operation

[0702] As a real-world use case, imagine a user is talking to a sales rep on the phone and says, "Please schedule a meeting with a customer next Tuesday in a conference room in my office." The system would automatically execute the following process:

[0703] 1. The call will be recorded and an audio file will be created using your device's built-in microphone and audio recording app.

[0704] 2. The audio file is uploaded to the server.

[0705] 3. The server converts the received voice file into text data using a voice recognition system, and the generated text is, "Please schedule a meeting with the client for next Tuesday. The location should be in the office conference room."

[0706] 4. This text data is fed into a generative AI model, which generates a summary: "Customer meeting: Tuesday, office conference room."

[0707] 5. The summary is stored in a database and linked to the call history.

[0708] 6. When the user checks their call history, a summary of "Customer Meeting: Tuesday, Office Conference Room" is instantly displayed.

[0709] In this way, the present invention allows users to record and quickly check the contents of calls, and improves work efficiency without overlooking important information.

[0710] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0711] Step 1:

[0712] The device automatically starts recording audio when the user starts a call. Specifically, the device's call management app detects the call start event and activates the audio recording module. Once recording begins, audio data is collected in real time and temporarily stored in the device's internal memory.

[0713] Input: User call start event

[0714] Output: Audio data being recorded

[0715] Step 2:

[0716] The device stops recording when the call ends and saves the collected audio data in a specific folder. The call management app detects the call end event and stops the recording module, completing the recording. The recording file is then uploaded to the server using the HTTP protocol.

[0717] Input: Call end event and recording audio data

[0718] Output: Audio file uploaded to the server

[0719] Step 3:

[0720] The server receives the HTTP POST request and saves the uploaded audio file on the storage server. Next, it calls the speech recognition system to convert the audio data into text data. This process takes the audio file and outputs it as text using the speech recognition API. For example, if the audio data says "The next meeting is on Tuesday," it will be converted into text data.

[0721] Input: Uploaded audio file

[0722] Output: Converted text data

[0723] Step 4:

[0724] The server inputs the generated text data into the generative AI model to generate a summary. Specifically, it forms a prompt sentence and sends an API request to the generative model. The text "The next meeting is on Tuesday" is input into the generative AI model, and the summary "Next meeting: Tuesday" is returned.

[0725] Input: Generated text data

[0726] Output: Summarized text data

[0727] Step 5:

[0728] The server stores the summary in a database and links it to the call history, which contains metadata such as the call date and time and the caller's phone number, and associates the summary with that information.

[0729] Input: Summarized text data

[0730] Output: A link between the summary stored in the database and the call history

[0731] Step 6:

[0732] When a user checks the call history on their device, the device sends a request to the server to retrieve the stored summary. When the user opens the call history app and taps on a specific call history, the summary linked to that call history is displayed on the device.

[0733] Input: User operation to check call history

[0734] Output: A summary of the call displayed on your device

[0735] (Application example 1)

[0736] 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."

[0737] While current call recording systems can accurately record call content, it is difficult to quickly and efficiently understand the content. Furthermore, they lack the functionality to store call records on the cloud and play them back as needed, creating a need for improved security and operational efficiency.

[0738] 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.

[0739] In this invention, the server includes means for recording the contents of the call, means for converting the recorded voice data into text data using a voice recognition engine, means for summarizing the text data using a generative model, means for saving and displaying the summarized text data in association with the call history, means for uploading and saving the voice data and the summarized text to cloud storage, and means for restoring and playing back the uploaded voice data. This enables quick and efficient understanding of the contents of the call, and improves security and business efficiency through data management on the cloud.

[0740] "Call content" refers to all words and information spoken during a call.

[0741] "Recording means" refers to any device or method that physically or digitally records the audio of a call.

[0742] "Audio Data" refers to a digital audio file that records the contents of a call.

[0743] A "voice recognition engine" refers to software or algorithms that have the function of analyzing voice data and converting it into text data.

[0744] "Text data" refers to the textual information of the speech content generated by a speech recognition engine.

[0745] A "generative model" refers to a machine learning model used to generate a summary or output in a specific format from input data.

[0746] "Means for summarizing" refers to a method or device that analyzes text data, extracts only the important points, and shortens them.

[0747] "Call history" refers to databases and log information that record the date and time of past calls, the callers, etc.

[0748] The "means for storing and displaying in association" refers to a method or device for linking the summarized text data to the call history, and storing and displaying the data so that the user can easily view it.

[0749] "Cloud storage" refers to an online storage service that allows you to store data remotely via the Internet.

[0750] "Uploading and storage means" refers to the methods and devices for transferring and securely storing audio data and summary text in cloud storage.

[0751] "Means for restoring and playing" refers to a method or device for downloading audio data stored in cloud storage and playing it back as the original audio.

[0752] An embodiment of the present invention will now be described in detail. This system is mainly composed of three elements: a terminal, a server, and a user.

[0753] Recording and uploading audio

[0754] The device automatically starts recording audio when the user starts a call. The recorded audio data is temporarily stored on the device when the call ends and then uploaded to cloud storage. For example, if you are using smart glasses, the microphone will capture the audio and the recording file will be saved in cloud storage.

[0755] Speech-to-text conversion

[0756] When the server receives the uploaded audio file, a speech recognition engine is activated and converts the audio data into text data. The speech recognition engine used in this process uses, for example, the Python speech_recognition library. For example, if a user says, "The next meeting is on Tuesday," this is accurately recorded as text.

[0757] Text summary

[0758] The server generates a summary using a generative model (e.g., an AI model such as BERT or GPT) to further simplify the converted text data. For example, the text "The next meeting is on Tuesday" is summarized as "Meeting: Tuesday." This summary is generated by inputting a prompt sentence to the generative AI model.

[0759] Save summary and link to call history

[0760] The server stores the summarized text data in a cloud database and links it to the user's call history. This allows the summary to be displayed instantly when the user checks the call history. For example, by simply tapping on a call history, the summary "Meeting: Tuesday" can be displayed.

[0761] Displaying notes in call history

[0762] When a user checks the call history, the device will display the associated summary, allowing the user to easily check the content of the call. For example, the summary "Meeting: Tuesday" will be displayed on the smart glasses display.

[0763] Data management in cloud storage

[0764] The recorded audio data and generated summary text are uploaded to cloud storage and stored securely, allowing for efficient management of large amounts of data. The stored audio data can also be restored and played back as needed.

[0765] Specific prompt examples

[0766] Conversation: "I have a conference call with a client tomorrow at 2 PM. Please send the materials by email."

[0767] Summary: "Meeting: Tomorrow at 2pm, materials to be sent."

[0768] An example of a prompt to input to a generative AI model: "Please summarize the following text: I have a conference call with a client tomorrow at 2 PM. Please also send the materials by email."

[0769] The above is a form for implementing the present invention. By using this system, users can quickly and efficiently understand the content of calls, and furthermore, data management on the cloud can improve security and business efficiency.

[0770] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0771] Step 1:

[0772] When a call is initiated, the device automatically records the call. The input for the recording is the audio from the call, and the output is audio data (e.g., a .wav file). This audio data is stored in the device's temporary memory.

[0773] Step 2:

[0774] When the call ends, the device automatically uploads the recorded audio data to cloud storage. The input is the audio data generated in step 1, and the output is the audio file stored in cloud storage.

[0775] Step 3:

[0776] The server receives the audio data uploaded from the cloud storage. The input is the audio data stored in the cloud storage, and the output is the audio data stored in the server's storage device.

[0777] Step 4:

[0778] The server uses a speech recognition engine to convert the received voice data into text data. The input is voice data (e.g., a .wav file), and the output is the text of the conversation. Specifically, the server uses the Python speech_recognition library to analyze the voice data and convert it into text information.

[0779] Step 5:

[0780] The server uses a generative AI model to summarize the converted text data. The input is the text data generated in step 4, and the output is summarized text information. A prompt sentence is used to summarize the input text for the generative AI model (e.g., GPT-3). An example of a specific prompt sentence is, "Please summarize the following text: I have a conference call with a client tomorrow at 2 p.m. Please also send the materials by email."

