System

Abstract

A system is provided.SOLUTION: A system comprising: means for recording a call; means for transmitting the recorded call data to a server; means for converting the call data to text at the server; means for summarizing the converted text; means for returning the summarized text to a terminal; and means for user confirmation of the returned text and summary.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In modern digital communication, emails and chats are text-based, making it easy to check the history of conversations. However, with telephone conversations, no records of the conversation are kept, and users must take their own notes, making it difficult to review the content of long or important conversations later. This problem not only hinders the organization of information in work and daily life, but also complicates the process of checking when a problem occurs. The objective of the present invention is to make it possible to easily check the content of telephone conversations, thereby helping to organize information and prevent problems. [Means for solving the problem]

[0005] To solve the above problems, the present invention provides the following means: a system including a means for recording telephone conversations, a means for transmitting the recorded call data to a server, a means for converting the call data to text on the server, a means for summarizing the converted text, a means for returning the summarized text to the terminal, and a means for the user to check the returned text and summary. This system automatically converts the contents of telephone conversations into text, allowing the user to easily check the necessary information. In addition, the summarized text and detailed call contents can be easily viewed from the call history screen, making it possible to organize information more efficiently and providing reliable evidence in the event of a problem.

[0006] A "call" is a voice communication conducted using a telephone.

[0007] "Recording" refers to saving a conversation or sound on a recording medium.

[0008] "Call data" refers to the audio data of recorded conversations.

[0009] A "server" is a computer system that provides services to multiple terminals over a network.

[0010] "Convert to text" refers to the process of converting audio data into written information.

[0011] A summary is a short summary of the main points and important information extracted from a longer text.

[0012] "Terminal" refers to a computer device, such as a smartphone or PC, that connects to a server and sends, receives, and processes call data.

[0013] "Return" refers to the act of sending data back from the server to the terminal.

[0014] "Confirmation" refers to the act of a user viewing the provided data or information and checking its contents. [Brief explanation of the drawings]

[0015] [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

[0016] 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.

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

[0018] 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).

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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."

[0023] [First embodiment]

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

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form 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.

[0030] 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.

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

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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."

[0036] The present invention provides a system that automatically converts the contents of a telephone conversation into text and summarizes it so that the user can easily check it. Specific embodiments of the system will be described below.

[0037] Overall system configuration

[0038] This system consists of a user's device, a server, and related software components. The user's device records and transmits the call data, while the server analyzes, converts, and summarizes the call data. The converted data and summary are then sent to the user's device, where the user can view them on the call history screen.

[0039] User enables call transcription option

[0040] The user operates the device's settings screen and enables the "call transcription option." This setting causes all subsequent calls to be recorded and analyzed.

[0041] The device records the call

[0042] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records the audio data in real time and temporarily stores it in an internal buffer. When the call ends, the recording data is saved.

[0043] The device sends the recording data to the server

[0044] After the call ends, the device compresses the recording and sends it over a secure channel to the server, where the transmission process includes encryption to protect the user's privacy.

[0045] The server receives and analyzes the recorded data.

[0046] Once the server receives the recording, it immediately stores it in a database. It then converts the recording into text using a speech recognition AI engine. This AI engine uses advanced speech recognition technology and can handle noise and speaker characteristics.

[0047] The server summarizes the converted text

[0048] The converted data is then analyzed by a generative AI model to create a summary. This model is able to quickly extract important information and key points and summarize them as a summary text. Specific keywords and context are also taken into account to improve the quality of the summary.

[0049] The server sends the summarized text back to the terminal

[0050] The generated text and summary are then re-encrypted and sent to the device over a secure channel, a process that allows users to view the call on their own device without worrying about information leaking to the outside world.

[0051] Users can view the transcript and summary on the call history screen.

[0052] Users can view the written text and summaries by operating the call history screen on their device. This screen displays a summary of each call and allows users to view detailed text if needed. This feature allows users to easily review long or important calls later.

[0053] Specific examples

[0054] Example: Business meeting call

[0055] 1. A user has a business meeting scheduled and enables the call transcription option.

[0056] 2. When the meeting starts, the device will automatically record the call.

[0057] 3. When the meeting ends, the device sends the recording data to the server.

[0058] 4. The server receives the recording data and begins analyzing it.

[0059] 5. A generative AI model converts the call content into text and then generates a summary.

[0060] 6. The server sends the generated text and summary to the terminal.

[0061] 7. After the meeting, the user opens the call history screen to review the key points and detailed discussions from the meeting.

[0062] In this way, the system of the present invention automatically converts telephone conversations into text and generates summaries, allowing users to effectively manage and review important information.

[0063] The processing flow will be explained below.

[0064] Step 1:

[0065] The user enables the call transcription option. Specifically, they open the device's settings screen, select "Call transcription options," and save the settings. This setting information is stored in the device's internal memory.

[0066] Step 2:

[0067] When a user makes or receives a call, the device detects the start of the call and automatically activates the recording function.

[0068] Step 3:

[0069] The device records the audio data during a call in real time and temporarily stores it in a buffer. The recorded data is encrypted for security reasons and temporarily stored.

[0070] Step 4:

[0071] The user ends the call. When the device detects the end of the call, it completes buffering the recorded data and begins preparing to save it as a file in a storage device.

[0072] Step 5:

[0073] The device compresses the recorded data and prepares it for transmission. The compression process is performed to ensure data quality.

[0074] Step 6:

[0075] The device establishes a secure communication channel with the server, encrypts the recording data, and transmits it to the server, with security protocols applied during data transfer.

[0076] Step 7:

[0077] The server receives the recording data and stores it securely in a database, where it is checked for integrity and prepared for further processing.

[0078] Step 8:

[0079] The server starts the speech recognition AI engine and begins analyzing the recorded data, which then starts the process of converting the voice data into text.

[0080] Step 9:

[0081] The server inputs the text data into a generative AI model to generate a summary, which extracts key information from the text and creates a summary.

[0082] Step 10:

[0083] The generated text and summaries are stored in a database on the server, and this storage process ensures that data inconsistencies do not occur.

[0084] Step 11:

[0085] The server encrypts the generated text and summary and sends it back to the device over a secure communication channel, ensuring that the returned data is transmitted securely.

[0086] Step 12:

[0087] The device receives and decrypts the data from the server, and the received data is stored in local storage.

[0088] Step 13:

[0089] The user opens the call history screen on their device and checks the text and summary of the call. The user can easily refer to the necessary information from the history screen.

[0090] Example 1

[0091] 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."

[0092] In recent years, there has been an increasing need to review the contents of business meetings and important phone calls later. However, with existing technology, it has been difficult to accurately and efficiently convert phone calls into text and summarize the key points. In particular, there have been challenges with the length of phone calls, the effects of noise, and the quality of the summaries of the converted text.

[0093] 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.

[0094] In this invention, the server includes means for converting call data into text using speech recognition technology, means for summarizing the converted text using a generative AI model, and means for returning the encrypted summary text to the terminal. This allows the user to efficiently check the call content by automatically and accurately converting the call content into text and summarizing important information.

[0095] The "call transcription option" refers to a setting function that records calls and converts the audio data into text for a summary.

[0096] "Means for recording" refers to the function of collecting audio during a call in real time and recording it as data.

[0097] "Means for compressing and transmitting via a secure communication channel" refers to the function of compressing recorded audio data using an algorithm that reduces the data size, encrypting it, and transmitting it using a communication protocol that prevents information leakage.

[0098] "Means of converting to text using speech recognition technology" refers to technology that analyzes recorded speech data and converts it into corresponding text data.

[0099] "Method of summarizing using a generative AI model" refers to artificial intelligence technology for extracting important information from large amounts of text data and generating concise summaries.

[0100] "Means for returning encrypted summary text to the terminal" refers to a technique for encrypting the generated text and summary to protect them and transmitting them securely to the user terminal.

[0101] The "call history screen" refers to the interface that the user uses to check the contents of past calls.

[0102] The present invention provides a system for automatically converting telephone conversations into text and summarizing them so that users can easily check them. Specific embodiments of the present invention will be described below.

[0103] Overall system configuration

[0104] This system consists of a user's device, a server, and related software components. The user's device records and transmits the call data, while the server analyzes, converts, and summarizes the call data. The converted data and summary are then sent to the user's device, where the user can view them on the call history screen.

[0105] User enables call transcription option

[0106] Users can activate the "call transcription option" by going to the device's settings screen. This setting ensures that all subsequent calls are recorded and analyzed, allowing users to easily review important calls later.

[0107] The device records the call

[0108] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records the audio data in real time and temporarily stores it in an internal buffer. When the call ends, the recording data is saved.

[0109] The device compresses and encrypts the recorded data and sends it to the server.

[0110] After the call ends, the device compresses the recording and encrypts it using an encryption algorithm such as AES, then transmits the recording to the server over a secure channel using the HTTPS protocol, thereby protecting the user's privacy.

[0111] The server receives and analyzes the recorded data.

[0112] When the server receives the recorded data, it first decrypts it. Then, it stores the received recorded data in a database (e.g., MySQL, MongoDB, etc.). Here, each call data is assigned a unique ID for management. The stored recorded data is then called into a speech recognition AI engine (e.g., a general cloud speech recognition service) to convert the voice data into text. At this time, a noise reduction algorithm is also applied to improve recognition accuracy.

[0113] The server summarizes the text using a generative AI model

[0114] The converted data is then fed into a generative AI model (e.g., a typical natural language generation model) to generate a summary. This process extracts the key parts of the text and incorporates specific keywords and contextual information to create a concise summary.

[0115] The server encrypts the generated text and summary and sends it to the device.

[0116] The generated text and summary are then encrypted again using an encryption algorithm (e.g., AES) and then sent to the device via a secure communication protocol (e.g., HTTPS). This process allows the user to view the call on their own device, but the information is not leaked to the outside world.

[0117] Users see the text and summary on the call history screen

[0118] Users can view the textual content and summary of calls by opening the call history screen on their device. This screen displays a summary of each call and allows users to view detailed text as needed, allowing users to effectively manage and check important information.

[0119] Specific examples

[0120] Example: Business meeting call

[0121] 1. A user has a business meeting scheduled, so they enable the "Call transcription option" in their device settings.

[0122] 2. When a meeting call begins, the device's recording function will automatically start and buffer the call audio in real time.

[0123] 3. After the meeting ends, the device compresses the recording data, encrypts it using AES, and then sends it to the server via HTTPS.

[0124] 4. The server receives the recording, decrypts it, and stores it in a database.

[0125] 5. The server uses voice recognition technology to convert the recording into text.

[0126] 6. The converted text is summarized using a generative AI model, which extracts key points.

[0127] 7. The generated text and summary are again AES encrypted and sent to the terminal via the HTTPS protocol.

[0128] 8. After the meeting, the user opens the call history screen on their device and checks the summary and detailed discussion content.

[0129] Example prompts to input to the generative AI model

[0130] "Please transcribe the following audio recording and write a summary: Audio recording: [audio file]"

[0131] The system automatically converts phone conversations into text and generates summaries, allowing users to effectively manage and review important information.

[0132] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0133] Step 1:

[0134] The user enables the call transcription option.

[0135] The user operates the device's settings screen to enable the "call transcription option." This setting causes all subsequent calls to be recorded and analyzed. The input is the user's operation, and the output is the setting change that enables the option.

[0136] Step 2:

[0137] Record calls on your device.

[0138] When a user starts a phone call, the device automatically starts the recording function. The input is the call start trigger, and the output is the recorded voice data. The recorded voice data is temporarily stored in a buffer.

[0139] Step 3:

[0140] The recorded data is compressed and encrypted and sent to the server.

[0141] Once the call is over, the device compresses the recording and applies AES encryption to it. The input is the recorded audio data, and the output is the compressed and encrypted audio data. This data is sent to the server using the HTTPS protocol.

[0142] Step 4:

[0143] The server receives the recording data and stores it in a database.

[0144] When the server receives the recorded data, it first decrypts the data. The input is the received compressed encrypted data, and the output is the decrypted audio data. Then, it stores the recorded data in the database.

[0145] Step 5:

[0146] The server converts the audio data into text.

[0147] The server inputs the stored voice data into a voice recognition AI engine and converts it into text data. The input is voice data and the output is text data. The voice recognition engine also runs a noise reduction algorithm to improve recognition accuracy.

[0148] Step 6:

[0149] The server summarizes the text using a generative AI model.

[0150] The server inputs the converted text into a generative AI model to extract important information and generate a summary. The input is text data, and the output is a summary text. The generated summary is constructed taking into account specific keywords and context, improving its quality.

[0151] Step 7:

[0152] The generated text and summary are encrypted and sent to the terminal.

[0153] The server encrypts the generated text and summary using AES and sends them to the terminal via HTTPS. The input is the generated text and summary, and the output is the encrypted text and summary.

[0154] Step 8:

[0155] The user sees the text and summary on the call history screen.

[0156] The user can open the call history screen on their device and check the text of the call and the summary. The input is the encrypted text and the decrypted data of the summary, and the output is the text and summary displayed on the call history screen. The user can also view the detailed text if necessary.

[0157] (Application example 1)

[0158] 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."

[0159] Conventional call recording systems simply convert recorded call data into text, and there is no way to determine whether the call content contains confidential information, making it impossible to reduce the risk of information leaks. Furthermore, they lacked a mechanism for notifying administrators of important information in real time, which required the time and effort of checking the information later. This created a need for improvements to enhance information security within companies.

[0160] 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.

[0161] In this invention, the server includes a means for marking the contents of a call that contain confidential or important information based on the results of text analysis, and a means for notifying an administrator when confidential information is detected, thereby enabling real-time monitoring of the contents of a call and immediate notification of important information.

[0162] "Call recording means" refers to devices and programs for automatically recording user-initiated telephone calls.

[0163] "Means for transmitting recorded call data to a server" refers to the communication means and encryption technology for securely transferring call data from a recording device to a server device.

[0164] "Means for converting call data to text on the server" refers to a server program and related technology for converting call data into text format using voice recognition technology.

[0165] "Means for summarizing converted text" refers to generative AI models and programs for analyzing text data, extracting key points, and generating concise summaries.

[0166] "Means for returning summarized text to the terminal" refers to a server program and communication means for securely transferring the generated summary text to the user's terminal.

[0167] "Means for the user to review the returned text and summary" refers to an interface and program that allows the user to view and review the returned text and summary on their own terminal.

[0168] "Means for marking call content containing confidential or sensitive information based on the results of text analysis" refers to programs and algorithms that analyze the generated text and identify and locate confidential or sensitive information contained therein.

[0169] "Means for notifying administrators when confidential information is detected" refers to programs and communication means for sending alerts and notifications to administrators in real time when confidential information is included in the content of a call.

[0170] The present invention relates to a system for converting telephone conversation content into text in real time and detecting and notifying confidential information. Specific embodiments of the system will be described below.

[0171] Overall system configuration

[0172] The system consists of a user's device, a server, and related software components. The user's device records calls and transmits the data, while the server analyzes the call data, converts it to text, generates summaries, and detects and notifies users of confidential information. The converted data, summaries, and detection notifications are then sent to the user's device, where they can view them on their call history screen.

[0173] Hardware and Software Configuration

[0174] Device: An application that runs on a smartphone or PC and has recording and data transmission functions.

[0175] Server: A cloud service such as AWS or Google Cloud that uses Google Cloud Speech-to-Text as a speech recognition engine, OpenAI's GPT-4 for text analysis and summary generation, and runs programs and algorithms for sensitive information detection and notification.

[0176] Communication Method: A secure encrypted communication channel is used to send and receive call data.

[0177] System Operation

[0178] 1. User enables the call transcription option:

[0179] The user operates the device's settings screen and enables the "call transcription option," which causes all subsequent calls to be recorded and analyzed.

[0180] 2. Your device will record the call:

[0181] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records the audio data in real time and temporarily stores it in an internal buffer. When the call ends, the recording data is saved.

[0182] 3. The device sends the recording data to the server:

[0183] After the call ends, the device compresses the recording and sends it over a secure channel to the server, where the transmission process includes encryption to protect the user's privacy.

[0184] 4. The server receives and analyzes the recording:

[0185] Once the server receives the recording, it immediately stores it in a database. It then converts the recording into text using a speech recognition AI engine. This AI engine uses advanced speech recognition technology and can handle noise and speaker characteristics.

[0186] 5. The server summarizes the converted text:

[0187] The converted data is then analyzed by a generative AI model to create a summary. This generative AI model is capable of quickly extracting important information and key points and summarizing them as a summary text.

[0188] 6. The server detects and notifies you of sensitive information:

[0189] The generated text and summary are simultaneously analyzed to detect whether it contains sensitive or important information, and if so, the sensitive information is marked based on specific keywords or context, and administrators are notified in real time.