[0781] Step 6:

[0782] The server associates the summarized text data with the call history and stores it in a database. The input is the summarized text data and the call history information, and the output is the summary text linked to the call history. This allows the relevant summary information to be displayed immediately when the user checks the call history.

[0783] Step 7:

[0784] When a user checks the call history, the terminal displays the associated summary text. The input is the summary text obtained from the call history, and the output is the summary information displayed on the user interface, allowing the user to quickly check the key points of the call.

[0785] Step 8:

[0786] The server or device provides the functionality to restore and play back the voice data stored in the cloud storage as needed. The input is the voice file stored in the cloud storage, and the output is the voice output from the playback device (e.g., speaker or headset). This allows the user to reconfirm the contents of the call.

[0787] 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.

[0788] An embodiment of the present invention will be described in detail. The present invention is a system that combines a system that automatically takes notes on phone conversations and allows users to check the notes with a single touch from their call history, with an emotion engine that recognizes the user's emotions. This system is composed of three entities: a terminal, a server, and a user.

[0789] Recording and uploading audio

[0790] When a user starts a call, the device automatically starts recording the audio. The audio data of the call is recorded and uploaded to the server when the call ends.

[0791] Speech-to-text conversion

[0792] The server receives the uploaded audio file and activates a speech recognition engine to convert the audio data into text data. For example, if a user says, "The next meeting is on Tuesday," the audio is recorded as text: "The next meeting is on Tuesday."

[0793] Emotion recognition

[0794] The server inputs the voice data into an emotion engine to analyze the user's emotions. The emotion engine recognizes emotions from the voice and outputs the results as numerical data. For example, from the statement "The next meeting is on Tuesday," it can identify emotions such as "happiness" or "sadness."

[0795] Summarize text and add sentiment

[0796] The server inputs the converted text data into a generative model (e.g., GPT-3) to summarize the text data. Furthermore, the emotion engine adds the emotion data recognized by the text data to the summary as supplementary information. For example, the text "The next meeting is on Tuesday" becomes "Next meeting: Tuesday (emotion: happy)."

[0797] Save summary and link to call history

[0798] The server stores the summarized text data in a database and associates it with the user's call history, so that when the user checks their call history, the summary text and emotion data are instantly displayed.

[0799] Displaying notes in call history

[0800] When a user checks their call history, the device retrieves summary data and emotion data associated with the call history and displays them on the screen. For example, it displays "Next meeting: Tuesday (Emotion: Happiness)." This allows the user to easily understand the content of the call and the emotion expressed at the time.

[0801] Specific examples

[0802] Here's a concrete example: A user has a phone conversation with a colleague, with the following content and emotions:

[0803] "Schedule a meeting with the client next Tuesday in a conference room at our office." (Emotion: Happiness)

[0804] The call is recorded and the audio is uploaded to a server.

[0805] The server converts the speech to text, resulting in the text "Please schedule a meeting with the client for next Tuesday in a conference room in our office."

[0806] The emotion engine recognizes "happiness" from this statement

[0807] The generative model summarizes the text as "Customer meeting: Tuesday, office conference room."

[0808] The server adds the emotion data "happiness" to the summary text, resulting in "Customer meeting: Tuesday, office conference room (emotion: happiness)."

[0809] This summary and emotion data are stored in a database and linked to call history.

[0810] When a user checks their call history, they see a summary of the call, "Customer Meeting: Tuesday, Office Conference Room (Emotion: Happiness)," along with emotion data.

[0811] This system allows users to manage not only the content of a call, but also their emotions during the call. By checking emotional data, users can more accurately grasp the nuances and importance of the conversation, contributing to improved work efficiency and communication quality.

[0812] The processing flow will be explained below.

[0813] Step 1:

[0814] The user starts a call. This causes the device to automatically prepare to start recording audio. Specifically, the device's recording function is turned on, and the device is ready to record the audio data of the call.

[0815] Step 2:

[0816] The user ends the call. When the call ends, the device stops recording and saves the recorded audio data as a file. This audio file is uploaded to the server as soon as the call ends.

[0817] Step 3:

[0818] The server receives the uploaded audio file. The server then activates a speech recognition engine and converts the audio data into text data. For example, the server passes the audio file to the speech recognition engine and obtains the text data, "The next meeting is on Tuesday."

[0819] Step 4:

[0820] The server inputs the voice data into an emotion engine to analyze the user's emotions. The emotion engine recognizes emotions from the voice and outputs the results as numerical data. For example, from the statement "The next meeting is on Tuesday," it can identify emotions such as "happiness" or "sadness."

[0821] Step 5:

[0822] The server inputs the converted text data into a generative model, which then summarizes the text data into a concise form. For example, the text data "The next meeting is on Tuesday" is summarized as "Next meeting: Tuesday."

[0823] Step 6:

[0824] The server adds the emotion data recognized by the emotion engine to the summarized text data. This adds the emotion data as supplementary information to the summary text. For example, "Next meeting: Tuesday (emotion: happy)".

[0825] Step 7:

[0826] The server stores the summary and emotion data in a database. At the same time, it associates the summary data with the user's call history. This allows the summary and emotion data to be referenced from the call history.

[0827] Step 8:

[0828] The user checks the call history. The device retrieves the summary data and emotion data associated with the call history and displays them on the screen. The user can easily check the related summary data and emotion data by simply tapping the call history.

[0829] Step 9:

[0830] The user checks the summary data and emotion data, and transcribes it into a memo or calendar as needed. This step allows the user to smoothly move on to the next action without forgetting important information about the call content.

[0831] Example 2

[0832] 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."

[0833] Conventional call management systems only record call content and convert it into text, but do not associate it with the user's call history or recognize emotions. This makes it difficult for users to easily understand detailed call records, especially important call content and emotional nuances. Furthermore, summarizing conversations and extracting key points is time-consuming and laborious. The objective of this invention is to solve these problems and provide a system that enables efficient call content management and immediate understanding of important information.

[0834] 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 a means for recording the contents of the call, a means for converting the recorded voice data into text data using a voice recognition engine, a means for analyzing the user's emotions from the text data using an emotion analysis engine, a means for summarizing the converted text data using a generative model, a means for adding emotion data to the summarized text data, and a means for saving and displaying the summarized text data and emotion data in association with the call history. This not only allows the user to easily understand the contents of the call, but also allows the user to simultaneously manage emotions during the call, making it possible to instantly check the nuances and important points of the conversation.

[0835] "Means for recording phone calls" refers to the ability to capture audio when a call begins and save that audio as a file when the call ends.

[0836] "Speech recognition engine" refers to software or hardware technology that has the function of analyzing voice data and converting it into corresponding text data.

[0837] "Means for analyzing user emotions from text data using an emotion analysis engine" refers to a function that receives text data as input, identifies the user's emotions from the content, and outputs them as a number or category.

[0838] "Means for summarizing converted text data using a generative model" refers to the function of using a generative AI model to summarize long text data and extract only the important points.

[0839] "Means for adding emotion data to summarized text data" refers to a function that adds emotion analysis results to summarized text and provides them as supplementary information.

[0840] "Means for storing and displaying summarized text data and emotional data in association with call history" refers to a function that stores summarized text data and emotional data in a database and links them to call history, thereby displaying this data simultaneously when the user checks the history.

[0841] According to an embodiment of the present invention, a system is provided that automates the entire process of recording, converting to text, analyzing emotions, and summarizing call content. This system records call content and associates it with a user's call history, enabling efficient management of call content and emotion information.

[0842] Recording and uploading audio

[0843] When a user starts a call, the device automatically starts recording audio. This is achieved using the device's recording functionality (e.g., the MediaRecorder class in Android or the AVAudioRecorder class in iOS). The recorded audio data is saved as a temporary file. When the call ends, the device uploads this audio data to the server using an HTTP POST request.