[0190] 7. The server sends the summarized text and notification back to the device:

[0191] The generated text, summary, and confidential information notification are then re-encrypted and sent to the device over a secure channel. This process ensures that the user can view the call and notification on their own device, without the risk of information leaking to the outside world.

[0192] 8. User checks on call history screen:

[0193] Users can view written text and summaries, as well as confidential information notifications, by operating the call history screen on their device. This screen displays a summary of each call, and detailed text can be viewed as needed. This feature allows users to easily review long calls or important conversation content later. In addition, administrators can detect confidential information leaks in real time.

[0194] Specific examples

[0195] Example: Internal call monitoring

[0196] 1. A user installs a security call monitoring application to monitor internal company calls.

[0197] 2. Designated important calls are automatically recorded, encrypted, and sent to the server.

[0198] 3. On the server side, a voice recognition engine converts the call contents into text and analyzes whether it contains confidential information.

[0199] 4. The generative AI model creates a summary of the call data and presents it to the administrator through an interface.

[0200] 5. If necessary, monitor in real time and send an immediate alert if any unauthorized information leakage is detected.

[0201] Example prompt sentence:

[0202] "Please summarize the following call:\n\n{call content}\n\n"

[0203] In this way, the system of the present invention can improve information security within a company and effectively manage important information through automatic text conversion and summary generation of call content, as well as real-time detection and notification of confidential information.

[0204] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0205] Step 1:

[0206] The user enables the call transcription option. The user operates the device's settings screen and enables the "call transcription option." This setting causes all calls to be recorded and analyzed. The input is the user setting information, and the output is the enabled state of the recording function.

[0207] Step 2:

[0208] The device records the call. When the user starts a phone call, the recording function is automatically activated. During the call, the device records audio data in real time and temporarily stores it in an internal buffer. After the call ends, the recording data is saved. The input is the call audio data, and the output is the recorded audio file.

[0209] Step 3:

[0210] The device sends the recorded data to the server. After the call ends, the device compresses the recorded data and sends it to the server over a secure channel. This transmission process includes data encryption. The input is the recorded audio file, and the output is the encrypted data being transmitted.

[0211] Step 4:

[0212] The server receives and analyzes the recorded data. Once the server receives the recorded data, it stores it in a database. Next, it converts the recorded data into text using a speech recognition AI engine. This engine can handle noise and speaker characteristics. The input is encrypted voice data, and the output is text data.

[0213] Step 5:

[0214] The server summarizes the converted text. The converted data is analyzed by a generative AI model to extract important information and key points and create a summary. Prompt sentences are used in this process. The input is text data and the output is a summary text.

[0215] Step 6:

[0216] The server marks sensitive information based on the results of text analysis. The generated text and summary are analyzed to identify sensitive or important information based on specific keywords and context. The input is the summary text, and the output is the marked text.

[0217] Step 7:

[0218] The server notifies the administrator if sensitive information is detected. If sensitive information is detected, it sends alerts and notifications to the administrator in real time. The input is the marked text and the output is a notification to the administrator.

[0219] Step 8:

[0220] The server returns the summarized text and notification to the terminal. The generated text, summary, and notification of confidential information are sent to the terminal via a secure communication channel. The input is the marked text, summary, and notification, and the output is the text and summary displayed on the user's terminal.

[0221] Step 9:

[0222] The user checks the call history screen. The user can operate the call history screen on their device to check the written text, summary, and notification of confidential information. The input is the text, summary, and notification sent to the device, and the output is the user's confirmation result.

[0223] 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.

[0224] This invention combines a system that automatically converts the contents of a telephone conversation into text and summarizes it so that the user can easily check it, with an emotion engine that recognizes the user's emotions. Specific embodiments of this invention are described below.

[0225] Overall system configuration

[0226] This system consists of a user's device, a server, an emotion engine, and related software components. The user's device records calls and transmits the data, while the server analyzes the call data, converts it to text, recognizes emotions, and generates summaries. The converted data, emotion information, and summaries are then sent to the user's device, where the user can view them on the call history screen.

[0227] User enables call transcription option

[0228] The user operates the device's settings screen and enables the "call transcription option." This setting causes all subsequent calls to be recorded and analyzed. The user's emotion recognition function is also enabled at the same time.

[0229] The device records the call

[0230] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records audio data in real time and temporarily stores it in a buffer. The recording data is encrypted for security reasons and temporarily stored.

[0231] The device sends the recording data to the server

[0232] After the call ends, the device compresses the recording and sends it over a secure channel to the server, where the transmission process includes encryption to protect the user's privacy.

[0233] The server receives and analyzes the recorded data.

[0234] When the server receives the recorded data, it first stores it in a database. Then, it converts the recorded data into text using a speech recognition AI engine. This AI engine uses advanced speech recognition technology and can handle noise and speaker characteristics.

[0235] Emotion recognition by emotion engine

[0236] The server analyzes the recorded data and recognizes the user's emotions using an emotion engine, which analyzes the tone of the voice and the words chosen to express emotions such as joy, anger, sadness, and happiness.

[0237] The server summarizes the converted text

[0238] The text-converted data, including emotional information, is then fed into a generative AI model to create a summary. This generative AI model is capable of quickly extracting key information and key points and summarizing them as emotionally relevant summary text. Specific keywords and context are also taken into account to improve the quality of the summary.

[0239] The server sends the summarized text back to the terminal

[0240] The generated text, summary, and user emotion information are then encrypted again and sent to the device via a secure channel. This process allows users to view the call content on their own devices, eliminating the risk of information leaking to the outside world.

[0241] Users can view the transcript and summary on the call history screen.

[0242] Users can view the written text, summaries, and recognized emotional information by operating the call history screen on their device. This screen displays a summary of each call, and users can view detailed text as needed. Emotional information is also displayed visually, allowing users to understand the emotional flow of the conversation.

[0243] Specific examples

[0244] Example: Customer support call

[0245] 1. The user (customer support representative) enables the call transcription option and emotion recognition feature.

[0246] 2. When a support call begins, the device will automatically record the call and collect emotional information.

[0247] 3. When the call ends, the device sends the recording data and emotional information to the server.

[0248] 4. The server receives the recording data and begins analyzing it.

[0249] 5. Voice recognition AI converts the call content into text, and the emotion engine analyzes emotional information.

[0250] 6. The generative AI model summarizes the call content and adjusts the quality of the summary based on emotional information.

[0251] 7. The server sends the generated text, summary, and emotion information to the terminal.

[0252] 8. The support agent opens the call history screen to see the main points of the call, detailed discussions, and the customer's emotional state.

[0253] In this way, the system of the present invention automatically converts telephone conversations into text and generates summaries with emotional information, allowing users to effectively manage and review important information. Furthermore, summaries that incorporate emotional information also allow users to grasp the emotional flow of the conversation, making them extremely useful in scenarios such as customer service.

[0254] The processing flow will be explained below.

[0255] Step 1:

[0256] The user enables the call transcription option and emotion recognition function. Specifically, they open the device's settings screen, select "Call transcription option" and "Emotion recognition function," and save the settings. This setting information is stored in the device's internal memory.

[0257] Step 2:

[0258] When a user makes or receives a call, the device detects the start of the call and automatically activates the recording and emotion recognition functions.

[0259] Step 3:

[0260] The device records the voice data during a call in real time and temporarily stores it in a buffer. At the same time, the emotion recognition function analyzes the voice data and detects the user's emotions in real time.

[0261] Step 4:

[0262] The user ends the call. When the device detects the end of the call, it completes buffering the recorded data and begins preparing to save it as a file on a storage device. Emotional information is also saved at the same time.

[0263] Step 5:

[0264] The device compresses the recorded data and emotional information and prepares it for transmission. The data compression process is processed to ensure quality.

[0265] Step 6:

[0266] The device establishes a secure communication channel with the server, encrypts the recorded data and emotional information, and then transmits the encrypted data to the server, where a security protocol is applied.

[0267] Step 7:

[0268] The server receives the recordings and emotional information, stores them securely in a database, ensures the integrity of the received data, and prepares it for analysis.

[0269] Step 8:

[0270] The server starts the speech recognition AI engine and begins analyzing the recorded data, converting the audio data into text data.

[0271] Step 9:

[0272] The server inputs the text data into a generative AI model to generate a summary, which also takes emotional information into account. The AI ​​model extracts important information and key points and provides a view that responds to changes in emotional state.

[0273] Step 10:

[0274] The generated text, summary, and sentiment information are stored in a database on the server, and this storage process ensures that data inconsistencies do not occur.

[0275] Step 11:

[0276] The server encrypts the generated text, summary, and emotion information and sends it back to the device through a secure communication channel. The returned data is transmitted with security in mind.

[0277] Step 12:

[0278] The device receives and decrypts the data from the server, and the received data is stored in local storage.

[0279] Step 13:

[0280] The user can open the call history screen on their device and check the text and summary, as well as the recognized emotion information. The history screen allows them to easily refer to the necessary information and changes in emotion. This is useful for analyzing whether the problem has been gradually resolved or whether the customer is satisfied.

[0281] As a concrete example, during a customer support call, the user (support representative) enables the call transcription option and emotion recognition function. When the meeting begins, the call is recorded and emotion information is collected. After the call ends, the recording data and emotion information are sent to the server for analysis. The generated text, summary, and emotion information are returned to the terminal, and the support representative can check the call content and changes in the customer's emotion from the history screen. This process allows the representative to improve the quality of customer support.

[0282] Example 2

[0283] 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."

[0284] Efficiently recording phone call content and managing important information is a time-consuming and labor-intensive process. Furthermore, manual transcription and summarization of call content can lack accuracy and consistency. Furthermore, it is difficult to properly recognize emotional information during a call and use it to evaluate the conversation.

[0285] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0286] In this invention, the server includes means for utilizing a speech recognition engine to convert call data into text, means for utilizing an emotion recognition engine to analyze emotional information, and means for utilizing a generative AI model to summarize the converted text including the emotional information. This allows efficient text conversion of the call content, recognition of emotional information, and summarization to be performed automatically, enabling the user to quickly and accurately grasp important information.

[0287] A "means for recording a call" is a function or device that records the audio of a call in real time and saves it in a format that can be analyzed later.

[0288] The "means for transmitting recorded call data to a server" refers to a function or device for securely compressing recorded voice data and transmitting it to a server via an encrypted communication channel.

[0289] "Means utilizing a speech recognition engine at a server to convert call data to text" means a function or device that operates an engine with speech recognition technology used to convert recorded voice data into text format.

[0290] "Means for utilizing an emotion recognition engine to analyze emotion information" refers to a function or device that operates an engine used to recognize and analyze a user's emotion from the converted text data.

[0291] "Means utilizing a generative AI model to summarize transformed text containing emotional information" refers to a function or device that operates a generative model using artificial intelligence to summarize text data containing emotional information.

[0292] The "means for returning the summarized text to the terminal" is a function or device for securely encrypting the generated summarized text and transmitting it to the user's terminal.

[0293] The "means for the user to check the returned text and summary" is a function or device that allows the user to view and check the text data and summary received at the terminal.

[0294] This invention combines a system that automatically converts telephone conversation content into text and summarizes it for easy user confirmation with an emotion engine that recognizes the user's emotions. Specific embodiments of this invention are described below.

[0295] Overall system configuration

[0296] This system consists of a user's device, a server, an emotion recognition engine, and related software components. The user's device records calls and transmits the data, while the server analyzes the call data, converts it to text, recognizes emotions, and generates summaries. The converted data, emotion information, and summaries are then sent to the user's device, where the user can view them on the call history screen.

[0297] Hardware and Software Configuration

[0298] Device: This could be a mobile device such as a smartphone or tablet, which requires recording software and encryption capabilities.

[0299] Server: A high-performance cloud or on-premise server is preferred, including a database and an AI engine for analytics.

[0300] Speech recognition engine: Uses highly accurate speech recognition technology such as Google Cloud Speech-to-Text.

[0301] Emotion recognition engine: Can use IBM Watson Tone Analyzer, etc. to analyze emotions from voice and text.

[0302] Generative AI models: Use models capable of automatic summary generation and natural language processing, such as OpenAI GPT-4.

[0303] System Operation

[0304] The system operates as follows.

[0305] User enables call transcription option

[0306] The user operates the device's settings screen and enables the "call transcription option." This setting causes all subsequent calls to be recorded and analyzed. The user's emotion recognition function is also enabled at the same time.

[0307] The device records the call

[0308] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records audio data in real time and temporarily stores it in a buffer. For security reasons, the recording data is encrypted using AES-256 encryption before being temporarily stored.

[0309] The device sends the recording data to the server

[0310] After the call ends, the device compresses the recording and sends it to the server via a secure channel (SSL / TLS protocol). This transmission process includes encryption, ensuring user privacy.

[0311] The server receives and analyzes the recorded data.

[0312] The server receives the recording and first stores it in a database. It then converts the recording into text using a speech recognition engine such as Google Cloud Speech-to-Text. This engine has advanced speech recognition technology and can handle noise and speaker characteristics.

[0313] Emotion recognition by emotion engine

[0314] The server inputs the converted text data into an emotion recognition engine such as IBM Watson Tone Analyzer to analyze emotions. The emotion engine analyzes the text content, tone of voice, and selected words to infer emotions such as joy, anger, sadness, and happiness.

[0315] The server summarizes the converted text

[0316] The converted text and sentiment information are then fed into a generative AI model, such as OpenAI GPT-4, to create a summary. This model quickly extracts key information and key points, and generates a mature summary that also reflects sentiment. Specific keywords and context are also taken into account, resulting in a high-quality summary.

[0317] The server sends the summarized text back to the terminal

[0318] The generated text, summary, and emotion information are then encrypted again and sent to the device via a secure channel. This process allows the user to review the call contents on their device, minimizing the risk of information leakage.

[0319] Users can view the transcript and summary on the call history screen.

[0320] Users can view the written text, summary, and recognized emotion information on the call history screen of their device. This screen displays a summary of each call and allows users to view detailed text. Emotion information is also displayed visually, allowing users to understand the flow of emotions.

[0321] Specific examples

[0322] Customer Support Calls

[0323] 1. The user (customer support representative) enables the call transcription option and emotion recognition feature.

[0324] 2. When a support call begins, the device will automatically record the call and collect emotional information.

[0325] 3. When the call ends, the device sends the recording data and emotional information to the server.

[0326] 4. The server receives the recording data and begins analyzing it.

[0327] 5. A voice recognition AI (e.g., Google Cloud Speech-to-Text) converts the call content into text, and an emotion engine (e.g., IBM Watson Tone Analyzer) analyzes emotions.

[0328] 6. A generative AI model (e.g., OpenAI GPT-4) summarizes the call and adjusts the quality of the summary based on emotional information.

[0329] 7. The server sends the generated text, summary, and emotion information to the terminal.

[0330] 8. The support representative opens the call history screen and checks the main points of the call, detailed discussions, and the customer's emotional state.

[0331] Prompt Sentence Examples

[0332] "Please transcribe and summarize this call and include the user's sentiment."

[0333] This system automatically converts phone conversations into text and generates summaries with emotional information, allowing users to efficiently manage and check important information. Furthermore, summaries that take emotional information into account also allow users to grasp the emotional flow of the conversation, making them extremely useful in customer service and other situations.

[0334] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0335] Step 1:

[0336] The user enables the call transcription option. In this step, the user operates the device's settings screen and turns on the "call transcription option." This setting will record all future calls and also enable the user's emotion recognition function. The input is the user's operation, and the output is the confirmation of the setting change.

[0337] Step 2:

[0338] A user initiates a phone call. When the call begins, the device automatically activates the recording function. During the call, the device records audio data in real time and temporarily stores it in a buffer. The hardware used is the device's built-in microphone, and the software used is a recording application. For security reasons, the recorded data is encrypted using the AES-256 encryption method and temporarily stored. The input is the start of the call, and the output is the encrypted recorded data.

[0339] Step 3:

[0340] After the call ends, the device compresses the recording and sends it to the server using a secure communication protocol (SSL / TLS). During this transmission process, the compressed recording is further encrypted with AES-256. The input is the encrypted recording, and the output is the data sent to the server.

[0341] Step 4:

[0342] The server receives the recording and stores it in a database. It then converts the recording into text using a speech recognition engine (e.g., Google Cloud Speech-to-Text). This engine uses noise filtering and speech enhancement techniques to convert the recording into highly accurate text. The input is the recording, and the output is text.

[0343] Step 5:

[0344] The server sends the text data to an emotion recognition engine (e.g., IBM Watson Tone Analyzer) for emotion recognition. The emotion recognition engine analyzes the content of the text, the tone of the voice, and the selected words to infer emotions such as joy, anger, sadness, and happiness. The input is text data, and the output is text data with emotional information.

[0345] Step 6:

[0346] The server inputs text data containing emotional information into a generative AI model (e.g., OpenAI GPT-4) to create a summary. This generative AI model quickly extracts important information and key points and generates a summary that reflects emotional information. The input is text data with emotional information, and the output is a summary text.