[0844] Speech-to-text conversion

[0845] The server receives the uploaded audio file. It invokes a speech recognition engine (e.g., Google Speech-to-Text API) to convert the audio data into text data. Specifically, the audio file is sent as an API request, and the server receives the text data as a response. For example, if a user says, "The next meeting is on Tuesday," this speech is converted into the text data, "The next meeting is on Tuesday."

[0846] Emotion recognition

[0847] The server inputs the voice data into a sentiment analysis engine (e.g., IBM Watson's Tone Analyzer). The sentiment analysis engine analyzes the user's emotions from the voice and text data and outputs the results as numerical or categorical data. For example, the server sends the text data "The next meeting is on Tuesday" to the sentiment analysis engine and receives the sentiment analysis result of "Happy."

[0848] Summarize text and add sentiment

[0849] The server then uses a generative AI model (e.g., OpenAI GPT-3) to summarize the converted text data. The server sends the generative AI model a prompt like this:

[0850] "Please summarize the following conversation: The next meeting is on Tuesday."

[0851] The generative AI model responds to this prompt sentence and generates a summary such as "Next meeting: Tuesday." Furthermore, the sentiment analysis engine adds emotional data recognized by the summary text to the summary text. For example, emotional data is added to the summary sentence, such as "Next meeting: Tuesday (Emotion: Happy)."

[0852] Save summary data and link to call history

[0853] The server stores the summarized text data and emotion data in a database and associates them with the user's call history. Specifically, the server links the summarized data to specific records in the call history database, enabling efficient management of call content and emotion information.

[0854] Displaying notes in call history

[0855] When a user checks their call history, the device retrieves the relevant summary data and emotion data from the server and displays it on the screen. For example, it might display "Next meeting: Tuesday (Emotion: Happiness)," allowing the user to understand the content of the call and the emotion expressed at the time at a glance. This allows users to quickly check detailed information about particularly important calls.

[0856] As described above, the present invention provides a system that effectively integrates all processes from recording call content to emotion recognition, summarization, storage, and display, which will greatly contribute to improving work efficiency and the quality of communication.

[0857] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0858] Step 1:

[0859] When a user starts a call, the device automatically starts recording audio. This is achieved by using the call start event as a trigger to launch the device's recording function (e.g., the MediaRecorder class in Android or the AVAudioRecorder class in iOS). The input is the call start event, and the output is the audio data being recorded. The recorded audio data is saved as a temporary file on the device.

[0860] Step 2:

[0861] When the call ends, the device uploads the audio data stored in the temporary file to the server. Specifically, it uses an HTTP POST request to send the audio file to a specific endpoint on the server. The input is the recorded audio file, and the output is the audio data uploaded to the server. The audio file is then sent to the server and ready for the next processing step.

[0862] Step 3:

[0863] The server receives the uploaded audio file. The server invokes a speech recognition engine (e.g., Google Speech-to-Text API) to convert the audio data into text data. The input is the audio file uploaded to the server, and the output is the converted text data. Specifically, the server sends the audio file as an API request and receives text data as a response. For example, the generated text is "The next meeting is on Tuesday."

[0864] Step 4:

[0865] The server inputs the voice data into a sentiment analysis engine (e.g., IBM Watson's Tone Analyzer). The input is converted text data, and the output is analyzed emotional data. The sentiment analysis engine analyzes the user's emotions from the voice and text data and outputs the results as numerical or categorical data. The server sends the text data to the API endpoint of the sentiment analysis engine and receives the sentiment analysis result, for example, "happy."

[0866] Step 5:

[0867] The server inputs the converted text data into a generative AI model (e.g., OpenAI GPT-3) and summarizes the text data. The input is the converted text data and a prompt, and the output is the summary text. The server sends the following prompt to the generative AI model:

[0868] "Please summarize the following conversation: The next meeting is on Tuesday."

[0869] The generative AI model responds to this prompt and generates a summary such as "Next meeting: Tuesday."

[0870] Step 6:

[0871] The server adds emotional data to the summarized text data. The input is the summary text and emotional data, and the output is the summary text with the added emotional data. For example, emotional data is added to the summary sentence, such as "Next meeting: Tuesday (emotion: happy)." As a result, the content of the call and the emotions felt at the time can be understood at a glance.

[0872] Step 7:

[0873] The server stores the summarized text data and emotion data in a database and associates it with the user's call history. The input is the summarized text and emotion data, and the output is data associated with the call history. Specifically, the summary data and emotion data are linked to each call history record. This process allows the necessary information to be managed quickly and efficiently.

[0874] Step 8:

[0875] When a user checks their call history, the device retrieves the relevant summary data and emotion data from the server and displays them on the screen. The input is a request for call history, and the output is a screen displaying the summary data and emotion data. For example, information is displayed in the format "Next meeting: Tuesday (emotion: happy)." As a result, the user can easily check the content of the call and the emotion at the time.

[0876] (Application example 2)

[0877] 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."

[0878] Autonomous vehicles require a means to efficiently record conversations and instructions between passengers and quickly obtain necessary information by summarizing it. However, conventional technology only records voice data and does not analyze or summarize emotional data, making it difficult to grasp the nuances and importance of the information. This makes it easy to overlook passenger intentions and important information, increasing the risk of problems occurring.

[0879] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0880] In this invention, the server includes means for recording the contents of the call, means for converting the recorded voice data into text data using a voice recognition engine, means for summarizing the text data using a generative model, means for saving and displaying the summarized text data together with emotion data added by an emotion recognition engine, and means for saving and displaying the data in association with the call history. This not only records the voice but also performs summarization and emotion analysis, making it possible to prevent important information from being overlooked and to quickly and accurately grasp the intentions and importance of passengers.

[0881] A "call recording device" is any device or software that is capable of adequately capturing and digitally recording the audio of a call.

[0882] "Means for converting recorded voice data into text using a voice recognition engine" means software or algorithms that use voice recognition technology to convert recorded voice files into text format.

[0883] "Means for summarizing text data using a generative model" refers to an algorithm that uses a generative AI model to extract key information from long text data and output it as a short summary.

[0884] The "means for saving and displaying summarized text data together with emotion data added by an emotion recognition engine" refers to a system for adding emotion information analyzed by an emotion recognition system to summarized text data, saving the data, and displaying it so that the user can easily access it.

[0885] "Means for storing and displaying data in association with call history" refers to technology for linking and associating data containing summaries and emotional information with call history, thereby picking up and displaying the content of past calls and emotions.

[0886] MODE FOR CARRYING OUT THE INVENTION

[0887] The system for carrying out the present invention mainly has the following configuration.

[0888] Recording and uploading audio

[0889] When a user starts a call, the device automatically starts recording audio. It uses the microphone to capture audio data and continues recording for the duration of the call. When the call ends, the recorded audio data is uploaded to the server.

[0890] Speech-to-text conversion

[0891] When the server receives the recorded voice data, it activates a speech recognition engine and converts the voice data into text data. In this case, the speech_recognition library is used. For example, if a user says, "Please make the next stop Shibuya Station," the speech is recorded as text saying, "Please make the next stop Shibuya Station."

[0892] Emotion recognition

[0893] The server inputs the converted text data into an emotion recognition engine to analyze the user's emotions. A custom module called EmotionRecognition is used for this emotion recognition. For example, the same utterance "Please make the next stop Shibuya Station" can identify the emotion "neutral."

[0894] Summarize text and add sentiment

[0895] The server then inputs the converted text data into a generative AI model to summarize it. This summarization is performed using a generative model such as T5 from the Transformers library. Emotion data recognized by an emotion recognition engine is then added to the summary. For example, the text "Please make the next stop Shibuya Station" is summarized as "Stops at Shibuya Station," and the emotion data "neutral" is added to make it "Stops at Shibuya Station (Emotion: Neutral)."

[0896] Save to database and link to call history

[0897] Once the summary and emotion data are created, the server stores them in a database and associates them with the call history, so that when a user checks their call history, the summary and emotion data are instantly displayed.

[0898] Displaying notes in call history

[0899] When a user checks their call history, the device retrieves summary data and emotion data associated with the call history and displays them on the screen. Specifically, it displays the data in the form of "Shibuya Station Stop (Emotion: Neutral)." This system allows users to easily understand the content of the call and the emotion expressed at the time.

[0900] Specific examples

[0901] Scenario: A passenger gives instructions on board, such as "Next stop: Shibuya Station."

[0902] Example prompt sentence:

[0903] "Next stop please be Shibuya Station."

[0904] Example output:

[0905] Shibuya Station (Emotion: Neutral)

[0906] This invention converts voice data into text and automatically displays summaries and memos with emotion recognition, enabling quick and accurate understanding of information. This is expected to reduce troubles in self-driving vehicles and improve the passenger experience.

[0907] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0908] Step 1:

[0909] When the device detects the start of a call, it automatically starts recording audio. Specifically, the device's microphone captures the audio data and continues recording until the call ends. The recorded audio data is saved in raw audio format. The input is the audio during the call, and the output is the recorded audio file.

[0910] Step 2:

[0911] When the call ends, the device uploads the recorded audio data to the server. At this time, the audio data is sent to the server via the network. The input is the recorded audio file, and the output is the audio data stored on the server.

[0912] Step 3:

[0913] The server inputs the received voice data into a voice recognition engine and converts it into text data. This process uses the speech_recognition library to convert the voice file into text format. The input is voice data and the output is text data.

[0914] Step 4:

[0915] The server inputs the converted text data into an emotion recognition engine to analyze the user's emotions. The emotion recognition engine uses the EmotionRecognition module to extract emotional information from the text data. The input is text data, and the output is emotional data.

[0916] Step 5:

[0917] The server inputs the text data into a generative AI model to generate a summary. In this case, it uses the T5 model from the Transformers library to extract key information from long text data and generate a short summary. The input is text data, and the output is a summary.

[0918] Step 6:

[0919] The server adds emotional data to the generated summary to create the final summary. The server then integrates and formats the generated summary and emotional data. The input is the summary and emotional data, and the output is summary data that includes emotional information.

[0920] Step 7:

[0921] The server stores this summary data in a database and associates it with the call history. A new entry is added to the database and linked to the call history. The input is the summary data including emotion information, and the output is a database entry associated with the call history.

[0922] Step 8:

[0923] When a user checks the call history, the device retrieves the associated summary data from the database and displays it on the screen. The device sends a query to the database and displays the retrieved summary data and emotion information on the screen. The input is the operation to reference the call history, and the output is the summary data displayed on the screen.

[0924] 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.

[0925] 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.

[0926] 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.

[0927] [Fourth embodiment]

[0928] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0929] 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.

[0930] 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).

[0931] 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.

[0932] 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.

[0933] 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).

[0934] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0935] 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.

[0936] 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.

[0937] 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.

[0938] 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.

[0939] 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.

[0940] 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."

[0941] The present invention provides a system that automatically memos the contents of telephone conversations and allows users to check the memos with a single touch from the call history. This system is mainly composed of three entities: a terminal, a server, and a user.

[0942] Recording and uploading audio

[0943] The device automatically starts recording audio when the user starts a call. The recorded audio is saved when the call ends and uploaded to a server, allowing the call contents to be stored in digital form.

[0944] Speech-to-text conversion

[0945] When the server receives the uploaded audio file, a speech recognition engine is activated and converts the audio data into text data. For example, if a user says, "The next meeting is on Tuesday," this is recorded as text.

[0946] Text summary

[0947] The server inputs the converted text data into a generative model (e.g., GPT-3) to generate a summary. For example, the text "The next meeting is on Tuesday" is summarized as "Next meeting: Tuesday." This summary aims to simplify the information.

[0948] Save summary and link to call history

[0949] The server stores the summarized text data in a database and links it to the user's call history, so that when the user checks their call history, the summary text is immediately displayed.

[0950] Displaying notes in call history

[0951] When users check their call history, the device will display the associated summary, allowing them to easily check the content of the call. For example, simply tapping on a call history will display the summary "Next meeting: Tuesday."

[0952] Specific examples

[0953] Here's a concrete example: A user talks to a sales representative on the phone and says the following:

[0954] "Please schedule a meeting with a client for next Tuesday in the office conference room."

[0955] Calls are recorded and uploaded to a server

[0956] The server converts the speech to text, generating the following text: "Please schedule a meeting with the client for next Tuesday in a conference room in our office."

[0957] The generative model summarizes this as "Customer meeting: Tuesday, office conference room."

[0958] This summary is stored in a database and linked to the call history.

[0959] When users check their call history, a summary is displayed, allowing them to see the details of the call at a glance.

[0960] In this way, the present invention allows users to record and quickly check the contents of calls, thereby preventing users from overlooking important information and improving work efficiency.

[0961] The processing flow will be explained below.

[0962] Step 1:

[0963] The user starts a call. This causes the device to automatically prepare to start recording audio. Specifically, the device's recording function is turned on, and the device is ready to record the audio data of the call.

[0964] Step 2:

[0965] The user ends the call. When the call ends, the device stops recording and saves the recorded audio data as a file. This audio file is uploaded to the server as soon as the call ends.

[0966] Step 3:

[0967] The server receives the uploaded audio file. The server then activates a speech recognition engine and converts the audio data into text data. For example, the server passes the audio file to the speech recognition engine and obtains the text data, "The next meeting is on Tuesday."

[0968] Step 4:

[0969] The server inputs the converted text data into a generative model, which then summarizes the text data into a concise form. For example, the text data "The next meeting is on Tuesday" is summarized as "Next meeting: Tuesday."

[0970] Step 5:

[0971] The server saves the summarized text data in a database. At the same time, it associates this summary data with the user's call history, allowing the summary data to be referenced from the call history.

[0972] Step 6:

[0973] The user checks the call history. The device retrieves the summary data associated with the call history and displays it on the screen. The user can easily check the related summary data by simply tapping the call history.

[0974] Step 7:

[0975] The user checks the summary data and transcribes it into a memo or calendar as necessary. This step allows the user to smoothly move on to the next action without forgetting any important information about the call content.

[0976] Example 1

[0977] 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."

[0978] Conventional call recording systems require the manual recording of call content, which can lead to important information being overlooked. Furthermore, reviewing call content takes time, and there are few ways to improve work efficiency. Therefore, there is a need for a system that automatically records call content and can efficiently summarize and search it.

[0979] 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.

[0980] In this invention, the server includes a means for recording the contents of the call, a means for converting the recorded voice data into text data using a voice recognition system, a means for summarizing the text data using a generative AI model, a means for storing and displaying the summarized text data in association with the communication history, and a means for displaying the summary when the user checks the communication history. This enables the contents of the call to be automatically recorded and efficiently summarized and searched.

[0981] "Call content" refers to information and conversations exchanged over the telephone or other voice communications.

[0982] "Recording" refers to the process of storing audio data in digital or analog form.

[0983] "Audio Data" refers to audio information recorded in digital or analog format.

[0984] A "speech recognition system" refers to the technology that allows a computer to understand human speech and convert it into text.

[0985] "Text data" refers to digital data expressed as character information.

[0986] A "generative AI model" refers to an artificial intelligence model that learns using large amounts of data and makes predictions and generates results.

[0987] A "summary" is a short sentence that concisely summarizes the original information.

[0988] "Communication history" refers to a record of calls and messages made, including information such as the caller, recipient, and time.

[0989] "Storage" refers to keeping data or information in a fixed location.

[0990] "Display" refers to outputting information such as text or graphics to a screen or display.

[0991] "User" refers to a person who uses a system or application.

[0992] The present invention is a system that automatically records and summarizes the contents of phone calls and associates them with call history. This system is mainly composed of three entities: a terminal, a server, and a user.

[0993] Recording and uploading audio

[0994] When a user starts a call, the device automatically starts recording audio. This function uses the device's built-in microphone and a voice recording app (for example, the Voice Memos app on iOS). When the call ends, the recorded audio file is saved in a specific folder on the device. The recorded file is then uploaded to the server using a POST request using the HTTP protocol.