[0347] Step 7:

[0348] The generated text, summary, and sentiment information are again encrypted with AES-256 and sent to the device via a secure communication protocol, with the summary text and sentiment information as input and the encrypted data as output.

[0349] Step 8:

[0350] Users can view written text, summaries, and emotion information on the call history screen on their device. This screen displays a summary of each call and allows users to view detailed text. Emotion information is displayed visually, allowing users to understand the emotional flow of the conversation. The input is data returned from the server, and the output is text and emotion information displayed on the user's device.

[0351] (Application example 2)

[0352] 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."

[0353] In modern security services, it is difficult to quickly and accurately grasp the content of customer telephone conversations and extract and manage important information. Furthermore, if emotional information during a call could be grasped, it would be possible to respond based on the customer's psychological state, but this is also difficult to do using conventional means. The present invention aims to provide a system that automatically converts the content of a call into text, summarizes it, analyzes emotional information, and utilizes this information in customer service.

[0354] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for converting call data into text, means for analyzing the call data using an emotion engine and generating emotion information, and means for summarizing the text using the generated emotion information. This automatically converts the call content into text and summarizes it, and also analyzes the customer's emotion information, making it possible to quickly and accurately grasp important information in customer service and to respond appropriately based on the customer's psychological state.

[0355] "Call Data" refers to audio information recorded during a telephone call.

[0356] "Text conversion means" refers to software or hardware that performs the process of converting audio data into written information.

[0357] "Server" refers to a computer system that performs processes such as receiving call data, analyzing it, converting it into text, generating summaries, and collecting and transmitting emotional information.

[0358] An "emotion engine" refers to software that has algorithms or functions for analyzing voice and text data to estimate the speaker's emotional state.

[0359] "Summarization tool" refers to software or algorithms that extract key information from long text data and provide it in a shortened form.

[0360] "Means for returning" refers to the communication protocol or software for retransmitting data processed by the server to the user's terminal.

[0361] "Means for user confirmation" refers to an interface or software that allows a user to view the text, summary, and emotional information of the call using a terminal.

[0362] "Transcription options" refers to the settings that allow a user to enable a feature on the device that automatically transcribes phone conversations into text.

[0363] The "call history screen" refers to the display screen of the terminal on which the user can check the contents of past calls, their summaries, and emotional information.

[0364] "Generative AI models" refer to algorithms or software equipped with artificial intelligence technology used to summarize text data and analyze emotional information.

[0365] The system of the present invention is for smooth customer service in the field of security services. The system includes a means for recording calls, a means for transmitting the recorded data to a server, a means for analyzing the data on the server, a means for converting the data into text and generating a summary, and a means for analyzing emotional information. This makes it possible to improve the quality of customer service.

[0366] Hardware and software used

[0367] Terminal

[0368] This is a device such as a smartphone or smart glasses that users use when making phone calls. The device automatically records the call and temporarily stores the data. The recorded data is encrypted for security reasons.

[0369] server

[0370] The server is a computer system that receives and analyzes the recorded data. The software used includes a speech recognition AI engine, an emotion recognition engine, and a generative AI model.

[0371] Data processing and calculation procedures

[0372] 1. Record and send calls

[0373] When a user starts a phone call, the device automatically starts recording and temporarily saves the call. After the call ends, the recording data is sent to the server via a secure communication channel.

[0374] 2. Converting audio data into text

[0375] Once the server receives the recording, it uses a speech recognition AI engine (for example, Google Speech-to-Text API) to convert the audio data into text, which is then passed on to the next analysis step.

[0376] 3. Emotion Analysis

[0377] The text data is analyzed by an emotion engine to estimate the emotional state of the customer and the operator. Software used includes EmotionRecognition. This information is stored along with the text data.

[0378] 4. Summary Generation

[0379] It uses generative AI models (for example, Hugging Face's T5 model) to extract key information from long text data and provide it in a shortened form, which also reflects emotional information.

[0380] 5. Return and confirmation of results

[0381] The generated text, summary, and emotion information are encrypted and sent to the terminal via a secure communication channel. The user can then view this information on the call history screen.

[0382] Specific examples

[0383] For example, this system works in the background when a security company operator is talking to a customer. At the end of the call, the operator can view a summary of the call and sentiment information within the application, and if the customer is feeling anxious, they can quickly provide additional support.

[0384] Prompt Sentence Examples

[0385] An example of a prompt is as follows:

[0386] Please summarize the following Japanese conversation text and detect any emotions present:

[0387] <conversation_text>

[0388] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0389] Step 1:

[0390] The user enables the call transcription option and emotion recognition feature, which automatically records calls and collects emotional information. At this point, the device is ready to record and transmit data from the next call.

[0391] Step 2:

[0392] The device records the call. When a user initiates a phone call, the device automatically starts recording. During the call, audio data is recorded in real time and the recording data is temporarily stored in a buffer. This recording data is encrypted for security purposes.

[0393] Step 3:

[0394] The device sends the recording to the server. After the call ends, the recording is compressed and sent to the server over a secure channel. The data is re-encrypted during this process, protecting the user's privacy.

[0395] Step 4:

[0396] The server receives and analyzes the recorded data. When the server receives the recorded data, it first stores it in a database. Next, it converts the recorded data into text using a speech recognition AI engine. One example of the software used is the Google Speech-to-Text API. The input at this stage is the recorded data, and the output is text information.

[0397] Step 5:

[0398] The server analyzes the emotions in the voice data. An emotion engine is used to analyze the text data and generate emotional information for the customer and the operator. The software used is EmotionRecognition. In this process, the text data is the input and the output is emotional information.

[0399] Step 6:

[0400] The server summarizes the converted text. The data converted into text is then summarized by a generative AI model. The software used is the Hugging Face T5 model. Specifically, the text data and generated emotion information are input, the necessary information is extracted, and a summary of one or several sentences is output.

[0401] Step 7:

[0402] The server sends the summarized text and emotion information to the terminal. The generated text and emotion information are encrypted again and sent to the terminal via a secure communication channel. Here, the input is the summarized text and emotion information, and the output is the data sent back to the terminal.

[0403] Step 8:

[0404] The user checks the text, summary, and emotional information on the call history screen. The user opens the call history screen on the device and views the returned text, summary, and emotional information. This allows the user to grasp the main points of the call and the emotional flow of the conversation.

[0405] 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.

[0406] 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.

[0407] 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.

[0408] [Second embodiment]

[0409] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0410] 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.

[0411] 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).

[0412] 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.

[0413] 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.

[0414] 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).

[0415] 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.

[0416] 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.

[0417] 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.

[0418] 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.

[0419] In the smart glasses 214, 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.

[0420] 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."

[0421] The present invention provides a system that automatically converts the contents of a telephone conversation into text and summarizes it so that the user can easily check it. Specific embodiments of the system will be described below.

[0422] Overall system configuration

[0423] This system consists of a user's device, a server, and related software components. The user's device records and transmits the call data, while the server analyzes, converts, and summarizes the call data. The converted data and summary are then sent to the user's device, where the user can view them on the call history screen.

[0424] User enables call transcription option

[0425] The user operates the device's settings screen and enables the "call transcription option." This setting causes all subsequent calls to be recorded and analyzed.

[0426] The device records the call

[0427] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records the audio data in real time and temporarily stores it in an internal buffer. When the call ends, the recording data is saved.

[0428] The device sends the recording data to the server

[0429] After the call ends, the device compresses the recording and sends it over a secure channel to the server, where the transmission process includes encryption to protect the user's privacy.

[0430] The server receives and analyzes the recorded data.

[0431] Once the server receives the recording, it immediately stores it in a database. It then converts the recording into text using a speech recognition AI engine. This AI engine uses advanced speech recognition technology and can handle noise and speaker characteristics.

[0432] The server summarizes the converted text

[0433] The converted data is then analyzed by a generative AI model to create a summary. This model is able to quickly extract important information and key points and summarize them as a summary text. Specific keywords and context are also taken into account to improve the quality of the summary.

[0434] The server sends the summarized text back to the terminal

[0435] The generated text and summary are then re-encrypted and sent to the device over a secure channel, a process that allows users to view the call on their own device without worrying about information leaking to the outside world.

[0436] Users can view the transcript and summary on the call history screen.

[0437] Users can view the written text and summaries by operating the call history screen on their device. This screen displays a summary of each call and allows users to view detailed text if needed. This feature allows users to easily review long or important calls later.

[0438] Specific examples

[0439] Example: Business meeting call

[0440] 1. A user has a business meeting scheduled and enables the call transcription option.

[0441] 2. When the meeting starts, the device will automatically record the call.

[0442] 3. When the meeting ends, the device sends the recording data to the server.

[0443] 4. The server receives the recording data and begins analyzing it.

[0444] 5. A generative AI model converts the call content into text and then generates a summary.

[0445] 6. The server sends the generated text and summary to the terminal.

[0446] 7. After the meeting, the user opens the call history screen to review the key points and detailed discussions from the meeting.

[0447] In this way, the system of the present invention automatically converts telephone conversations into text and generates summaries, allowing users to effectively manage and review important information.

[0448] The processing flow will be explained below.

[0449] Step 1:

[0450] The user enables the call transcription option. Specifically, they open the device's settings screen, select "Call transcription options," and save the settings. This setting information is stored in the device's internal memory.

[0451] Step 2:

[0452] When a user makes or receives a call, the device detects the start of the call and automatically activates the recording function.

[0453] Step 3:

[0454] The device records the audio data during a call in real time and temporarily stores it in a buffer. The recorded data is encrypted for security reasons and temporarily stored.

[0455] Step 4:

[0456] The user ends the call. When the device detects the end of the call, it completes buffering the recorded data and begins preparing to save it as a file in a storage device.

[0457] Step 5:

[0458] The device compresses the recorded data and prepares it for transmission. The compression process is performed to ensure data quality.

[0459] Step 6:

[0460] The device establishes a secure communication channel with the server, encrypts the recording data, and transmits it to the server, with security protocols applied during data transfer.

[0461] Step 7:

[0462] The server receives the recording data and stores it securely in a database, where it is checked for integrity and prepared for further processing.

[0463] Step 8:

[0464] The server starts the speech recognition AI engine and begins analyzing the recorded data, which then starts the process of converting the voice data into text.

[0465] Step 9:

[0466] The server inputs the text data into a generative AI model to generate a summary, which extracts key information from the text and creates a summary.

[0467] Step 10:

[0468] The generated text and summaries are stored in a database on the server, and this storage process ensures that data inconsistencies do not occur.

[0469] Step 11:

[0470] The server encrypts the generated text and summary and sends it back to the device over a secure communication channel, ensuring that the returned data is transmitted securely.

[0471] Step 12:

[0472] The device receives and decrypts the data from the server, and the received data is stored in local storage.

[0473] Step 13:

[0474] The user opens the call history screen on their device and checks the text and summary of the call. The user can easily refer to the necessary information from the history screen.

[0475] Example 1

[0476] 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."

[0477] In recent years, there has been an increasing need to review the contents of business meetings and important phone calls later. However, with existing technology, it has been difficult to accurately and efficiently convert phone calls into text and summarize the key points. In particular, there have been challenges with the length of phone calls, the effects of noise, and the quality of the summaries of the converted text.

[0478] 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.

[0479] In this invention, the server includes means for converting call data into text using speech recognition technology, means for summarizing the converted text using a generative AI model, and means for returning the encrypted summary text to the terminal. This allows the user to efficiently check the call content by automatically and accurately converting the call content into text and summarizing important information.

[0480] The "call transcription option" refers to a setting function that records calls and converts the audio data into text for a summary.

[0481] "Means for recording" refers to the function of collecting audio during a call in real time and recording it as data.

[0482] "Means for compressing and transmitting via a secure communication channel" refers to the function of compressing recorded audio data using an algorithm that reduces the data size, encrypting it, and transmitting it using a communication protocol that prevents information leakage.

[0483] "Means of converting to text using speech recognition technology" refers to technology that analyzes recorded speech data and converts it into corresponding text data.

[0484] "Method of summarizing using a generative AI model" refers to artificial intelligence technology for extracting important information from large amounts of text data and generating concise summaries.

[0485] "Means for returning encrypted summary text to the terminal" refers to a technique for encrypting the generated text and summary to protect them and transmitting them securely to the user terminal.

[0486] The "call history screen" refers to the interface that the user uses to check the contents of past calls.

[0487] The present invention provides a system for automatically converting telephone conversations into text and summarizing them so that users can easily check them. Specific embodiments of the present invention will be described below.

[0488] Overall system configuration

[0489] This system consists of a user's device, a server, and related software components. The user's device records and transmits the call data, while the server analyzes, converts, and summarizes the call data. The converted data and summary are then sent to the user's device, where the user can view them on the call history screen.

[0490] User enables call transcription option

[0491] Users can activate the "call transcription option" by going to the device's settings screen. This setting ensures that all subsequent calls are recorded and analyzed, allowing users to easily review important calls later.

[0492] The device records the call

[0493] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records the audio data in real time and temporarily stores it in an internal buffer. When the call ends, the recording data is saved.

[0494] The device compresses and encrypts the recorded data and sends it to the server.

[0495] After the call ends, the device compresses the recording and encrypts it using an encryption algorithm such as AES, then transmits the recording to the server over a secure channel using the HTTPS protocol, thereby protecting the user's privacy.

[0496] The server receives and analyzes the recorded data.

[0497] When the server receives the recorded data, it first decrypts it. Then, it stores the received recorded data in a database (e.g., MySQL, MongoDB, etc.). Here, each call data is assigned a unique ID for management. The stored recorded data is then called into a speech recognition AI engine (e.g., a general cloud speech recognition service) to convert the voice data into text. At this time, a noise reduction algorithm is also applied to improve recognition accuracy.

[0498] The server summarizes the text using a generative AI model

[0499] The converted data is then fed into a generative AI model (e.g., a typical natural language generation model) to generate a summary. This process extracts the key parts of the text and incorporates specific keywords and contextual information to create a concise summary.

[0500] The server encrypts the generated text and summary and sends it to the device.

[0501] The generated text and summary are then encrypted again using an encryption algorithm (e.g., AES) and then sent to the device via a secure communication protocol (e.g., HTTPS). This process allows the user to view the call on their own device, but the information is not leaked to the outside world.

[0502] Users see the text and summary on the call history screen

[0503] Users can view the textual content and summary of calls by opening the call history screen on their device. This screen displays a summary of each call and allows users to view detailed text as needed, allowing users to effectively manage and check important information.

[0504] Specific examples

[0505] Example: Business meeting call

[0506] 1. A user has a business meeting scheduled, so they enable the "Call transcription option" in their device settings.

[0507] 2. When a meeting call begins, the device's recording function will automatically start and buffer the call audio in real time.

[0508] 3. After the meeting ends, the device compresses the recording data, encrypts it using AES, and then sends it to the server via HTTPS.

[0509] 4. The server receives the recording, decrypts it, and stores it in a database.

[0510] 5. The server uses voice recognition technology to convert the recording into text.

[0511] 6. The converted text is summarized using a generative AI model, which extracts key points.

[0512] 7. The generated text and summary are again AES encrypted and sent to the terminal via the HTTPS protocol.

[0513] 8. After the meeting, the user opens the call history screen on their device and checks the summary and detailed discussion content.

[0514] Example prompts to input to the generative AI model

[0515] "Please transcribe the following audio recording and write a summary: Audio recording: [audio file]"

[0516] The system automatically converts phone conversations into text and generates summaries, allowing users to effectively manage and review important information.

[0517] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0518] Step 1:

[0519] The user enables the call transcription option.

[0520] The user operates the device's settings screen to enable the "call transcription option." This setting causes all subsequent calls to be recorded and analyzed. The input is the user's operation, and the output is the setting change that enables the option.

[0521] Step 2:

[0522] Record calls on your device.

[0523] When a user starts a phone call, the device automatically starts the recording function. The input is the call start trigger, and the output is the recorded voice data. The recorded voice data is temporarily stored in a buffer.

[0524] Step 3:

[0525] The recorded data is compressed and encrypted and sent to the server.

[0526] Once the call is over, the device compresses the recording and applies AES encryption to it. The input is the recorded audio data, and the output is the compressed and encrypted audio data. This data is sent to the server using the HTTPS protocol.

[0527] Step 4:

[0528] The server receives the recording data and stores it in a database.

[0529] When the server receives the recorded data, it first decrypts the data. The input is the received compressed encrypted data, and the output is the decrypted audio data. Then, it stores the recorded data in the database.

[0530] Step 5:

[0531] The server converts the audio data into text.

[0532] The server inputs the stored voice data into a voice recognition AI engine and converts it into text data. The input is voice data and the output is text data. The voice recognition engine also runs a noise reduction algorithm to improve recognition accuracy.

[0533] Step 6:

[0534] The server summarizes the text using a generative AI model.

[0535] The server inputs the converted text into a generative AI model to extract important information and generate a summary. The input is text data, and the output is a summary text. The generated summary is constructed taking into account specific keywords and context, improving its quality.

[0536] Step 7:

[0537] The generated text and summary are encrypted and sent to the terminal.

[0538] The server encrypts the generated text and summary using AES and sends them to the terminal via HTTPS. The input is the generated text and summary, and the output is the encrypted text and summary.

[0539] Step 8:

[0540] The user sees the text and summary on the call history screen.

[0541] The user can open the call history screen on their device and check the text of the call and the summary. The input is the encrypted text and the decrypted data of the summary, and the output is the text and summary displayed on the call history screen. The user can also view the detailed text if necessary.

[0542] (Application example 1)

[0543] 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."

[0544] Conventional call recording systems simply convert recorded call data into text, and there is no way to determine whether the call content contains confidential information, making it impossible to reduce the risk of information leaks. Furthermore, they lacked a mechanism for notifying administrators of important information in real time, which required the time and effort of checking the information later. This created a need for improvements to enhance information security within companies.

[0545] 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.

[0546] In this invention, the server includes a means for marking the contents of a call that contain confidential or important information based on the results of text analysis, and a means for notifying an administrator when confidential information is detected, thereby enabling real-time monitoring of the contents of a call and immediate notification of important information.

[0547] "Call recording means" refers to devices and programs for automatically recording user-initiated telephone calls.

[0548] "Means for transmitting recorded call data to a server" refers to the communication means and encryption technology for securely transferring call data from a recording device to a server device.

[0549] "Means for converting call data to text on the server" refers to a server program and related technology for converting call data into text format using voice recognition technology.

[0550] "Means for summarizing converted text" refers to generative AI models and programs for analyzing text data, extracting key points, and generating concise summaries.

[0551] "Means for returning summarized text to the terminal" refers to a server program and communication means for securely transferring the generated summary text to the user's terminal.

[0552] "Means for the user to review the returned text and summary" refers to an interface and program that allows the user to view and review the returned text and summary on their own terminal.

[0553] "Means for marking call content containing confidential or sensitive information based on the results of text analysis" refers to programs and algorithms that analyze the generated text and identify and locate confidential or sensitive information contained therein.

[0554] "Means for notifying administrators when confidential information is detected" refers to programs and communication means for sending alerts and notifications to administrators in real time when confidential information is included in the content of a call.

[0555] The present invention relates to a system for converting telephone conversation content into text in real time and detecting and notifying confidential information. Specific embodiments of the system will be described below.

[0556] Overall system configuration

[0557] The system consists of a user's device, a server, and related software components. The user's device records calls and transmits the data, while the server analyzes the call data, converts it to text, generates summaries, and detects and notifies users of confidential information. The converted data, summaries, and detection notifications are then sent to the user's device, where they can view them on their call history screen.

[0558] Hardware and Software Configuration

[0559] Device: An application that runs on a smartphone or PC and has recording and data transmission functions.

[0560] Server: A cloud service such as AWS or Google Cloud that uses Google Cloud Speech-to-Text as a speech recognition engine, OpenAI's GPT-4 for text analysis and summary generation, and runs programs and algorithms for sensitive information detection and notification.

[0561] Communication Method: A secure encrypted communication channel is used to send and receive call data.

[0562] System Operation

[0563] 1. User enables the call transcription option:

[0564] The user operates the device's settings screen and enables the "call transcription option," which causes all subsequent calls to be recorded and analyzed.

[0565] 2. Your device will record the call:

[0566] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records the audio data in real time and temporarily stores it in an internal buffer. When the call ends, the recording data is saved.

[0567] 3. The device sends the recording data to the server:

[0568] After the call ends, the device compresses the recording and sends it over a secure channel to the server, where the transmission process includes encryption to protect the user's privacy.

[0569] 4. The server receives and analyzes the recording:

[0570] Once the server receives the recording, it immediately stores it in a database. It then converts the recording into text using a speech recognition AI engine. This AI engine uses advanced speech recognition technology and can handle noise and speaker characteristics.

[0571] 5. The server summarizes the converted text:

[0572] The converted data is then analyzed by a generative AI model to create a summary. This generative AI model is capable of quickly extracting important information and key points and summarizing them as a summary text.

[0573] 6. The server detects and notifies you of sensitive information:

[0574] The generated text and summary are simultaneously analyzed to detect whether it contains sensitive or important information, and if so, the sensitive information is marked based on specific keywords or context, and administrators are notified in real time.

[0575] 7. The server sends the summarized text and notification back to the device:

[0576] The generated text, summary, and confidential information notification are then re-encrypted and sent to the device over a secure channel. This process ensures that the user can view the call and notification on their own device, without the risk of information leaking to the outside world.

[0577] 8. User checks on call history screen:

[0578] Users can view written text and summaries, as well as confidential information notifications, by operating the call history screen on their device. This screen displays a summary of each call, and detailed text can be viewed as needed. This feature allows users to easily review long calls or important conversation content later. In addition, administrators can detect confidential information leaks in real time.

[0579] Specific examples

[0580] Example: Internal call monitoring

[0581] 1. A user installs a security call monitoring application to monitor internal company calls.

[0582] 2. Designated important calls are automatically recorded, encrypted, and sent to the server.

[0583] 3. On the server side, a voice recognition engine converts the call contents into text and analyzes whether it contains confidential information.

[0584] 4. The generative AI model creates a summary of the call data and presents it to the administrator through an interface.

[0585] 5. If necessary, monitor in real time and send an immediate alert if any unauthorized information leakage is detected.

[0586] Example prompt sentence:

[0587] "Please summarize the following call:\n\n{call content}\n\n"

[0588] In this way, the system of the present invention can improve information security within a company and effectively manage important information through automatic text conversion and summary generation of call content, as well as real-time detection and notification of confidential information.

[0589] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0590] Step 1:

[0591] The user enables the call transcription option. The user operates the device's settings screen and enables the "call transcription option." This setting causes all calls to be recorded and analyzed. The input is the user setting information, and the output is the enabled state of the recording function.

[0592] Step 2:

[0593] The device records the call. When the user starts a phone call, the recording function is automatically activated. During the call, the device records audio data in real time and temporarily stores it in an internal buffer. After the call ends, the recording data is saved. The input is the call audio data, and the output is the recorded audio file.

[0594] Step 3:

[0595] The device sends the recorded data to the server. After the call ends, the device compresses the recorded data and sends it to the server over a secure channel. This transmission process includes data encryption. The input is the recorded audio file, and the output is the encrypted data being transmitted.

[0596] Step 4:

[0597] The server receives and analyzes the recorded data. Once the server receives the recorded data, it stores it in a database. Next, it converts the recorded data into text using a speech recognition AI engine. This engine can handle noise and speaker characteristics. The input is encrypted voice data, and the output is text data.

[0598] Step 5:

[0599] The server summarizes the converted text. The converted data is analyzed by a generative AI model to extract important information and key points and create a summary. Prompt sentences are used in this process. The input is text data and the output is a summary text.

[0600] Step 6:

[0601] The server marks sensitive information based on the results of text analysis. The generated text and summary are analyzed to identify sensitive or important information based on specific keywords and context. The input is the summary text, and the output is the marked text.

[0602] Step 7:

[0603] The server notifies the administrator if sensitive information is detected. If sensitive information is detected, it sends alerts and notifications to the administrator in real time. The input is the marked text and the output is a notification to the administrator.

[0604] Step 8:

[0605] The server returns the summarized text and notification to the terminal. The generated text, summary, and notification of confidential information are sent to the terminal via a secure communication channel. The input is the marked text, summary, and notification, and the output is the text and summary displayed on the user's terminal.

[0606] Step 9:

[0607] The user checks the call history screen. The user can operate the call history screen on their device to check the written text, summary, and notification of confidential information. The input is the text, summary, and notification sent to the device, and the output is the user's confirmation result.

[0608] 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.

[0609] This invention combines a system that automatically converts the contents of a telephone conversation into text and summarizes it so that the user can easily check it, with an emotion engine that recognizes the user's emotions. Specific embodiments of this invention are described below.

[0610] Overall system configuration

[0611] This system consists of a user's device, a server, an emotion engine, and related software components. The user's device records calls and transmits the data, while the server analyzes the call data, converts it to text, recognizes emotions, and generates summaries. The converted data, emotion information, and summaries are then sent to the user's device, where the user can view them on the call history screen.

[0612] User enables call transcription option

[0613] The user operates the device's settings screen and enables the "call transcription option." This setting causes all subsequent calls to be recorded and analyzed. The user's emotion recognition function is also enabled at the same time.

[0614] The device records the call

[0615] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records audio data in real time and temporarily stores it in a buffer. The recording data is encrypted for security reasons and temporarily stored.

[0616] The device sends the recording data to the server

[0617] After the call ends, the device compresses the recording and sends it over a secure channel to the server, where the transmission process includes encryption to protect the user's privacy.

[0618] The server receives and analyzes the recorded data.

[0619] When the server receives the recorded data, it first stores it in a database. Then, it converts the recorded data into text using a speech recognition AI engine. This AI engine uses advanced speech recognition technology and can handle noise and speaker characteristics.

[0620] Emotion recognition by emotion engine

[0621] The server analyzes the recorded data and recognizes the user's emotions using an emotion engine, which analyzes the tone of the voice and the words chosen to express emotions such as joy, anger, sadness, and happiness.

[0622] The server summarizes the converted text

[0623] The text-converted data, including emotional information, is then fed into a generative AI model to create a summary. This generative AI model is capable of quickly extracting key information and key points and summarizing them as emotionally relevant summary text. Specific keywords and context are also taken into account to improve the quality of the summary.

[0624] The server sends the summarized text back to the terminal

[0625] The generated text, summary, and user emotion information are then encrypted again and sent to the device via a secure channel. This process allows users to view the call content on their own devices, eliminating the risk of information leaking to the outside world.

[0626] Users can view the transcript and summary on the call history screen.

[0627] Users can view the written text, summaries, and recognized emotional information by operating the call history screen on their device. This screen displays a summary of each call, and users can view detailed text as needed. Emotional information is also displayed visually, allowing users to understand the emotional flow of the conversation.

[0628] Specific examples

[0629] Example: Customer support call

[0630] 1. The user (customer support representative) enables the call transcription option and emotion recognition feature.

[0631] 2. When a support call begins, the device will automatically record the call and collect emotional information.

[0632] 3. When the call ends, the device sends the recording data and emotional information to the server.

[0633] 4. The server receives the recording data and begins analyzing it.

[0634] 5. Voice recognition AI converts the call content into text, and the emotion engine analyzes emotional information.

[0635] 6. The generative AI model summarizes the call content and adjusts the quality of the summary based on emotional information.

[0636] 7. The server sends the generated text, summary, and emotion information to the terminal.

[0637] 8. The support agent opens the call history screen to see the main points of the call, detailed discussions, and the customer's emotional state.

[0638] In this way, the system of the present invention automatically converts telephone conversations into text and generates summaries with emotional information, allowing users to effectively manage and review important information. Furthermore, summaries that incorporate emotional information also allow users to grasp the emotional flow of the conversation, making them extremely useful in scenarios such as customer service.

[0639] The processing flow will be explained below.

[0640] Step 1:

[0641] The user enables the call transcription option and emotion recognition function. Specifically, they open the device's settings screen, select "Call transcription option" and "Emotion recognition function," and save the settings. This setting information is stored in the device's internal memory.

[0642] Step 2:

[0643] When a user makes or receives a call, the device detects the start of the call and automatically activates the recording and emotion recognition functions.

[0644] Step 3:

[0645] The device records the voice data during a call in real time and temporarily stores it in a buffer. At the same time, the emotion recognition function analyzes the voice data and detects the user's emotions in real time.

[0646] Step 4:

[0647] The user ends the call. When the device detects the end of the call, it completes buffering the recorded data and begins preparing to save it as a file on a storage device. Emotional information is also saved at the same time.

[0648] Step 5:

[0649] The device compresses the recorded data and emotional information and prepares it for transmission. The data compression process is processed to ensure quality.

[0650] Step 6:

[0651] The device establishes a secure communication channel with the server, encrypts the recorded data and emotional information, and then transmits the encrypted data to the server, where a security protocol is applied.

[0652] Step 7:

[0653] The server receives the recordings and emotional information, stores them securely in a database, ensures the integrity of the received data, and prepares it for analysis.

[0654] Step 8:

[0655] The server starts the speech recognition AI engine and begins analyzing the recorded data, converting the audio data into text data.

[0656] Step 9:

[0657] The server inputs the text data into a generative AI model to generate a summary, which also takes emotional information into account. The AI ​​model extracts important information and key points and provides a view that responds to changes in emotional state.

[0658] Step 10:

[0659] The generated text, summary, and sentiment information are stored in a database on the server, and this storage process ensures that data inconsistencies do not occur.

[0660] Step 11:

[0661] The server encrypts the generated text, summary, and emotion information and sends it back to the device through a secure communication channel. The returned data is transmitted with security in mind.

[0662] Step 12:

[0663] The device receives and decrypts the data from the server, and the received data is stored in local storage.

[0664] Step 13:

[0665] The user can open the call history screen on their device and check the text and summary, as well as the recognized emotion information. The history screen allows them to easily refer to the necessary information and changes in emotion. This is useful for analyzing whether the problem has been gradually resolved or whether the customer is satisfied.

[0666] As a concrete example, during a customer support call, the user (support representative) enables the call transcription option and emotion recognition function. When the meeting begins, the call is recorded and emotion information is collected. After the call ends, the recording data and emotion information are sent to the server for analysis. The generated text, summary, and emotion information are returned to the terminal, and the support representative can check the call content and changes in the customer's emotion from the history screen. This process allows the representative to improve the quality of customer support.

[0667] Example 2

[0668] 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."

[0669] Efficiently recording phone call content and managing important information is a time-consuming and labor-intensive process. Furthermore, manual transcription and summarization of call content can lack accuracy and consistency. Furthermore, it is difficult to properly recognize emotional information during a call and use it to evaluate the conversation.

[0670] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0671] In this invention, the server includes means for utilizing a speech recognition engine to convert call data into text, means for utilizing an emotion recognition engine to analyze emotional information, and means for utilizing a generative AI model to summarize the converted text including the emotional information. This allows efficient text conversion of the call content, recognition of emotional information, and summarization to be performed automatically, enabling the user to quickly and accurately grasp important information.

[0672] A "means for recording a call" is a function or device that records the audio of a call in real time and saves it in a format that can be analyzed later.

[0673] The "means for transmitting recorded call data to a server" refers to a function or device for securely compressing recorded voice data and transmitting it to a server via an encrypted communication channel.

[0674] "Means utilizing a speech recognition engine at a server to convert call data to text" means a function or device that operates an engine with speech recognition technology used to convert recorded voice data into text format.

[0675] "Means for utilizing an emotion recognition engine to analyze emotion information" refers to a function or device that operates an engine used to recognize and analyze a user's emotion from the converted text data.

[0676] "Means utilizing a generative AI model to summarize transformed text containing emotional information" refers to a function or device that operates a generative model using artificial intelligence to summarize text data containing emotional information.

[0677] The "means for returning the summarized text to the terminal" is a function or device for securely encrypting the generated summarized text and transmitting it to the user's terminal.

[0678] The "means for the user to check the returned text and summary" is a function or device that allows the user to view and check the text data and summary received at the terminal.

[0679] This invention combines a system that automatically converts telephone conversation content into text and summarizes it for easy user confirmation with an emotion engine that recognizes the user's emotions. Specific embodiments of this invention are described below.

[0680] Overall system configuration

[0681] This system consists of a user's device, a server, an emotion recognition engine, and related software components. The user's device records calls and transmits the data, while the server analyzes the call data, converts it to text, recognizes emotions, and generates summaries. The converted data, emotion information, and summaries are then sent to the user's device, where the user can view them on the call history screen.

[0682] Hardware and Software Configuration

[0683] Device: This could be a mobile device such as a smartphone or tablet, which requires recording software and encryption capabilities.

[0684] Server: A high-performance cloud or on-premise server is preferred, including a database and an AI engine for analytics.

[0685] Speech recognition engine: Uses highly accurate speech recognition technology such as Google Cloud Speech-to-Text.

[0686] Emotion recognition engine: Can use IBM Watson Tone Analyzer, etc. to analyze emotions from voice and text.

[0687] Generative AI models: Use models capable of automatic summary generation and natural language processing, such as OpenAI GPT-4.

[0688] System Operation

[0689] The system operates as follows.

[0690] User enables call transcription option

[0691] The user operates the device's settings screen and enables the "call transcription option." This setting causes all subsequent calls to be recorded and analyzed. The user's emotion recognition function is also enabled at the same time.