[0995] Speech-to-text conversion

[0996] When the server receives the uploaded audio file, a speech recognition system (e.g., Google Cloud Speech-to-Text API) is activated to convert the audio data into text data. At this time, the audio file is saved in a storage server (e.g., Amazon S3) and then analyzed by the speech recognition system. For example, if a user says, "The next meeting is on Tuesday," this content is recorded as text data.

[0997] Text summary

[0998] The server inputs the generated text data into a generative AI model (e.g., OpenAI's GPT-3) to generate a summary. As a specific example, the server receives the text "The next meeting is on Tuesday" and sends the prompt "Please summarize this text: 'The next meeting is on Tuesday.'" to the generative model. As a result, the generative model generates the summary "Next meeting: Tuesday."

[0999] Save summary and link to call history

[1000] The server stores the summary in a database (e.g., MySQL) and links it to the user's call history. This linking allows the user to instantly retrieve the summary when checking the call history. For example, by associating the call summary data with the call date and time and the other party's phone number, the call history can be managed effectively.

[1001] Displaying notes in call history

[1002] When a user checks the call history on their device, the device requests the summary stored on the server. The server then sends back the corresponding summary, which is then displayed in the device's call history app. This process allows users to instantly view summaries such as "Next meeting: Tuesday" by simply tapping on the call history.

[1003] Example operation

[1004] As a real-world use case, imagine a user is talking to a sales rep on the phone and says, "Please schedule a meeting with a customer next Tuesday in a conference room in my office." The system would automatically execute the following process:

[1005] 1. The call will be recorded and an audio file will be created using your device's built-in microphone and audio recording app.

[1006] 2. The audio file is uploaded to the server.

[1007] 3. The server converts the received voice file into text data using a voice recognition system, and the generated text is, "Please schedule a meeting with the client for next Tuesday. The location should be in the office conference room."

[1008] 4. This text data is fed into a generative AI model, which generates a summary: "Customer meeting: Tuesday, office conference room."

[1009] 5. The summary is stored in a database and linked to the call history.

[1010] 6. When the user checks their call history, a summary of "Customer Meeting: Tuesday, Office Conference Room" is instantly displayed.

[1011] In this way, the present invention allows users to record and quickly check the contents of calls, and improves work efficiency without overlooking important information.

[1012] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1013] Step 1:

[1014] The device automatically starts recording audio when the user starts a call. Specifically, the device's call management app detects the call start event and activates the audio recording module. Once recording begins, audio data is collected in real time and temporarily stored in the device's internal memory.

[1015] Input: User call start event

[1016] Output: Audio data being recorded

[1017] Step 2:

[1018] The device stops recording when the call ends and saves the collected audio data in a specific folder. The call management app detects the call end event and stops the recording module, completing the recording. The recording file is then uploaded to the server using the HTTP protocol.

[1019] Input: Call end event and recording audio data

[1020] Output: Audio file uploaded to the server

[1021] Step 3:

[1022] The server receives the HTTP POST request and saves the uploaded audio file on the storage server. Next, it calls the speech recognition system to convert the audio data into text data. This process takes the audio file and outputs it as text using the speech recognition API. For example, if the audio data says "The next meeting is on Tuesday," it will be converted into text data.

[1023] Input: Uploaded audio file

[1024] Output: Converted text data

[1025] Step 4:

[1026] The server inputs the generated text data into the generative AI model to generate a summary. Specifically, it forms a prompt sentence and sends an API request to the generative model. The text "The next meeting is on Tuesday" is input into the generative AI model, and the summary "Next meeting: Tuesday" is returned.

[1027] Input: Generated text data

[1028] Output: Summarized text data

[1029] Step 5:

[1030] The server stores the summary in a database and links it to the call history, which contains metadata such as the call date and time and the caller's phone number, and associates the summary with that information.

[1031] Input: Summarized text data

[1032] Output: A link between the summary stored in the database and the call history

[1033] Step 6:

[1034] When a user checks the call history on their device, the device sends a request to the server to retrieve the stored summary. When the user opens the call history app and taps on a specific call history, the summary linked to that call history is displayed on the device.

[1035] Input: User operation to check call history

[1036] Output: A summary of the call displayed on your device

[1037] (Application example 1)

[1038] 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."

[1039] While current call recording systems can accurately record call content, it is difficult to quickly and efficiently understand the content. Furthermore, they lack the functionality to store call records on the cloud and play them back as needed, creating a need for improved security and operational efficiency.

[1040] 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.

[1041] In this invention, the server includes means for recording the contents of the call, means for converting the recorded voice data into text data using a voice recognition engine, means for summarizing the text data using a generative model, means for saving and displaying the summarized text data in association with the call history, means for uploading and saving the voice data and the summarized text to cloud storage, and means for restoring and playing back the uploaded voice data. This enables quick and efficient understanding of the contents of the call, and improves security and business efficiency through data management on the cloud.

[1042] "Call content" refers to all words and information spoken during a call.

[1043] "Recording means" refers to any device or method that physically or digitally records the audio of a call.

[1044] "Audio Data" refers to a digital audio file that records the contents of a call.

[1045] A "voice recognition engine" refers to software or algorithms that have the function of analyzing voice data and converting it into text data.

[1046] "Text data" refers to the textual information of the speech content generated by a speech recognition engine.

[1047] A "generative model" refers to a machine learning model used to generate a summary or output in a specific format from input data.

[1048] "Means for summarizing" refers to a method or device that analyzes text data, extracts only the important points, and shortens them.

[1049] "Call history" refers to databases and log information that record the date and time of past calls, the callers, etc.

[1050] The "means for storing and displaying in association" refers to a method or device for linking the summarized text data to the call history, and storing and displaying the data so that the user can easily view it.

[1051] "Cloud storage" refers to an online storage service that allows you to store data remotely via the Internet.

[1052] "Uploading and storage means" refers to the methods and devices for transferring and securely storing audio data and summary text in cloud storage.

[1053] "Means for restoring and playing" refers to a method or device for downloading audio data stored in cloud storage and playing it back as the original audio.

[1054] An embodiment of the present invention will now be described in detail. This system is mainly composed of three elements: a terminal, a server, and a user.

[1055] Recording and uploading audio

[1056] The device automatically starts recording audio when the user starts a call. The recorded audio data is temporarily stored on the device when the call ends and then uploaded to cloud storage. For example, if you are using smart glasses, the microphone will capture the audio and the recording file will be saved in cloud storage.

[1057] Speech-to-text conversion

[1058] When the server receives the uploaded audio file, a speech recognition engine is activated and converts the audio data into text data. The speech recognition engine used in this process uses, for example, the Python speech_recognition library. For example, if a user says, "The next meeting is on Tuesday," this is accurately recorded as text.

[1059] Text summary

[1060] The server generates a summary using a generative model (e.g., an AI model such as BERT or GPT) to further simplify the converted text data. For example, the text "The next meeting is on Tuesday" is summarized as "Meeting: Tuesday." This summary is generated by inputting a prompt sentence to the generative AI model.

[1061] Save summary and link to call history

[1062] The server stores the summarized text data in a cloud database and links it to the user's call history. This allows the summary to be displayed instantly when the user checks the call history. For example, by simply tapping on a call history, the summary "Meeting: Tuesday" can be displayed.

[1063] Displaying notes in call history

[1064] When a user checks the call history, the device will display the associated summary, allowing the user to easily check the content of the call. For example, the summary "Meeting: Tuesday" will be displayed on the smart glasses display.

[1065] Data management in cloud storage

[1066] The recorded audio data and generated summary text are uploaded to cloud storage and stored securely, allowing for efficient management of large amounts of data. The stored audio data can also be restored and played back as needed.

[1067] Specific prompt examples

[1068] Conversation: "I have a conference call with a client tomorrow at 2 PM. Please send the materials by email."

[1069] Summary: "Meeting: Tomorrow at 2pm, materials to be sent."

[1070] An example of a prompt to input to a generative AI model: "Please summarize the following text: I have a conference call with a client tomorrow at 2 PM. Please also send the materials by email."