[0692] The device records the call

[0693] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records audio data in real time and temporarily stores it in a buffer. For security reasons, the recording data is encrypted using AES-256 encryption before being temporarily stored.

[0694] The device sends the recording data to the server

[0695] After the call ends, the device compresses the recording and sends it to the server via a secure channel (SSL / TLS protocol). This transmission process includes encryption, ensuring user privacy.

[0696] The server receives and analyzes the recorded data.

[0697] The server receives the recording and first stores it in a database. It then converts the recording into text using a speech recognition engine such as Google Cloud Speech-to-Text. This engine has advanced speech recognition technology and can handle noise and speaker characteristics.

[0698] Emotion recognition by emotion engine

[0699] The server inputs the converted text data into an emotion recognition engine such as IBM Watson Tone Analyzer to analyze emotions. The emotion engine analyzes the text content, tone of voice, and selected words to infer emotions such as joy, anger, sadness, and happiness.

[0700] The server summarizes the converted text

[0701] The converted text and sentiment information are then fed into a generative AI model, such as OpenAI GPT-4, to create a summary. This model quickly extracts key information and key points, and generates a mature summary that also reflects sentiment. Specific keywords and context are also taken into account, resulting in a high-quality summary.

[0702] The server sends the summarized text back to the terminal

[0703] The generated text, summary, and emotion information are then encrypted again and sent to the device via a secure channel. This process allows the user to review the call contents on their device, minimizing the risk of information leakage.

[0704] Users can view the transcript and summary on the call history screen.

[0705] Users can view the written text, summary, and recognized emotion information on the call history screen of their device. This screen displays a summary of each call and allows users to view detailed text. Emotion information is also displayed visually, allowing users to understand the flow of emotions.

[0706] Specific examples

[0707] Customer Support Calls

[0708] 1. The user (customer support representative) enables the call transcription option and emotion recognition feature.

[0709] 2. When a support call begins, the device will automatically record the call and collect emotional information.

[0710] 3. When the call ends, the device sends the recording data and emotional information to the server.

[0711] 4. The server receives the recording data and begins analyzing it.

[0712] 5. A voice recognition AI (e.g., Google Cloud Speech-to-Text) converts the call content into text, and an emotion engine (e.g., IBM Watson Tone Analyzer) analyzes emotions.

[0713] 6. A generative AI model (e.g., OpenAI GPT-4) summarizes the call and adjusts the quality of the summary based on emotional information.

[0714] 7. The server sends the generated text, summary, and emotion information to the terminal.

[0715] 8. The support representative opens the call history screen and checks the main points of the call, detailed discussions, and the customer's emotional state.

[0716] Prompt Sentence Examples

[0717] "Please transcribe and summarize this call and include the user's sentiment."

[0718] This system automatically converts phone conversations into text and generates summaries with emotional information, allowing users to efficiently manage and check important information. Furthermore, summaries that take emotional information into account also allow users to grasp the emotional flow of the conversation, making them extremely useful in customer service and other situations.

[0719] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0720] Step 1:

[0721] The user enables the call transcription option. In this step, the user operates the device's settings screen and turns on the "call transcription option." This setting will record all future calls and also enable the user's emotion recognition function. The input is the user's operation, and the output is the confirmation of the setting change.

[0722] Step 2:

[0723] A user initiates a phone call. When the call begins, the device automatically activates the recording function. During the call, the device records audio data in real time and temporarily stores it in a buffer. The hardware used is the device's built-in microphone, and the software used is a recording application. For security reasons, the recorded data is encrypted using the AES-256 encryption method and temporarily stored. The input is the start of the call, and the output is the encrypted recorded data.

[0724] Step 3:

[0725] After the call ends, the device compresses the recording and sends it to the server using a secure communication protocol (SSL / TLS). During this transmission process, the compressed recording is further encrypted with AES-256. The input is the encrypted recording, and the output is the data sent to the server.

[0726] Step 4:

[0727] The server receives the recording and stores it in a database. It then converts the recording into text using a speech recognition engine (e.g., Google Cloud Speech-to-Text). This engine uses noise filtering and speech enhancement techniques to convert the recording into highly accurate text. The input is the recording, and the output is text.

[0728] Step 5:

[0729] The server sends the text data to an emotion recognition engine (e.g., IBM Watson Tone Analyzer) for emotion recognition. The emotion recognition engine analyzes the content of the text, the tone of the voice, and the selected words to infer emotions such as joy, anger, sadness, and happiness. The input is text data, and the output is text data with emotional information.

[0730] Step 6:

[0731] The server inputs text data containing emotional information into a generative AI model (e.g., OpenAI GPT-4) to create a summary. This generative AI model quickly extracts important information and key points and generates a summary that reflects emotional information. The input is text data with emotional information, and the output is a summary text.

[0732] Step 7:

[0733] The generated text, summary, and sentiment information are again encrypted with AES-256 and sent to the device via a secure communication protocol, with the summary text and sentiment information as input and the encrypted data as output.

[0734] Step 8:

[0735] Users can view written text, summaries, and emotion information on the call history screen on their device. This screen displays a summary of each call and allows users to view detailed text. Emotion information is displayed visually, allowing users to understand the emotional flow of the conversation. The input is data returned from the server, and the output is text and emotion information displayed on the user's device.

[0736] (Application example 2)

[0737] 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."

[0738] In modern security services, it is difficult to quickly and accurately grasp the content of customer telephone conversations and extract and manage important information. Furthermore, if emotional information during a call could be grasped, it would be possible to respond based on the customer's psychological state, but this is also difficult to do using conventional means. The present invention aims to provide a system that automatically converts the content of a call into text, summarizes it, analyzes emotional information, and utilizes this information in customer service.

[0739] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for converting call data into text, means for analyzing the call data using an emotion engine and generating emotion information, and means for summarizing the text using the generated emotion information. This automatically converts the call content into text and summarizes it, and also analyzes the customer's emotion information, making it possible to quickly and accurately grasp important information in customer service and to respond appropriately based on the customer's psychological state.

[0740] "Call Data" refers to audio information recorded during a telephone call.

[0741] "Text conversion means" refers to software or hardware that performs the process of converting audio data into written information.

[0742] "Server" refers to a computer system that performs processes such as receiving call data, analyzing it, converting it into text, generating summaries, and collecting and transmitting emotional information.

[0743] An "emotion engine" refers to software that has algorithms or functions for analyzing voice and text data to estimate the speaker's emotional state.

[0744] "Summarization tool" refers to software or algorithms that extract key information from long text data and provide it in a shortened form.

[0745] "Means for returning" refers to the communication protocol or software for retransmitting data processed by the server to the user's terminal.

[0746] "Means for user confirmation" refers to an interface or software that allows a user to view the text, summary, and emotional information of the call using a terminal.

[0747] "Transcription options" refers to the settings that allow a user to enable a feature on the device that automatically transcribes phone conversations into text.

[0748] The "call history screen" refers to the display screen of the terminal on which the user can check the contents of past calls, their summaries, and emotional information.

[0749] "Generative AI models" refer to algorithms or software equipped with artificial intelligence technology used to summarize text data and analyze emotional information.

[0750] The system of the present invention is for smooth customer service in the field of security services. The system includes a means for recording calls, a means for transmitting the recorded data to a server, a means for analyzing the data on the server, a means for converting the data into text and generating a summary, and a means for analyzing emotional information. This makes it possible to improve the quality of customer service.

[0751] Hardware and software used

[0752] Terminal

[0753] This is a device such as a smartphone or smart glasses that users use when making phone calls. The device automatically records the call and temporarily stores the data. The recorded data is encrypted for security reasons.

[0754] server

[0755] The server is a computer system that receives and analyzes the recorded data. The software used includes a speech recognition AI engine, an emotion recognition engine, and a generative AI model.

[0756] Data processing and calculation procedures

[0757] 1. Record and send calls

[0758] When a user starts a phone call, the device automatically starts recording and temporarily saves the call. After the call ends, the recording data is sent to the server via a secure communication channel.

[0759] 2. Converting audio data into text

[0760] Once the server receives the recording, it uses a speech recognition AI engine (for example, Google Speech-to-Text API) to convert the audio data into text, which is then passed on to the next analysis step.

[0761] 3. Emotion Analysis

[0762] The text data is analyzed by an emotion engine to estimate the emotional state of the customer and the operator. Software used includes EmotionRecognition. This information is stored along with the text data.

[0763] 4. Summary Generation

[0764] It uses generative AI models (for example, Hugging Face's T5 model) to extract key information from long text data and provide it in a shortened form, which also reflects emotional information.

[0765] 5. Return and confirmation of results

[0766] The generated text, summary, and emotion information are encrypted and sent to the terminal via a secure communication channel. The user can then view this information on the call history screen.

[0767] Specific examples

[0768] For example, this system works in the background when a security company operator is talking to a customer. At the end of the call, the operator can view a summary of the call and sentiment information within the application, and if the customer is feeling anxious, they can quickly provide additional support.

[0769] Prompt Sentence Examples

[0770] An example of a prompt is as follows:

[0771] Please summarize the following Japanese conversation text and detect any emotions present:

[0772] <conversation_text>

[0773] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0774] Step 1:

[0775] The user enables the call transcription option and emotion recognition feature, which automatically records calls and collects emotional information. At this point, the device is ready to record and transmit data from the next call.

[0776] Step 2:

[0777] The device records the call. When a user initiates a phone call, the device automatically starts recording. During the call, audio data is recorded in real time and the recording data is temporarily stored in a buffer. This recording data is encrypted for security purposes.

[0778] Step 3:

[0779] The device sends the recording to the server. After the call ends, the recording is compressed and sent to the server over a secure channel. The data is re-encrypted during this process, protecting the user's privacy.

[0780] Step 4:

[0781] The server receives and analyzes the recorded data. When the server receives the recorded data, it first stores it in a database. Next, it converts the recorded data into text using a speech recognition AI engine. One example of the software used is the Google Speech-to-Text API. The input at this stage is the recorded data, and the output is text information.

[0782] Step 5:

[0783] The server analyzes the emotions in the voice data. An emotion engine is used to analyze the text data and generate emotional information for the customer and the operator. The software used is EmotionRecognition. In this process, the text data is the input and the output is emotional information.

[0784] Step 6:

[0785] The server summarizes the converted text. The data converted into text is then summarized by a generative AI model. The software used is the Hugging Face T5 model. Specifically, the text data and generated emotion information are input, the necessary information is extracted, and a summary of one or several sentences is output.

[0786] Step 7:

[0787] The server sends the summarized text and emotion information to the terminal. The generated text and emotion information are encrypted again and sent to the terminal via a secure communication channel. Here, the input is the summarized text and emotion information, and the output is the data sent back to the terminal.

[0788] Step 8:

[0789] The user checks the text, summary, and emotional information on the call history screen. The user opens the call history screen on the device and views the returned text, summary, and emotional information. This allows the user to grasp the main points of the call and the emotional flow of the conversation.

[0790] 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.

[0791] 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.

[0792] 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.

[0793] [Third embodiment]

[0794] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0795] 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.

[0796] 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).

[0797] 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.

[0798] 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.

[0799] 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).

[0800] 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.

[0801] 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.

[0802] 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.

[0803] 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.

[0804] 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.

[0805] 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."

[0806] The present invention provides a system that automatically converts the contents of a telephone conversation into text and summarizes it so that the user can easily check it. Specific embodiments of the system will be described below.

[0807] Overall system configuration

[0808] This system consists of a user's device, a server, and related software components. The user's device records and transmits the call data, while the server analyzes, converts, and summarizes the call data. The converted data and summary are then sent to the user's device, where the user can view them on the call history screen.

[0809] User enables call transcription option

[0810] The user operates the device's settings screen and enables the "call transcription option." This setting causes all subsequent calls to be recorded and analyzed.

[0811] The device records the call

[0812] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records the audio data in real time and temporarily stores it in an internal buffer. When the call ends, the recording data is saved.

[0813] The device sends the recording data to the server

[0814] After the call ends, the device compresses the recording and sends it over a secure channel to the server, where the transmission process includes encryption to protect the user's privacy.

[0815] The server receives and analyzes the recorded data.

[0816] Once the server receives the recording, it immediately stores it in a database. It then converts the recording into text using a speech recognition AI engine. This AI engine uses advanced speech recognition technology and can handle noise and speaker characteristics.

[0817] The server summarizes the converted text

[0818] The converted data is then analyzed by a generative AI model to create a summary. This model is able to quickly extract important information and key points and summarize them as a summary text. Specific keywords and context are also taken into account to improve the quality of the summary.

[0819] The server sends the summarized text back to the terminal

[0820] The generated text and summary are then re-encrypted and sent to the device over a secure channel, a process that allows users to view the call on their own device without worrying about information leaking to the outside world.

[0821] Users can view the transcript and summary on the call history screen.

[0822] Users can view the written text and summaries by operating the call history screen on their device. This screen displays a summary of each call and allows users to view detailed text if needed. This feature allows users to easily review long or important calls later.

[0823] Specific examples

[0824] Example: Business meeting call

[0825] 1. A user has a business meeting scheduled and enables the call transcription option.

[0826] 2. When the meeting starts, the device will automatically record the call.

[0827] 3. When the meeting ends, the device sends the recording data to the server.

[0828] 4. The server receives the recording data and begins analyzing it.

[0829] 5. A generative AI model converts the call content into text and then generates a summary.

[0830] 6. The server sends the generated text and summary to the terminal.

[0831] 7. After the meeting, the user opens the call history screen to review the key points and detailed discussions from the meeting.

[0832] In this way, the system of the present invention automatically converts telephone conversations into text and generates summaries, allowing users to effectively manage and review important information.

[0833] The processing flow will be explained below.

[0834] Step 1:

[0835] The user enables the call transcription option. Specifically, they open the device's settings screen, select "Call transcription options," and save the settings. This setting information is stored in the device's internal memory.

[0836] Step 2:

[0837] When a user makes or receives a call, the device detects the start of the call and automatically activates the recording function.

[0838] Step 3:

[0839] The device records the audio data during a call in real time and temporarily stores it in a buffer. The recorded data is encrypted for security reasons and temporarily stored.

[0840] Step 4:

[0841] The user ends the call. When the device detects the end of the call, it completes buffering the recorded data and begins preparing to save it as a file in a storage device.

[0842] Step 5:

[0843] The device compresses the recorded data and prepares it for transmission. The compression process is performed to ensure data quality.

[0844] Step 6:

[0845] The device establishes a secure communication channel with the server, encrypts the recording data, and transmits it to the server, with security protocols applied during data transfer.

[0846] Step 7:

[0847] The server receives the recording data and stores it securely in a database, where it is checked for integrity and prepared for further processing.

[0848] Step 8:

[0849] The server starts the speech recognition AI engine and begins analyzing the recorded data, which then starts the process of converting the voice data into text.

[0850] Step 9:

[0851] The server inputs the text data into a generative AI model to generate a summary, which extracts key information from the text and creates a summary.

[0852] Step 10:

[0853] The generated text and summaries are stored in a database on the server, and this storage process ensures that data inconsistencies do not occur.

[0854] Step 11:

[0855] The server encrypts the generated text and summary and sends it back to the device over a secure communication channel, ensuring that the returned data is transmitted securely.

[0856] Step 12:

[0857] The device receives and decrypts the data from the server, and the received data is stored in local storage.

[0858] Step 13:

[0859] The user opens the call history screen on their device and checks the text and summary of the call. The user can easily refer to the necessary information from the history screen.

[0860] Example 1

[0861] 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."

[0862] In recent years, there has been an increasing need to review the contents of business meetings and important phone calls later. However, with existing technology, it has been difficult to accurately and efficiently convert phone calls into text and summarize the key points. In particular, there have been challenges with the length of phone calls, the effects of noise, and the quality of the summaries of the converted text.

[0863] 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.

[0864] In this invention, the server includes means for converting call data into text using speech recognition technology, means for summarizing the converted text using a generative AI model, and means for returning the encrypted summary text to the terminal. This allows the user to efficiently check the call content by automatically and accurately converting the call content into text and summarizing important information.

[0865] The "call transcription option" refers to a setting function that records calls and converts the audio data into text for a summary.

[0866] "Means for recording" refers to the function of collecting audio during a call in real time and recording it as data.

[0867] "Means for compressing and transmitting via a secure communication channel" refers to the function of compressing recorded audio data using an algorithm that reduces the data size, encrypting it, and transmitting it using a communication protocol that prevents information leakage.

[0868] "Means of converting to text using speech recognition technology" refers to technology that analyzes recorded speech data and converts it into corresponding text data.

[0869] "Method of summarizing using a generative AI model" refers to artificial intelligence technology for extracting important information from large amounts of text data and generating concise summaries.