[1071] The above is a form for implementing the present invention. By using this system, users can quickly and efficiently understand the content of calls, and furthermore, data management on the cloud can improve security and business efficiency.

[1072] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1073] Step 1:

[1074] When a call is initiated, the device automatically records the call. The input for the recording is the audio from the call, and the output is audio data (e.g., a .wav file). This audio data is stored in the device's temporary memory.

[1075] Step 2:

[1076] When the call ends, the device automatically uploads the recorded audio data to cloud storage. The input is the audio data generated in step 1, and the output is the audio file stored in cloud storage.

[1077] Step 3:

[1078] The server receives the audio data uploaded from the cloud storage. The input is the audio data stored in the cloud storage, and the output is the audio data stored in the server's storage device.

[1079] Step 4:

[1080] The server uses a speech recognition engine to convert the received voice data into text data. The input is voice data (e.g., a .wav file), and the output is the text of the conversation. Specifically, the server uses the Python speech_recognition library to analyze the voice data and convert it into text information.

[1081] Step 5:

[1082] The server uses a generative AI model to summarize the converted text data. The input is the text data generated in step 4, and the output is summarized text information. A prompt sentence is used to summarize the input text for the generative AI model (e.g., GPT-3). An example of a specific prompt sentence is, "Please summarize the following text: I have a conference call with a client tomorrow at 2 p.m. Please also send the materials by email."

[1083] Step 6:

[1084] The server associates the summarized text data with the call history and stores it in a database. The input is the summarized text data and the call history information, and the output is the summary text linked to the call history. This allows the relevant summary information to be displayed immediately when the user checks the call history.

[1085] Step 7:

[1086] When a user checks the call history, the terminal displays the associated summary text. The input is the summary text obtained from the call history, and the output is the summary information displayed on the user interface, allowing the user to quickly check the key points of the call.

[1087] Step 8:

[1088] The server or device provides the functionality to restore and play back the voice data stored in the cloud storage as needed. The input is the voice file stored in the cloud storage, and the output is the voice output from the playback device (e.g., speaker or headset). This allows the user to reconfirm the contents of the call.

[1089] 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.

[1090] An embodiment of the present invention will be described in detail. The present invention is a system that combines a system that automatically takes notes on phone conversations and allows users to check the notes with a single touch from their call history, with an emotion engine that recognizes the user's emotions. This system is composed of three entities: a terminal, a server, and a user.

[1091] Recording and uploading audio

[1092] When a user starts a call, the device automatically starts recording the audio. The audio data of the call is recorded and uploaded to the server when the call ends.

[1093] Speech-to-text conversion

[1094] The server receives the uploaded audio file and activates a speech recognition engine to convert the audio data into text data. For example, if a user says, "The next meeting is on Tuesday," the audio is recorded as text: "The next meeting is on Tuesday."

[1095] Emotion recognition

[1096] The server inputs the voice data into an emotion engine to analyze the user's emotions. The emotion engine recognizes emotions from the voice and outputs the results as numerical data. For example, from the statement "The next meeting is on Tuesday," it can identify emotions such as "happiness" or "sadness."

[1097] Summarize text and add sentiment

[1098] The server inputs the converted text data into a generative model (e.g., GPT-3) to summarize the text data. Furthermore, the emotion engine adds the emotion data recognized by the text data to the summary as supplementary information. For example, the text "The next meeting is on Tuesday" becomes "Next meeting: Tuesday (emotion: happy)."

[1099] Save summary and link to call history

[1100] The server stores the summarized text data in a database and associates it with the user's call history, so that when the user checks their call history, the summary text and emotion data are instantly displayed.

[1101] Displaying notes in call history

[1102] When a user checks their call history, the device retrieves summary data and emotion data associated with the call history and displays them on the screen. For example, it displays "Next meeting: Tuesday (Emotion: Happiness)." This allows the user to easily understand the content of the call and the emotion expressed at the time.

[1103] Specific examples

[1104] Here's a concrete example: A user has a phone conversation with a colleague, with the following content and emotions:

[1105] "Schedule a meeting with the client next Tuesday in a conference room at our office." (Emotion: Happiness)

[1106] The call is recorded and the audio is uploaded to a server.

[1107] The server converts the speech to text, resulting in the text "Please schedule a meeting with the client for next Tuesday in a conference room in our office."

[1108] The emotion engine recognizes "happiness" from this statement

[1109] The generative model summarizes the text as "Customer meeting: Tuesday, office conference room."

[1110] The server adds the emotion data "happiness" to the summary text, resulting in "Customer meeting: Tuesday, office conference room (emotion: happiness)."

[1111] This summary and emotion data are stored in a database and linked to call history.

[1112] When a user checks their call history, they see a summary of the call, "Customer Meeting: Tuesday, Office Conference Room (Emotion: Happiness)," along with emotion data.

[1113] This system allows users to manage not only the content of a call, but also their emotions during the call. By checking emotional data, users can more accurately grasp the nuances and importance of the conversation, contributing to improved work efficiency and communication quality.

[1114] The processing flow will be explained below.

[1115] Step 1:

[1116] The user starts a call. This causes the device to automatically prepare to start recording audio. Specifically, the device's recording function is turned on, and the device is ready to record the audio data of the call.

[1117] Step 2:

[1118] The user ends the call. When the call ends, the device stops recording and saves the recorded audio data as a file. This audio file is uploaded to the server as soon as the call ends.

[1119] Step 3:

[1120] The server receives the uploaded audio file. The server then activates a speech recognition engine and converts the audio data into text data. For example, the server passes the audio file to the speech recognition engine and obtains the text data, "The next meeting is on Tuesday."

[1121] Step 4:

[1122] The server inputs the voice data into an emotion engine to analyze the user's emotions. The emotion engine recognizes emotions from the voice and outputs the results as numerical data. For example, from the statement "The next meeting is on Tuesday," it can identify emotions such as "happiness" or "sadness."

[1123] Step 5:

[1124] The server inputs the converted text data into a generative model, which then summarizes the text data into a concise form. For example, the text data "The next meeting is on Tuesday" is summarized as "Next meeting: Tuesday."

[1125] Step 6:

[1126] The server adds the emotion data recognized by the emotion engine to the summarized text data. This adds the emotion data as supplementary information to the summary text. For example, "Next meeting: Tuesday (emotion: happy)".

[1127] Step 7:

[1128] The server stores the summary and emotion data in a database. At the same time, it associates the summary data with the user's call history. This allows the summary and emotion data to be referenced from the call history.

[1129] Step 8:

[1130] The user checks the call history. The device retrieves the summary data and emotion data associated with the call history and displays them on the screen. The user can easily check the related summary data and emotion data by simply tapping the call history.

[1131] Step 9:

[1132] The user checks the summary data and emotion data, and transcribes it into a memo or calendar as needed. This step allows the user to smoothly move on to the next action without forgetting important information about the call content.

[1133] Example 2

[1134] 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."

[1135] Conventional call management systems only record call content and convert it into text, but do not associate it with the user's call history or recognize emotions. This makes it difficult for users to easily understand detailed call records, especially important call content and emotional nuances. Furthermore, summarizing conversations and extracting key points is time-consuming and laborious. The objective of this invention is to solve these problems and provide a system that enables efficient call content management and immediate understanding of important information.

[1136] 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 a means for recording the contents of the call, a means for converting the recorded voice data into text data using a voice recognition engine, a means for analyzing the user's emotions from the text data using an emotion analysis engine, a means for summarizing the converted text data using a generative model, a means for adding emotion data to the summarized text data, and a means for saving and displaying the summarized text data and emotion data in association with the call history. This not only allows the user to easily understand the contents of the call, but also allows the user to simultaneously manage emotions during the call, making it possible to instantly check the nuances and important points of the conversation.

[1137] "Means for recording phone calls" refers to the ability to capture audio when a call begins and save that audio as a file when the call ends.

[1138] "Speech recognition engine" refers to software or hardware technology that has the function of analyzing voice data and converting it into corresponding text data.

[1139] "Means for analyzing user emotions from text data using an emotion analysis engine" refers to a function that receives text data as input, identifies the user's emotions from the content, and outputs them as a number or category.

[1140] "Means for summarizing converted text data using a generative model" refers to the function of using a generative AI model to summarize long text data and extract only the important points.

[1141] "Means for adding emotion data to summarized text data" refers to a function that adds emotion analysis results to summarized text and provides them as supplementary information.

[1142] "Means for storing and displaying summarized text data and emotional data in association with call history" refers to a function that stores summarized text data and emotional data in a database and links them to call history, thereby displaying this data simultaneously when the user checks the history.

[1143] According to an embodiment of the present invention, a system is provided that automates the entire process of recording, converting to text, analyzing emotions, and summarizing call content. This system records call content and associates it with a user's call history, enabling efficient management of call content and emotion information.

[1144] Recording and uploading audio

[1145] When a user starts a call, the device automatically starts recording audio. This is achieved using the device's recording functionality (e.g., the MediaRecorder class in Android or the AVAudioRecorder class in iOS). The recorded audio data is saved as a temporary file. When the call ends, the device uploads this audio data to the server using an HTTP POST request.

[1146] Speech-to-text conversion

[1147] The server receives the uploaded audio file. It invokes a speech recognition engine (e.g., Google Speech-to-Text API) to convert the audio data into text data. Specifically, the audio file is sent as an API request, and the server receives the text data as a response. For example, if a user says, "The next meeting is on Tuesday," this speech is converted into the text data, "The next meeting is on Tuesday."

[1148] Emotion recognition

[1149] The server inputs the voice data into a sentiment analysis engine (e.g., IBM Watson's Tone Analyzer). The sentiment analysis engine analyzes the user's emotions from the voice and text data and outputs the results as numerical or categorical data. For example, the server sends the text data "The next meeting is on Tuesday" to the sentiment analysis engine and receives the sentiment analysis result of "Happy."

[1150] Summarize text and add sentiment

[1151] The server then uses a generative AI model (e.g., OpenAI GPT-3) to summarize the converted text data. The server sends the generative AI model a prompt like this:

[1152] "Please summarize the following conversation: The next meeting is on Tuesday."

[1153] The generative AI model responds to this prompt sentence and generates a summary such as "Next meeting: Tuesday." Furthermore, the sentiment analysis engine adds emotional data recognized by the summary text to the summary text. For example, emotional data is added to the summary sentence, such as "Next meeting: Tuesday (Emotion: Happy)."

[1154] Save summary data and link to call history

[1155] The server stores the summarized text data and emotion data in a database and associates them with the user's call history. Specifically, the server links the summarized data to specific records in the call history database, enabling efficient management of call content and emotion information.

[1156] Displaying notes in call history

[1157] When a user checks their call history, the device retrieves the relevant summary data and emotion data from the server and displays it on the screen. For example, it might display "Next meeting: Tuesday (Emotion: Happiness)," allowing the user to understand the content of the call and the emotion expressed at the time at a glance. This allows users to quickly check detailed information about particularly important calls.

[1158] As described above, the present invention provides a system that effectively integrates all processes from recording call content to emotion recognition, summarization, storage, and display, which will greatly contribute to improving work efficiency and the quality of communication.

[1159] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1160] Step 1:

[1161] When a user starts a call, the device automatically starts recording audio. This is achieved by using the call start event as a trigger to launch the device's recording function (e.g., the MediaRecorder class in Android or the AVAudioRecorder class in iOS). The input is the call start event, and the output is the audio data being recorded. The recorded audio data is saved as a temporary file on the device.

[1162] Step 2:

[1163] When the call ends, the device uploads the audio data stored in the temporary file to the server. Specifically, it uses an HTTP POST request to send the audio file to a specific endpoint on the server. The input is the recorded audio file, and the output is the audio data uploaded to the server. The audio file is then sent to the server and ready for the next processing step.

[1164] Step 3:

[1165] The server receives the uploaded audio file. The server invokes a speech recognition engine (e.g., Google Speech-to-Text API) to convert the audio data into text data. The input is the audio file uploaded to the server, and the output is the converted text data. Specifically, the server sends the audio file as an API request and receives text data as a response. For example, the generated text is "The next meeting is on Tuesday."

[1166] Step 4:

[1167] The server inputs the voice data into a sentiment analysis engine (e.g., IBM Watson's Tone Analyzer). The input is converted text data, and the output is analyzed emotional data. The sentiment analysis engine analyzes the user's emotions from the voice and text data and outputs the results as numerical or categorical data. The server sends the text data to the API endpoint of the sentiment analysis engine and receives the sentiment analysis result, for example, "happy."

[1168] Step 5:

[1169] The server inputs the converted text data into a generative AI model (e.g., OpenAI GPT-3) and summarizes the text data. The input is the converted text data and a prompt, and the output is the summary text. The server sends the following prompt to the generative AI model:

[1170] "Please summarize the following conversation: The next meeting is on Tuesday."

[1171] The generative AI model responds to this prompt and generates a summary such as "Next meeting: Tuesday."

[1172] Step 6:

[1173] The server adds emotional data to the summarized text data. The input is the summary text and emotional data, and the output is the summary text with the added emotional data. For example, emotional data is added to the summary sentence, such as "Next meeting: Tuesday (emotion: happy)." As a result, the content of the call and the emotions felt at the time can be understood at a glance.

[1174] Step 7:

[1175] The server stores the summarized text data and emotion data in a database and associates it with the user's call history. The input is the summarized text and emotion data, and the output is data associated with the call history. Specifically, the summary data and emotion data are linked to each call history record. This process allows the necessary information to be managed quickly and efficiently.

[1176] Step 8:

[1177] When a user checks their call history, the device retrieves the relevant summary data and emotion data from the server and displays them on the screen. The input is a request for call history, and the output is a screen displaying the summary data and emotion data. For example, information is displayed in the format "Next meeting: Tuesday (emotion: happy)." As a result, the user can easily check the content of the call and the emotion at the time.

[1178] (Application example 2)

[1179] 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."

[1180] Autonomous vehicles require a means to efficiently record conversations and instructions between passengers and quickly obtain necessary information by summarizing it. However, conventional technology only records voice data and does not analyze or summarize emotional data, making it difficult to grasp the nuances and importance of the information. This makes it easy to overlook passenger intentions and important information, increasing the risk of problems occurring.

[1181] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1182] In this invention, the server includes means for recording the contents of the call, means for converting the recorded voice data into text data using a voice recognition engine, means for summarizing the text data using a generative model, means for saving and displaying the summarized text data together with emotion data added by an emotion recognition engine, and means for saving and displaying the data in association with the call history. This not only records the voice but also performs summarization and emotion analysis, making it possible to prevent important information from being overlooked and to quickly and accurately grasp the intentions and importance of passengers.

[1183] A "call recording device" is any device or software that is capable of adequately capturing and digitally recording the audio of a call.

[1184] "Means for converting recorded voice data into text using a voice recognition engine" means software or algorithms that use voice recognition technology to convert recorded voice files into text format.

[1185] "Means for summarizing text data using a generative model" refers to an algorithm that uses a generative AI model to extract key information from long text data and output it as a short summary.

[1186] The "means for saving and displaying summarized text data together with emotion data added by an emotion recognition engine" refers to a system for adding emotion information analyzed by an emotion recognition system to summarized text data, saving the data, and displaying it so that the user can easily access it.

[1187] "Means for storing and displaying data in association with call history" refers to technology for linking and associating data containing summaries and emotional information with call history, thereby picking up and displaying the content of past calls and emotions.

[1188] MODE FOR CARRYING OUT THE INVENTION

[1189] The system for carrying out the present invention mainly has the following configuration.

[1190] Recording and uploading audio

[1191] When a user starts a call, the device automatically starts recording audio. It uses the microphone to capture audio data and continues recording for the duration of the call. When the call ends, the recorded audio data is uploaded to the server.