[0870] "Means for returning encrypted summary text to the terminal" refers to a technique for encrypting the generated text and summary to protect them and transmitting them securely to the user terminal.

[0871] The "call history screen" refers to the interface that the user uses to check the contents of past calls.

[0872] The present invention provides a system for automatically converting telephone conversations into text and summarizing them so that users can easily check them. Specific embodiments of the present invention will be described below.

[0873] Overall system configuration

[0874] This system consists of a user's device, a server, and related software components. The user's device records and transmits the call data, while the server analyzes, converts, and summarizes the call data. The converted data and summary are then sent to the user's device, where the user can view them on the call history screen.

[0875] User enables call transcription option

[0876] Users can activate the "call transcription option" by going to the device's settings screen. This setting ensures that all subsequent calls are recorded and analyzed, allowing users to easily review important calls later.

[0877] The device records the call

[0878] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records the audio data in real time and temporarily stores it in an internal buffer. When the call ends, the recording data is saved.

[0879] The device compresses and encrypts the recorded data and sends it to the server.

[0880] After the call ends, the device compresses the recording and encrypts it using an encryption algorithm such as AES, then transmits the recording to the server over a secure channel using the HTTPS protocol, thereby protecting the user's privacy.

[0881] The server receives and analyzes the recorded data.

[0882] When the server receives the recorded data, it first decrypts it. Then, it stores the received recorded data in a database (e.g., MySQL, MongoDB, etc.). Here, each call data is assigned a unique ID for management. The stored recorded data is then called into a speech recognition AI engine (e.g., a general cloud speech recognition service) to convert the voice data into text. At this time, a noise reduction algorithm is also applied to improve recognition accuracy.

[0883] The server summarizes the text using a generative AI model

[0884] The converted data is then fed into a generative AI model (e.g., a typical natural language generation model) to generate a summary. This process extracts the key parts of the text and incorporates specific keywords and contextual information to create a concise summary.

[0885] The server encrypts the generated text and summary and sends it to the device.

[0886] The generated text and summary are then encrypted again using an encryption algorithm (e.g., AES) and then sent to the device via a secure communication protocol (e.g., HTTPS). This process allows the user to view the call on their own device, but the information is not leaked to the outside world.

[0887] Users see the text and summary on the call history screen

[0888] Users can view the textual content and summary of calls by opening the call history screen on their device. This screen displays a summary of each call and allows users to view detailed text as needed, allowing users to effectively manage and check important information.

[0889] Specific examples

[0890] Example: Business meeting call

[0891] 1. A user has a business meeting scheduled, so they enable the "Call transcription option" in their device settings.

[0892] 2. When a meeting call begins, the device's recording function will automatically start and buffer the call audio in real time.

[0893] 3. After the meeting ends, the device compresses the recording data, encrypts it using AES, and then sends it to the server via HTTPS.

[0894] 4. The server receives the recording, decrypts it, and stores it in a database.

[0895] 5. The server uses voice recognition technology to convert the recording into text.

[0896] 6. The converted text is summarized using a generative AI model, which extracts key points.

[0897] 7. The generated text and summary are again AES encrypted and sent to the terminal via the HTTPS protocol.

[0898] 8. After the meeting, the user opens the call history screen on their device and checks the summary and detailed discussion content.

[0899] Example prompts to input to the generative AI model

[0900] "Please transcribe the following audio recording and write a summary: Audio recording: [audio file]"

[0901] The system automatically converts phone conversations into text and generates summaries, allowing users to effectively manage and review important information.

[0902] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0903] Step 1:

[0904] The user enables the call transcription option.

[0905] The user operates the device's settings screen to enable the "call transcription option." This setting causes all subsequent calls to be recorded and analyzed. The input is the user's operation, and the output is the setting change that enables the option.

[0906] Step 2:

[0907] Record calls on your device.

[0908] When a user starts a phone call, the device automatically starts the recording function. The input is the call start trigger, and the output is the recorded voice data. The recorded voice data is temporarily stored in a buffer.

[0909] Step 3:

[0910] The recorded data is compressed and encrypted and sent to the server.

[0911] Once the call is over, the device compresses the recording and applies AES encryption to it. The input is the recorded audio data, and the output is the compressed and encrypted audio data. This data is sent to the server using the HTTPS protocol.

[0912] Step 4:

[0913] The server receives the recording data and stores it in a database.

[0914] When the server receives the recorded data, it first decrypts the data. The input is the received compressed encrypted data, and the output is the decrypted audio data. Then, it stores the recorded data in the database.

[0915] Step 5:

[0916] The server converts the audio data into text.

[0917] The server inputs the stored voice data into a voice recognition AI engine and converts it into text data. The input is voice data and the output is text data. The voice recognition engine also runs a noise reduction algorithm to improve recognition accuracy.

[0918] Step 6:

[0919] The server summarizes the text using a generative AI model.

[0920] The server inputs the converted text into a generative AI model to extract important information and generate a summary. The input is text data, and the output is a summary text. The generated summary is constructed taking into account specific keywords and context, improving its quality.

[0921] Step 7:

[0922] The generated text and summary are encrypted and sent to the terminal.

[0923] The server encrypts the generated text and summary using AES and sends them to the terminal via HTTPS. The input is the generated text and summary, and the output is the encrypted text and summary.

[0924] Step 8:

[0925] The user sees the text and summary on the call history screen.

[0926] The user can open the call history screen on their device and check the text of the call and the summary. The input is the encrypted text and the decrypted data of the summary, and the output is the text and summary displayed on the call history screen. The user can also view the detailed text if necessary.

[0927] (Application example 1)

[0928] 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."

[0929] Conventional call recording systems simply convert recorded call data into text, and there is no way to determine whether the call content contains confidential information, making it impossible to reduce the risk of information leaks. Furthermore, they lacked a mechanism for notifying administrators of important information in real time, which required the time and effort of checking the information later. This created a need for improvements to enhance information security within companies.

[0930] 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.

[0931] In this invention, the server includes a means for marking the contents of a call that contain confidential or important information based on the results of text analysis, and a means for notifying an administrator when confidential information is detected, thereby enabling real-time monitoring of the contents of a call and immediate notification of important information.

[0932] "Call recording means" refers to devices and programs for automatically recording user-initiated telephone calls.

[0933] "Means for transmitting recorded call data to a server" refers to the communication means and encryption technology for securely transferring call data from a recording device to a server device.

[0934] "Means for converting call data to text on the server" refers to a server program and related technology for converting call data into text format using voice recognition technology.

[0935] "Means for summarizing converted text" refers to generative AI models and programs for analyzing text data, extracting key points, and generating concise summaries.

[0936] "Means for returning summarized text to the terminal" refers to a server program and communication means for securely transferring the generated summary text to the user's terminal.

[0937] "Means for the user to review the returned text and summary" refers to an interface and program that allows the user to view and review the returned text and summary on their own terminal.

[0938] "Means for marking call content containing confidential or sensitive information based on the results of text analysis" refers to programs and algorithms that analyze the generated text and identify and locate confidential or sensitive information contained therein.

[0939] "Means for notifying administrators when confidential information is detected" refers to programs and communication means for sending alerts and notifications to administrators in real time when confidential information is included in the content of a call.

[0940] The present invention relates to a system for converting telephone conversation content into text in real time and detecting and notifying confidential information. Specific embodiments of the system will be described below.

[0941] Overall system configuration

[0942] The system consists of a user's device, a server, and related software components. The user's device records calls and transmits the data, while the server analyzes the call data, converts it to text, generates summaries, and detects and notifies users of confidential information. The converted data, summaries, and detection notifications are then sent to the user's device, where they can view them on their call history screen.

[0943] Hardware and Software Configuration

[0944] Device: An application that runs on a smartphone or PC and has recording and data transmission functions.

[0945] Server: A cloud service such as AWS or Google Cloud that uses Google Cloud Speech-to-Text as a speech recognition engine, OpenAI's GPT-4 for text analysis and summary generation, and runs programs and algorithms for sensitive information detection and notification.

[0946] Communication Method: A secure encrypted communication channel is used to send and receive call data.

[0947] System Operation

[0948] 1. User enables the call transcription option:

[0949] The user operates the device's settings screen and enables the "call transcription option," which causes all subsequent calls to be recorded and analyzed.

[0950] 2. Your device will record the call:

[0951] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records the audio data in real time and temporarily stores it in an internal buffer. When the call ends, the recording data is saved.

[0952] 3. The device sends the recording data to the server:

[0953] After the call ends, the device compresses the recording and sends it over a secure channel to the server, where the transmission process includes encryption to protect the user's privacy.

[0954] 4. The server receives and analyzes the recording:

[0955] Once the server receives the recording, it immediately stores it in a database. It then converts the recording into text using a speech recognition AI engine. This AI engine uses advanced speech recognition technology and can handle noise and speaker characteristics.

[0956] 5. The server summarizes the converted text:

[0957] The converted data is then analyzed by a generative AI model to create a summary. This generative AI model is capable of quickly extracting important information and key points and summarizing them as a summary text.

[0958] 6. The server detects and notifies you of sensitive information:

[0959] The generated text and summary are simultaneously analyzed to detect whether it contains sensitive or important information, and if so, the sensitive information is marked based on specific keywords or context, and administrators are notified in real time.

[0960] 7. The server sends the summarized text and notification back to the device:

[0961] The generated text, summary, and confidential information notification are then re-encrypted and sent to the device over a secure channel. This process ensures that the user can view the call and notification on their own device, without the risk of information leaking to the outside world.

[0962] 8. User checks on call history screen:

[0963] Users can view written text and summaries, as well as confidential information notifications, by operating the call history screen on their device. This screen displays a summary of each call, and detailed text can be viewed as needed. This feature allows users to easily review long calls or important conversation content later. In addition, administrators can detect confidential information leaks in real time.

[0964] Specific examples

[0965] Example: Internal call monitoring

[0966] 1. A user installs a security call monitoring application to monitor internal company calls.

[0967] 2. Designated important calls are automatically recorded, encrypted, and sent to the server.

[0968] 3. On the server side, a voice recognition engine converts the call contents into text and analyzes whether it contains confidential information.

[0969] 4. The generative AI model creates a summary of the call data and presents it to the administrator through an interface.

[0970] 5. If necessary, monitor in real time and send an immediate alert if any unauthorized information leakage is detected.

[0971] Example prompt sentence:

[0972] "Please summarize the following call:\n\n{call content}\n\n"

[0973] In this way, the system of the present invention can improve information security within a company and effectively manage important information through automatic text conversion and summary generation of call content, as well as real-time detection and notification of confidential information.

[0974] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0975] Step 1:

[0976] The user enables the call transcription option. The user operates the device's settings screen and enables the "call transcription option." This setting causes all calls to be recorded and analyzed. The input is the user setting information, and the output is the enabled state of the recording function.

[0977] Step 2:

[0978] The device records the call. When the user starts a phone call, the recording function is automatically activated. During the call, the device records audio data in real time and temporarily stores it in an internal buffer. After the call ends, the recording data is saved. The input is the call audio data, and the output is the recorded audio file.

[0979] Step 3:

[0980] The device sends the recorded data to the server. After the call ends, the device compresses the recorded data and sends it to the server over a secure channel. This transmission process includes data encryption. The input is the recorded audio file, and the output is the encrypted data being transmitted.

[0981] Step 4:

[0982] The server receives and analyzes the recorded data. Once the server receives the recorded data, it stores it in a database. Next, it converts the recorded data into text using a speech recognition AI engine. This engine can handle noise and speaker characteristics. The input is encrypted voice data, and the output is text data.

[0983] Step 5:

[0984] The server summarizes the converted text. The converted data is analyzed by a generative AI model to extract important information and key points and create a summary. Prompt sentences are used in this process. The input is text data and the output is a summary text.

[0985] Step 6:

[0986] The server marks sensitive information based on the results of text analysis. The generated text and summary are analyzed to identify sensitive or important information based on specific keywords and context. The input is the summary text, and the output is the marked text.

[0987] Step 7:

[0988] The server notifies the administrator if sensitive information is detected. If sensitive information is detected, it sends alerts and notifications to the administrator in real time. The input is the marked text and the output is a notification to the administrator.

[0989] Step 8:

[0990] The server returns the summarized text and notification to the terminal. The generated text, summary, and notification of confidential information are sent to the terminal via a secure communication channel. The input is the marked text, summary, and notification, and the output is the text and summary displayed on the user's terminal.

[0991] Step 9:

[0992] The user checks the call history screen. The user can operate the call history screen on their device to check the written text, summary, and notification of confidential information. The input is the text, summary, and notification sent to the device, and the output is the user's confirmation result.

[0993] 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.

[0994] This invention combines a system that automatically converts the contents of a telephone conversation into text and summarizes it so that the user can easily check it, with an emotion engine that recognizes the user's emotions. Specific embodiments of this invention are described below.

[0995] Overall system configuration

[0996] This system consists of a user's device, a server, an emotion engine, and related software components. The user's device records calls and transmits the data, while the server analyzes the call data, converts it to text, recognizes emotions, and generates summaries. The converted data, emotion information, and summaries are then sent to the user's device, where the user can view them on the call history screen.

[0997] User enables call transcription option

[0998] The user operates the device's settings screen and enables the "call transcription option." This setting causes all subsequent calls to be recorded and analyzed. The user's emotion recognition function is also enabled at the same time.

[0999] The device records the call

[1000] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records audio data in real time and temporarily stores it in a buffer. The recording data is encrypted for security reasons and temporarily stored.

[1001] The device sends the recording data to the server

[1002] After the call ends, the device compresses the recording and sends it over a secure channel to the server, where the transmission process includes encryption to protect the user's privacy.

[1003] The server receives and analyzes the recorded data.

[1004] When the server receives the recorded data, it first stores it in a database. Then, it converts the recorded data into text using a speech recognition AI engine. This AI engine uses advanced speech recognition technology and can handle noise and speaker characteristics.

[1005] Emotion recognition by emotion engine

[1006] The server analyzes the recorded data and recognizes the user's emotions using an emotion engine, which analyzes the tone of the voice and the words chosen to express emotions such as joy, anger, sadness, and happiness.

[1007] The server summarizes the converted text

[1008] The text-converted data, including emotional information, is then fed into a generative AI model to create a summary. This generative AI model is capable of quickly extracting key information and key points and summarizing them as emotionally relevant summary text. Specific keywords and context are also taken into account to improve the quality of the summary.

[1009] The server sends the summarized text back to the terminal

[1010] The generated text, summary, and user emotion information are then encrypted again and sent to the device via a secure channel. This process allows users to view the call content on their own devices, eliminating the risk of information leaking to the outside world.

[1011] Users can view the transcript and summary on the call history screen.

[1012] Users can view the written text, summaries, and recognized emotional information by operating the call history screen on their device. This screen displays a summary of each call, and users can view detailed text as needed. Emotional information is also displayed visually, allowing users to understand the emotional flow of the conversation.

[1013] Specific examples

[1014] Example: Customer support call

[1015] 1. The user (customer support representative) enables the call transcription option and emotion recognition feature.

[1016] 2. When a support call begins, the device will automatically record the call and collect emotional information.

[1017] 3. When the call ends, the device sends the recording data and emotional information to the server.

[1018] 4. The server receives the recording data and begins analyzing it.

[1019] 5. Voice recognition AI converts the call content into text, and the emotion engine analyzes emotional information.

[1020] 6. The generative AI model summarizes the call content and adjusts the quality of the summary based on emotional information.

[1021] 7. The server sends the generated text, summary, and emotion information to the terminal.

[1022] 8. The support agent opens the call history screen to see the main points of the call, detailed discussions, and the customer's emotional state.

[1023] In this way, the system of the present invention automatically converts telephone conversations into text and generates summaries with emotional information, allowing users to effectively manage and review important information. Furthermore, summaries that incorporate emotional information also allow users to grasp the emotional flow of the conversation, making them extremely useful in scenarios such as customer service.

[1024] The processing flow will be explained below.

[1025] Step 1:

[1026] The user enables the call transcription option and emotion recognition function. Specifically, they open the device's settings screen, select "Call transcription option" and "Emotion recognition function," and save the settings. This setting information is stored in the device's internal memory.

[1027] Step 2:

[1028] When a user makes or receives a call, the device detects the start of the call and automatically activates the recording and emotion recognition functions.

[1029] Step 3:

[1030] The device records the voice data during a call in real time and temporarily stores it in a buffer. At the same time, the emotion recognition function analyzes the voice data and detects the user's emotions in real time.

[1031] Step 4:

[1032] The user ends the call. When the device detects the end of the call, it completes buffering the recorded data and begins preparing to save it as a file on a storage device. Emotional information is also saved at the same time.

[1033] Step 5:

[1034] The device compresses the recorded data and emotional information and prepares it for transmission. The data compression process is processed to ensure quality.