[1192] Speech-to-text conversion

[1193] When the server receives the recorded voice data, it activates a speech recognition engine and converts the voice data into text data. In this case, the speech_recognition library is used. For example, if a user says, "Please make the next stop Shibuya Station," the speech is recorded as text saying, "Please make the next stop Shibuya Station."

[1194] Emotion recognition

[1195] The server inputs the converted text data into an emotion recognition engine to analyze the user's emotions. A custom module called EmotionRecognition is used for this emotion recognition. For example, the same utterance "Please make the next stop Shibuya Station" can identify the emotion "neutral."

[1196] Summarize text and add sentiment

[1197] The server then inputs the converted text data into a generative AI model to summarize it. This summarization is performed using a generative model such as T5 from the Transformers library. Emotion data recognized by an emotion recognition engine is then added to the summary. For example, the text "Please make the next stop Shibuya Station" is summarized as "Stops at Shibuya Station," and the emotion data "neutral" is added to make it "Stops at Shibuya Station (Emotion: Neutral)."

[1198] Save to database and link to call history

[1199] Once the summary and emotion data are created, the server stores them in a database and associates them with the call history, so that when a user checks their call history, the summary and emotion data are instantly displayed.

[1200] Displaying notes in call history

[1201] When a user checks their call history, the device retrieves summary data and emotion data associated with the call history and displays them on the screen. Specifically, it displays the data in the form of "Shibuya Station Stop (Emotion: Neutral)." This system allows users to easily understand the content of the call and the emotion expressed at the time.

[1202] Specific examples

[1203] Scenario: A passenger gives instructions on board, such as "Next stop: Shibuya Station."

[1204] Example prompt sentence:

[1205] "Next stop please be Shibuya Station."

[1206] Example output:

[1207] Shibuya Station (Emotion: Neutral)

[1208] This invention converts voice data into text and automatically displays summaries and memos with emotion recognition, enabling quick and accurate understanding of information. This is expected to reduce troubles in self-driving vehicles and improve the passenger experience.

[1209] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1210] Step 1:

[1211] When the device detects the start of a call, it automatically starts recording audio. Specifically, the device's microphone captures the audio data and continues recording until the call ends. The recorded audio data is saved in raw audio format. The input is the audio during the call, and the output is the recorded audio file.

[1212] Step 2:

[1213] When the call ends, the device uploads the recorded audio data to the server. At this time, the audio data is sent to the server via the network. The input is the recorded audio file, and the output is the audio data stored on the server.

[1214] Step 3:

[1215] The server inputs the received voice data into a voice recognition engine and converts it into text data. This process uses the speech_recognition library to convert the voice file into text format. The input is voice data and the output is text data.

[1216] Step 4:

[1217] The server inputs the converted text data into an emotion recognition engine to analyze the user's emotions. The emotion recognition engine uses the EmotionRecognition module to extract emotional information from the text data. The input is text data, and the output is emotional data.

[1218] Step 5:

[1219] The server inputs the text data into a generative AI model to generate a summary. In this case, it uses the T5 model from the Transformers library to extract key information from long text data and generate a short summary. The input is text data, and the output is a summary.

[1220] Step 6:

[1221] The server adds emotional data to the generated summary to create the final summary. The server then integrates and formats the generated summary and emotional data. The input is the summary and emotional data, and the output is summary data that includes emotional information.

[1222] Step 7:

[1223] The server stores this summary data in a database and associates it with the call history. A new entry is added to the database and linked to the call history. The input is the summary data including emotion information, and the output is a database entry associated with the call history.

[1224] Step 8:

[1225] When a user checks the call history, the device retrieves the associated summary data from the database and displays it on the screen. The device sends a query to the database and displays the retrieved summary data and emotion information on the screen. The input is the operation to reference the call history, and the output is the summary data displayed on the screen.

[1226] 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.

[1227] 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.

[1228] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1229] 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.

[1230] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1231] 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.

[1232] 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).

[1233] 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.

[1234] 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."

[1235] 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.

[1236] 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).

[1237] 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.

[1238] 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.

[1239] 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.

[1240] 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.

[1241] 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.

[1242] 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.

[1243] 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.

[1244] 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.

[1245] 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.

[1246] 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.

[1247] The following is further disclosed regarding the above embodiment.

[1248] (Claim 1)

[1249] a means for recording the contents of the call;

[1250] A means for converting the recorded voice data into text data using a voice recognition engine;

[1251] A means for summarizing text data using a generative model;

[1252] a means for storing and displaying the summarized text data in association with the call history;

[1253] A system including:

[1254] (Claim 2)

[1255] 10. The system of claim 1, wherein the system uses a generative model to summarize the converted text data.

[1256] (Claim 3)

[1257] 2. The system according to claim 1, wherein the recording of the contents of a call is automatically started and ended in conjunction with the start and end of the call.

[1258] "Example 1"

[1259] (Claim 1)

[1260] a means for recording the contents of the call;

[1261] A means for converting the recorded voice data into text data using a voice recognition system;

[1262] A means for summarizing text data using a generative AI model;

[1263] a means for storing and displaying the summarized text data in association with the communication history;

[1264] means for displaying a summary when a user reviews the communication history;

[1265] A system including:

[1266] (Claim 2)

[1267] 10. The system of claim 1, wherein the system uses a generative AI model to summarize the converted text data.

[1268] (Claim 3)

[1269] 2. The system according to claim 1, wherein recording of the contents of a call is automatically started and ended in conjunction with the start and end of communication.

[1270] "Application Example 1"

[1271] (Claim 1)

[1272] a means for recording the contents of the call;

[1273] A means for converting the recorded voice data into text data using a voice recognition engine;

[1274] A means for summarizing text data using a generative model;

[1275] a means for storing and displaying the summarized text data in association with the call history;

[1276] A means for uploading and storing the audio data and summary text in cloud storage;

[1277] A means for restoring and playing the uploaded audio data;

[1278] A system including:

[1279] (Claim 2)

[1280] 10. The system of claim 1, wherein the system uses a generative AI model to summarize the converted text data.

[1281] (Claim 3)

[1282] 2. The system according to claim 1, wherein the recording of the contents of a call is automatically started and ended in conjunction with the start and end of the call.

[1283] "Example 2: Combining Emotion Engines"

[1284] (Claim 1)

[1285] a means for recording the contents of the call;

[1286] A means for converting the recorded voice data into text data using a voice recognition engine;

[1287] A means for analyzing user emotions from text data using an emotion analysis engine;

[1288] A means for summarizing the converted text data using a generative model;

[1289] means for adding emotion data to the summarized text data;

[1290] a means for storing and displaying the summarized text data and emotion data in association with the call history;

[1291] A system including:

[1292] (Claim 2)

[1293] 10. The system of claim 1, wherein the system uses a generative model to summarize the converted text data.

[1294] (Claim 3)

[1295] 2. The system according to claim 1, wherein the recording of the contents of a call is automatically started and ended in conjunction with the start and end of the call.

[1296] "Application example 2 when combining emotion engines"

[1297] (Claim 1)

[1298] a means for recording the contents of the call;

[1299] A means for converting the recorded voice data into text data using a voice recognition engine;

[1300] A means for summarizing text data using a generative model;

[1301] a means for saving and displaying the summarized text data together with emotion data added by an emotion recognition engine;

[1302] means for storing and displaying data in association with call history;

[1303] A system including:

[1304] (Claim 2)

[1305] 10. The system of claim 1, wherein the system uses a generative model to summarize the converted text data.

[1306] (Claim 3)

[1307] 2. The system according to claim 1, wherein the recording of the contents of a call is automatically started and ended in conjunction with the start and end of the call. [Explanation of symbols]

[1308] 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 recording the contents of the call; A means for converting the recorded voice data into text data using a voice recognition engine; A means for summarizing text data using a generative model; a means for storing and displaying the summarized text data in association with the call history; A system including:

2. The system of claim 1 , wherein the system uses a generative model to summarize the transformed text data.

3. 2. The system according to claim 1, wherein the recording of the contents of a call is automatically started and ended in conjunction with the start and end of the call.

Images