[1035] Step 6:

[1036] The device establishes a secure communication channel with the server, encrypts the recorded data and emotional information, and then transmits the encrypted data to the server, where a security protocol is applied.

[1037] Step 7:

[1038] The server receives the recordings and emotional information, stores them securely in a database, ensures the integrity of the received data, and prepares it for analysis.

[1039] Step 8:

[1040] The server starts the speech recognition AI engine and begins analyzing the recorded data, converting the audio data into text data.

[1041] Step 9:

[1042] The server inputs the text data into a generative AI model to generate a summary, which also takes emotional information into account. The AI ​​model extracts important information and key points and provides a view that responds to changes in emotional state.

[1043] Step 10:

[1044] The generated text, summary, and sentiment information are stored in a database on the server, and this storage process ensures that data inconsistencies do not occur.

[1045] Step 11:

[1046] The server encrypts the generated text, summary, and emotion information and sends it back to the device through a secure communication channel. The returned data is transmitted with security in mind.

[1047] Step 12:

[1048] The device receives and decrypts the data from the server, and the received data is stored in local storage.

[1049] Step 13:

[1050] The user can open the call history screen on their device and check the text and summary, as well as the recognized emotion information. The history screen allows them to easily refer to the necessary information and changes in emotion. This is useful for analyzing whether the problem has been gradually resolved or whether the customer is satisfied.

[1051] As a concrete example, during a customer support call, the user (support representative) enables the call transcription option and emotion recognition function. When the meeting begins, the call is recorded and emotion information is collected. After the call ends, the recording data and emotion information are sent to the server for analysis. The generated text, summary, and emotion information are returned to the terminal, and the support representative can check the call content and changes in the customer's emotion from the history screen. This process allows the representative to improve the quality of customer support.

[1052] Example 2

[1053] 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."

[1054] Efficiently recording phone call content and managing important information is a time-consuming and labor-intensive process. Furthermore, manual transcription and summarization of call content can lack accuracy and consistency. Furthermore, it is difficult to properly recognize emotional information during a call and use it to evaluate the conversation.

[1055] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1056] In this invention, the server includes means for utilizing a speech recognition engine to convert call data into text, means for utilizing an emotion recognition engine to analyze emotional information, and means for utilizing a generative AI model to summarize the converted text including the emotional information. This allows efficient text conversion of the call content, recognition of emotional information, and summarization to be performed automatically, enabling the user to quickly and accurately grasp important information.

[1057] A "means for recording a call" is a function or device that records the audio of a call in real time and saves it in a format that can be analyzed later.

[1058] The "means for transmitting recorded call data to a server" refers to a function or device for securely compressing recorded voice data and transmitting it to a server via an encrypted communication channel.

[1059] "Means utilizing a speech recognition engine at a server to convert call data to text" means a function or device that operates an engine with speech recognition technology used to convert recorded voice data into text format.

[1060] "Means for utilizing an emotion recognition engine to analyze emotion information" refers to a function or device that operates an engine used to recognize and analyze a user's emotion from the converted text data.

[1061] "Means utilizing a generative AI model to summarize transformed text containing emotional information" refers to a function or device that operates a generative model using artificial intelligence to summarize text data containing emotional information.

[1062] The "means for returning the summarized text to the terminal" is a function or device for securely encrypting the generated summarized text and transmitting it to the user's terminal.

[1063] The "means for the user to check the returned text and summary" is a function or device that allows the user to view and check the text data and summary received at the terminal.

[1064] This invention combines a system that automatically converts telephone conversation content into text and summarizes it for easy user confirmation with an emotion engine that recognizes the user's emotions. Specific embodiments of this invention are described below.

[1065] Overall system configuration

[1066] This system consists of a user's device, a server, an emotion recognition engine, and related software components. The user's device records calls and transmits the data, while the server analyzes the call data, converts it to text, recognizes emotions, and generates summaries. The converted data, emotion information, and summaries are then sent to the user's device, where the user can view them on the call history screen.

[1067] Hardware and Software Configuration

[1068] Device: This could be a mobile device such as a smartphone or tablet, which requires recording software and encryption capabilities.

[1069] Server: A high-performance cloud or on-premise server is preferred, including a database and an AI engine for analytics.

[1070] Speech recognition engine: Uses highly accurate speech recognition technology such as Google Cloud Speech-to-Text.

[1071] Emotion recognition engine: Can use IBM Watson Tone Analyzer, etc. to analyze emotions from voice and text.

[1072] Generative AI models: Use models capable of automatic summary generation and natural language processing, such as OpenAI GPT-4.

[1073] System Operation

[1074] The system operates as follows.

[1075] User enables call transcription option

[1076] The user operates the device's settings screen and enables the "call transcription option." This setting causes all subsequent calls to be recorded and analyzed. The user's emotion recognition function is also enabled at the same time.

[1077] The device records the call

[1078] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records audio data in real time and temporarily stores it in a buffer. For security reasons, the recording data is encrypted using AES-256 encryption before being temporarily stored.

[1079] The device sends the recording data to the server

[1080] After the call ends, the device compresses the recording and sends it to the server via a secure channel (SSL / TLS protocol). This transmission process includes encryption, ensuring user privacy.

[1081] The server receives and analyzes the recorded data.

[1082] The server receives the recording and first stores it in a database. It then converts the recording into text using a speech recognition engine such as Google Cloud Speech-to-Text. This engine has advanced speech recognition technology and can handle noise and speaker characteristics.

[1083] Emotion recognition by emotion engine

[1084] The server inputs the converted text data into an emotion recognition engine such as IBM Watson Tone Analyzer to analyze emotions. The emotion engine analyzes the text content, tone of voice, and selected words to infer emotions such as joy, anger, sadness, and happiness.

[1085] The server summarizes the converted text

[1086] The converted text and sentiment information are then fed into a generative AI model, such as OpenAI GPT-4, to create a summary. This model quickly extracts key information and key points, and generates a mature summary that also reflects sentiment. Specific keywords and context are also taken into account, resulting in a high-quality summary.

[1087] The server sends the summarized text back to the terminal

[1088] The generated text, summary, and emotion information are then encrypted again and sent to the device via a secure channel. This process allows the user to review the call contents on their device, minimizing the risk of information leakage.

[1089] Users can view the transcript and summary on the call history screen.

[1090] Users can view the written text, summary, and recognized emotion information on the call history screen of their device. This screen displays a summary of each call and allows users to view detailed text. Emotion information is also displayed visually, allowing users to understand the flow of emotions.

[1091] Specific examples

[1092] Customer Support Calls

[1093] 1. The user (customer support representative) enables the call transcription option and emotion recognition feature.

[1094] 2. When a support call begins, the device will automatically record the call and collect emotional information.

[1095] 3. When the call ends, the device sends the recording data and emotional information to the server.

[1096] 4. The server receives the recording data and begins analyzing it.

[1097] 5. A voice recognition AI (e.g., Google Cloud Speech-to-Text) converts the call content into text, and an emotion engine (e.g., IBM Watson Tone Analyzer) analyzes emotions.

[1098] 6. A generative AI model (e.g., OpenAI GPT-4) summarizes the call and adjusts the quality of the summary based on emotional information.

[1099] 7. The server sends the generated text, summary, and emotion information to the terminal.

[1100] 8. The support representative opens the call history screen and checks the main points of the call, detailed discussions, and the customer's emotional state.

[1101] Prompt Sentence Examples

[1102] "Please transcribe and summarize this call and include the user's sentiment."

[1103] This system automatically converts phone conversations into text and generates summaries with emotional information, allowing users to efficiently manage and check important information. Furthermore, summaries that take emotional information into account also allow users to grasp the emotional flow of the conversation, making them extremely useful in customer service and other situations.

[1104] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1105] Step 1:

[1106] The user enables the call transcription option. In this step, the user operates the device's settings screen and turns on the "call transcription option." This setting will record all future calls and also enable the user's emotion recognition function. The input is the user's operation, and the output is the confirmation of the setting change.

[1107] Step 2:

[1108] A user initiates a phone call. When the call begins, the device automatically activates the recording function. During the call, the device records audio data in real time and temporarily stores it in a buffer. The hardware used is the device's built-in microphone, and the software used is a recording application. For security reasons, the recorded data is encrypted using the AES-256 encryption method and temporarily stored. The input is the start of the call, and the output is the encrypted recorded data.

[1109] Step 3:

[1110] After the call ends, the device compresses the recording and sends it to the server using a secure communication protocol (SSL / TLS). During this transmission process, the compressed recording is further encrypted with AES-256. The input is the encrypted recording, and the output is the data sent to the server.

[1111] Step 4:

[1112] The server receives the recording and stores it in a database. It then converts the recording into text using a speech recognition engine (e.g., Google Cloud Speech-to-Text). This engine uses noise filtering and speech enhancement techniques to convert the recording into highly accurate text. The input is the recording, and the output is text.

[1113] Step 5:

[1114] The server sends the text data to an emotion recognition engine (e.g., IBM Watson Tone Analyzer) for emotion recognition. The emotion recognition engine analyzes the content of the text, the tone of the voice, and the selected words to infer emotions such as joy, anger, sadness, and happiness. The input is text data, and the output is text data with emotional information.

[1115] Step 6:

[1116] The server inputs text data containing emotional information into a generative AI model (e.g., OpenAI GPT-4) to create a summary. This generative AI model quickly extracts important information and key points and generates a summary that reflects emotional information. The input is text data with emotional information, and the output is a summary text.

[1117] Step 7:

[1118] The generated text, summary, and sentiment information are again encrypted with AES-256 and sent to the device via a secure communication protocol, with the summary text and sentiment information as input and the encrypted data as output.

[1119] Step 8:

[1120] Users can view written text, summaries, and emotion information on the call history screen on their device. This screen displays a summary of each call and allows users to view detailed text. Emotion information is displayed visually, allowing users to understand the emotional flow of the conversation. The input is data returned from the server, and the output is text and emotion information displayed on the user's device.

[1121] (Application example 2)

[1122] 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."

[1123] In modern security services, it is difficult to quickly and accurately grasp the content of customer telephone conversations and extract and manage important information. Furthermore, if emotional information during a call could be grasped, it would be possible to respond based on the customer's psychological state, but this is also difficult to do using conventional means. The present invention aims to provide a system that automatically converts the content of a call into text, summarizes it, analyzes emotional information, and utilizes this information in customer service.

[1124] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for converting call data into text, means for analyzing the call data using an emotion engine and generating emotion information, and means for summarizing the text using the generated emotion information. This automatically converts the call content into text and summarizes it, and also analyzes the customer's emotion information, making it possible to quickly and accurately grasp important information in customer service and to respond appropriately based on the customer's psychological state.

[1125] "Call Data" refers to audio information recorded during a telephone call.

[1126] "Text conversion means" refers to software or hardware that performs the process of converting audio data into written information.

[1127] "Server" refers to a computer system that performs processes such as receiving call data, analyzing it, converting it into text, generating summaries, and collecting and transmitting emotional information.

[1128] An "emotion engine" refers to software that has algorithms or functions for analyzing voice and text data to estimate the speaker's emotional state.

[1129] "Summarization tool" refers to software or algorithms that extract key information from long text data and provide it in a shortened form.

[1130] "Means for returning" refers to the communication protocol or software for retransmitting data processed by the server to the user's terminal.

[1131] "Means for user confirmation" refers to an interface or software that allows a user to view the text, summary, and emotional information of the call using a terminal.

[1132] "Transcription options" refers to the settings that allow a user to enable a feature on the device that automatically transcribes phone conversations into text.

[1133] The "call history screen" refers to the display screen of the terminal on which the user can check the contents of past calls, their summaries, and emotional information.

[1134] "Generative AI models" refer to algorithms or software equipped with artificial intelligence technology used to summarize text data and analyze emotional information.

[1135] The system of the present invention is for smooth customer service in the field of security services. The system includes a means for recording calls, a means for transmitting the recorded data to a server, a means for analyzing the data on the server, a means for converting the data into text and generating a summary, and a means for analyzing emotional information. This makes it possible to improve the quality of customer service.

[1136] Hardware and software used

[1137] Terminal

[1138] This is a device such as a smartphone or smart glasses that users use when making phone calls. The device automatically records the call and temporarily stores the data. The recorded data is encrypted for security reasons.

[1139] server

[1140] The server is a computer system that receives and analyzes the recorded data. The software used includes a speech recognition AI engine, an emotion recognition engine, and a generative AI model.

[1141] Data processing and calculation procedures

[1142] 1. Record and send calls

[1143] When a user starts a phone call, the device automatically starts recording and temporarily saves the call. After the call ends, the recording data is sent to the server via a secure communication channel.

[1144] 2. Converting audio data into text

[1145] Once the server receives the recording, it uses a speech recognition AI engine (for example, Google Speech-to-Text API) to convert the audio data into text, which is then passed on to the next analysis step.

[1146] 3. Emotion Analysis

[1147] The text data is analyzed by an emotion engine to estimate the emotional state of the customer and the operator. Software used includes EmotionRecognition. This information is stored along with the text data.

[1148] 4. Summary Generation

[1149] It uses generative AI models (for example, Hugging Face's T5 model) to extract key information from long text data and provide it in a shortened form, which also reflects emotional information.

[1150] 5. Return and confirmation of results

[1151] The generated text, summary, and emotion information are encrypted and sent to the terminal via a secure communication channel. The user can then view this information on the call history screen.

[1152] Specific examples

[1153] For example, this system works in the background when a security company operator is talking to a customer. At the end of the call, the operator can view a summary of the call and sentiment information within the application, and if the customer is feeling anxious, they can quickly provide additional support.

[1154] Prompt Sentence Examples

[1155] An example of a prompt is as follows:

[1156] Please summarize the following Japanese conversation text and detect any emotions present:

[1157] <conversation_text>

[1158] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1159] Step 1:

[1160] The user enables the call transcription option and emotion recognition feature, which automatically records calls and collects emotional information. At this point, the device is ready to record and transmit data from the next call.

[1161] Step 2:

[1162] The device records the call. When a user initiates a phone call, the device automatically starts recording. During the call, audio data is recorded in real time and the recording data is temporarily stored in a buffer. This recording data is encrypted for security purposes.

[1163] Step 3:

[1164] The device sends the recording to the server. After the call ends, the recording is compressed and sent to the server over a secure channel. The data is re-encrypted during this process, protecting the user's privacy.

[1165] Step 4:

[1166] The server receives and analyzes the recorded data. When the server receives the recorded data, it first stores it in a database. Next, it converts the recorded data into text using a speech recognition AI engine. One example of the software used is the Google Speech-to-Text API. The input at this stage is the recorded data, and the output is text information.

[1167] Step 5:

[1168] The server analyzes the emotions in the voice data. An emotion engine is used to analyze the text data and generate emotional information for the customer and the operator. The software used is EmotionRecognition. In this process, the text data is the input and the output is emotional information.

[1169] Step 6:

[1170] The server summarizes the converted text. The data converted into text is then summarized by a generative AI model. The software used is the Hugging Face T5 model. Specifically, the text data and generated emotion information are input, the necessary information is extracted, and a summary of one or several sentences is output.

[1171] Step 7:

[1172] The server sends the summarized text and emotion information to the terminal. The generated text and emotion information are encrypted again and sent to the terminal via a secure communication channel. Here, the input is the summarized text and emotion information, and the output is the data sent back to the terminal.

[1173] Step 8:

[1174] The user checks the text, summary, and emotional information on the call history screen. The user opens the call history screen on the device and views the returned text, summary, and emotional information. This allows the user to grasp the main points of the call and the emotional flow of the conversation.

[1175] 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.

[1176] 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.

[1177] 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.

[1178] [Fourth embodiment]

[1179] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1180] 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.

[1181] 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).

[1182] 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.

[1183] 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.

[1184] 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).

[1185] 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.

[1186] 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.

[1187] 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.

[1188] 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.

[1189] 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.

[1190] 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.

[1191] 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."

[1192] The present invention provides a system that automatically converts the contents of a telephone conversation into text and summarizes it so that the user can easily check it. Specific embodiments of the system will be described below.

[1193] Overall system configuration

[1194] This system consists of a user's device, a server, and related software components. The user's device records and transmits the call data, while the server analyzes, converts, and summarizes the call data. The converted data and summary are then sent to the user's device, where the user can view them on the call history screen.

[1195] User enables call transcription option

[1196] The user operates the device's settings screen and enables the "call transcription option." This setting causes all subsequent calls to be recorded and analyzed.

[1197] The device records the call

[1198] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records the audio data in real time and temporarily stores it in an internal buffer. When the call ends, the recording data is saved.

[1199] The device sends the recording data to the server

[1200] After the call ends, the device compresses the recording and sends it over a secure channel to the server, where the transmission process includes encryption to protect the user's privacy.

[1201] The server receives and analyzes the recorded data.

[1202] Once the server receives the recording, it immediately stores it in a database. It then converts the recording into text using a speech recognition AI engine. This AI engine uses advanced speech recognition technology and can handle noise and speaker characteristics.

[1203] The server summarizes the converted text

[1204] The converted data is then analyzed by a generative AI model to create a summary. This model is able to quickly extract important information and key points and summarize them as a summary text. Specific keywords and context are also taken into account to improve the quality of the summary.

[1205] The server sends the summarized text back to the terminal

[1206] The generated text and summary are then re-encrypted and sent to the device over a secure channel, a process that allows users to view the call on their own device without worrying about information leaking to the outside world.

[1207] Users can view the transcript and summary on the call history screen.

[1208] Users can view the written text and summaries by operating the call history screen on their device. This screen displays a summary of each call and allows users to view detailed text if needed. This feature allows users to easily review long or important calls later.

[1209] Specific examples

[1210] Example: Business meeting call

[1211] 1. A user has a business meeting scheduled and enables the call transcription option.

[1212] 2. When the meeting starts, the device will automatically record the call.

[1213] 3. When the meeting ends, the device sends the recording data to the server.

[1214] 4. The server receives the recording data and begins analyzing it.

[1215] 5. A generative AI model converts the call content into text and then generates a summary.

[1216] 6. The server sends the generated text and summary to the terminal.

[1217] 7. After the meeting, the user opens the call history screen to review the key points and detailed discussions from the meeting.

[1218] In this way, the system of the present invention automatically converts telephone conversations into text and generates summaries, allowing users to effectively manage and review important information.

[1219] The processing flow will be explained below.

[1220] Step 1:

[1221] The user enables the call transcription option. Specifically, they open the device's settings screen, select "Call transcription options," and save the settings. This setting information is stored in the device's internal memory.

[1222] Step 2:

[1223] When a user makes or receives a call, the device detects the start of the call and automatically activates the recording function.

[1224] Step 3:

[1225] The device records the audio data during a call in real time and temporarily stores it in a buffer. The recorded data is encrypted for security reasons and temporarily stored.

[1226] Step 4:

[1227] The user ends the call. When the device detects the end of the call, it completes buffering the recorded data and begins preparing to save it as a file in a storage device.

[1228] Step 5:

[1229] The device compresses the recorded data and prepares it for transmission. The compression process is performed to ensure data quality.

[1230] Step 6:

[1231] The device establishes a secure communication channel with the server, encrypts the recording data, and transmits it to the server, with security protocols applied during data transfer.

[1232] Step 7:

[1233] The server receives the recording data and stores it securely in a database, where it is checked for integrity and prepared for further processing.

[1234] Step 8:

[1235] The server starts the speech recognition AI engine and begins analyzing the recorded data, which then starts the process of converting the voice data into text.

[1236] Step 9:

[1237] The server inputs the text data into a generative AI model to generate a summary, which extracts key information from the text and creates a summary.

[1238] Step 10:

[1239] The generated text and summaries are stored in a database on the server, and this storage process ensures that data inconsistencies do not occur.

[1240] Step 11:

[1241] The server encrypts the generated text and summary and sends it back to the device over a secure communication channel, ensuring that the returned data is transmitted securely.

[1242] Step 12:

[1243] The device receives and decrypts the data from the server, and the received data is stored in local storage.

[1244] Step 13:

[1245] The user opens the call history screen on their device and checks the text and summary of the call. The user can easily refer to the necessary information from the history screen.

[1246] Example 1

[1247] 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."

[1248] In recent years, there has been an increasing need to review the contents of business meetings and important phone calls later. However, with existing technology, it has been difficult to accurately and efficiently convert phone calls into text and summarize the key points. In particular, there have been challenges with the length of phone calls, the effects of noise, and the quality of the summaries of the converted text.

[1249] 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.

[1250] In this invention, the server includes means for converting call data into text using speech recognition technology, means for summarizing the converted text using a generative AI model, and means for returning the encrypted summary text to the terminal. This allows the user to efficiently check the call content by automatically and accurately converting the call content into text and summarizing important information.

[1251] The "call transcription option" refers to a setting function that records calls and converts the audio data into text for a summary.

[1252] "Means for recording" refers to the function of collecting audio during a call in real time and recording it as data.

[1253] "Means for compressing and transmitting via a secure communication channel" refers to the function of compressing recorded audio data using an algorithm that reduces the data size, encrypting it, and transmitting it using a communication protocol that prevents information leakage.

[1254] "Means of converting to text using speech recognition technology" refers to technology that analyzes recorded speech data and converts it into corresponding text data.

[1255] "Method of summarizing using a generative AI model" refers to artificial intelligence technology for extracting important information from large amounts of text data and generating concise summaries.

[1256] "Means for returning encrypted summary text to the terminal" refers to a technique for encrypting the generated text and summary to protect them and transmitting them securely to the user terminal.

[1257] The "call history screen" refers to the interface that the user uses to check the contents of past calls.

[1258] The present invention provides a system for automatically converting telephone conversations into text and summarizing them so that users can easily check them. Specific embodiments of the present invention will be described below.

[1259] Overall system configuration

[1260] This system consists of a user's device, a server, and related software components. The user's device records and transmits the call data, while the server analyzes, converts, and summarizes the call data. The converted data and summary are then sent to the user's device, where the user can view them on the call history screen.

[1261] User enables call transcription option

[1262] Users can activate the "call transcription option" by going to the device's settings screen. This setting ensures that all subsequent calls are recorded and analyzed, allowing users to easily review important calls later.

[1263] The device records the call

[1264] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records the audio data in real time and temporarily stores it in an internal buffer. When the call ends, the recording data is saved.

[1265] The device compresses and encrypts the recorded data and sends it to the server.

[1266] After the call ends, the device compresses the recording and encrypts it using an encryption algorithm such as AES, then transmits the recording to the server over a secure channel using the HTTPS protocol, thereby protecting the user's privacy.

[1267] The server receives and analyzes the recorded data.

[1268] When the server receives the recorded data, it first decrypts it. Then, it stores the received recorded data in a database (e.g., MySQL, MongoDB, etc.). Here, each call data is assigned a unique ID for management. The stored recorded data is then called into a speech recognition AI engine (e.g., a general cloud speech recognition service) to convert the voice data into text. At this time, a noise reduction algorithm is also applied to improve recognition accuracy.

[1269] The server summarizes the text using a generative AI model

[1270] The converted data is then fed into a generative AI model (e.g., a typical natural language generation model) to generate a summary. This process extracts the key parts of the text and incorporates specific keywords and contextual information to create a concise summary.

[1271] The server encrypts the generated text and summary and sends it to the device.

[1272] The generated text and summary are then encrypted again using an encryption algorithm (e.g., AES) and then sent to the device via a secure communication protocol (e.g., HTTPS). This process allows the user to view the call on their own device, but the information is not leaked to the outside world.

[1273] Users see the text and summary on the call history screen

[1274] Users can view the textual content and summary of calls by opening the call history screen on their device. This screen displays a summary of each call and allows users to view detailed text as needed, allowing users to effectively manage and check important information.

[1275] Specific examples

[1276] Example: Business meeting call

[1277] 1. A user has a business meeting scheduled, so they enable the "Call transcription option" in their device settings.

[1278] 2. When a meeting call begins, the device's recording function will automatically start and buffer the call audio in real time.

[1279] 3. After the meeting ends, the device compresses the recording data, encrypts it using AES, and then sends it to the server via HTTPS.

[1280] 4. The server receives the recording, decrypts it, and stores it in a database.

[1281] 5. The server uses voice recognition technology to convert the recording into text.

[1282] 6. The converted text is summarized using a generative AI model, which extracts key points.

[1283] 7. The generated text and summary are again AES encrypted and sent to the terminal via the HTTPS protocol.

[1284] 8. After the meeting, the user opens the call history screen on their device and checks the summary and detailed discussion content.

[1285] Example prompts to input to the generative AI model

[1286] "Please transcribe the following audio recording and write a summary: Audio recording: [audio file]"

[1287] The system automatically converts phone conversations into text and generates summaries, allowing users to effectively manage and review important information.

[1288] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1289] Step 1:

[1290] The user enables the call transcription option.

[1291] The user operates the device's settings screen to enable the "call transcription option." This setting causes all subsequent calls to be recorded and analyzed. The input is the user's operation, and the output is the setting change that enables the option.

[1292] Step 2:

[1293] Record calls on your device.

[1294] When a user starts a phone call, the device automatically starts the recording function. The input is the call start trigger, and the output is the recorded voice data. The recorded voice data is temporarily stored in a buffer.

[1295] Step 3:

[1296] The recorded data is compressed and encrypted and sent to the server.

[1297] Once the call is over, the device compresses the recording and applies AES encryption to it. The input is the recorded audio data, and the output is the compressed and encrypted audio data. This data is sent to the server using the HTTPS protocol.

[1298] Step 4:

[1299] The server receives the recording data and stores it in a database.

[1300] When the server receives the recorded data, it first decrypts the data. The input is the received compressed encrypted data, and the output is the decrypted audio data. Then, it stores the recorded data in the database.

[1301] Step 5:

[1302] The server converts the audio data into text.

[1303] The server inputs the stored voice data into a voice recognition AI engine and converts it into text data. The input is voice data and the output is text data. The voice recognition engine also runs a noise reduction algorithm to improve recognition accuracy.

[1304] Step 6:

[1305] The server summarizes the text using a generative AI model.

[1306] The server inputs the converted text into a generative AI model to extract important information and generate a summary. The input is text data, and the output is a summary text. The generated summary is constructed taking into account specific keywords and context, improving its quality.

[1307] Step 7:

[1308] The generated text and summary are encrypted and sent to the terminal.

[1309] The server encrypts the generated text and summary using AES and sends them to the terminal via HTTPS. The input is the generated text and summary, and the output is the encrypted text and summary.

[1310] Step 8:

[1311] The user sees the text and summary on the call history screen.

[1312] The user can open the call history screen on their device and check the text of the call and the summary. The input is the encrypted text and the decrypted data of the summary, and the output is the text and summary displayed on the call history screen. The user can also view the detailed text if necessary.

[1313] (Application example 1)

[1314] 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."

[1315] Conventional call recording systems simply convert recorded call data into text, and there is no way to determine whether the call content contains confidential information, making it impossible to reduce the risk of information leaks. Furthermore, they lacked a mechanism for notifying administrators of important information in real time, which required the time and effort of checking the information later. This created a need for improvements to enhance information security within companies.

[1316] 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.

[1317] In this invention, the server includes a means for marking the contents of a call that contain confidential or important information based on the results of text analysis, and a means for notifying an administrator when confidential information is detected, thereby enabling real-time monitoring of the contents of a call and immediate notification of important information.

[1318] "Call recording means" refers to devices and programs for automatically recording user-initiated telephone calls.

[1319] "Means for transmitting recorded call data to a server" refers to the communication means and encryption technology for securely transferring call data from a recording device to a server device.

[1320] "Means for converting call data to text on the server" refers to a server program and related technology for converting call data into text format using voice recognition technology.

[1321] "Means for summarizing converted text" refers to generative AI models and programs for analyzing text data, extracting key points, and generating concise summaries.

[1322] "Means for returning summarized text to the terminal" refers to a server program and communication means for securely transferring the generated summary text to the user's terminal.

[1323] "Means for the user to review the returned text and summary" refers to an interface and program that allows the user to view and review the returned text and summary on their own terminal.

[1324] "Means for marking call content containing confidential or sensitive information based on the results of text analysis" refers to programs and algorithms that analyze the generated text and identify and locate confidential or sensitive information contained therein.

[1325] "Means for notifying administrators when confidential information is detected" refers to programs and communication means for sending alerts and notifications to administrators in real time when confidential information is included in the content of a call.

[1326] The present invention relates to a system for converting telephone conversation content into text in real time and detecting and notifying confidential information. Specific embodiments of the system will be described below.

[1327] Overall system configuration

[1328] The system consists of a user's device, a server, and related software components. The user's device records calls and transmits the data, while the server analyzes the call data, converts it to text, generates summaries, and detects and notifies users of confidential information. The converted data, summaries, and detection notifications are then sent to the user's device, where they can view them on their call history screen.

[1329] Hardware and Software Configuration

[1330] Device: An application that runs on a smartphone or PC and has recording and data transmission functions.

[1331] Server: A cloud service such as AWS or Google Cloud that uses Google Cloud Speech-to-Text as a speech recognition engine, OpenAI's GPT-4 for text analysis and summary generation, and runs programs and algorithms for sensitive information detection and notification.

[1332] Communication Method: A secure encrypted communication channel is used to send and receive call data.

[1333] System Operation

[1334] 1. User enables the call transcription option:

[1335] The user operates the device's settings screen and enables the "call transcription option," which causes all subsequent calls to be recorded and analyzed.

[1336] 2. Your device will record the call:

[1337] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records the audio data in real time and temporarily stores it in an internal buffer. When the call ends, the recording data is saved.

[1338] 3. The device sends the recording data to the server:

[1339] After the call ends, the device compresses the recording and sends it over a secure channel to the server, where the transmission process includes encryption to protect the user's privacy.

[1340] 4. The server receives and analyzes the recording:

[1341] Once the server receives the recording, it immediately stores it in a database. It then converts the recording into text using a speech recognition AI engine. This AI engine uses advanced speech recognition technology and can handle noise and speaker characteristics.

[1342] 5. The server summarizes the converted text:

[1343] The converted data is then analyzed by a generative AI model to create a summary. This generative AI model is capable of quickly extracting important information and key points and summarizing them as a summary text.

[1344] 6. The server detects and notifies you of sensitive information:

[1345] The generated text and summary are simultaneously analyzed to detect whether it contains sensitive or important information, and if so, the sensitive information is marked based on specific keywords or context, and administrators are notified in real time.

[1346] 7. The server sends the summarized text and notification back to the device:

[1347] The generated text, summary, and confidential information notification are then re-encrypted and sent to the device over a secure channel. This process ensures that the user can view the call and notification on their own device, without the risk of information leaking to the outside world.

[1348] 8. User checks on call history screen:

[1349] Users can view written text and summaries, as well as confidential information notifications, by operating the call history screen on their device. This screen displays a summary of each call, and detailed text can be viewed as needed. This feature allows users to easily review long calls or important conversation content later. In addition, administrators can detect confidential information leaks in real time.

[1350] Specific examples

[1351] Example: Internal call monitoring

[1352] 1. A user installs a security call monitoring application to monitor internal company calls.

[1353] 2. Designated important calls are automatically recorded, encrypted, and sent to the server.

[1354] 3. On the server side, a voice recognition engine converts the call contents into text and analyzes whether it contains confidential information.

[1355] 4. The generative AI model creates a summary of the call data and presents it to the administrator through an interface.

[1356] 5. If necessary, monitor in real time and send an immediate alert if any unauthorized information leakage is detected.

[1357] Example prompt sentence:

[1358] "Please summarize the following call:\n\n{call content}\n\n"

[1359] In this way, the system of the present invention can improve information security within a company and effectively manage important information through automatic text conversion and summary generation of call content, as well as real-time detection and notification of confidential information.

[1360] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1361] Step 1:

[1362] The user enables the call transcription option. The user operates the device's settings screen and enables the "call transcription option." This setting causes all calls to be recorded and analyzed. The input is the user setting information, and the output is the enabled state of the recording function.

[1363] Step 2:

[1364] The device records the call. When the user starts a phone call, the recording function is automatically activated. During the call, the device records audio data in real time and temporarily stores it in an internal buffer. After the call ends, the recording data is saved. The input is the call audio data, and the output is the recorded audio file.

[1365] Step 3:

[1366] The device sends the recorded data to the server. After the call ends, the device compresses the recorded data and sends it to the server over a secure channel. This transmission process includes data encryption. The input is the recorded audio file, and the output is the encrypted data being transmitted.

[1367] Step 4:

[1368] The server receives and analyzes the recorded data. Once the server receives the recorded data, it stores it in a database. Next, it converts the recorded data into text using a speech recognition AI engine. This engine can handle noise and speaker characteristics. The input is encrypted voice data, and the output is text data.

[1369] Step 5:

[1370] The server summarizes the converted text. The converted data is analyzed by a generative AI model to extract important information and key points and create a summary. Prompt sentences are used in this process. The input is text data and the output is a summary text.

[1371] Step 6:

[1372] The server marks sensitive information based on the results of text analysis. The generated text and summary are analyzed to identify sensitive or important information based on specific keywords and context. The input is the summary text, and the output is the marked text.

[1373] Step 7:

[1374] The server notifies the administrator if sensitive information is detected. If sensitive information is detected, it sends alerts and notifications to the administrator in real time. The input is the marked text and the output is a notification to the administrator.

[1375] Step 8:

[1376] The server returns the summarized text and notification to the terminal. The generated text, summary, and notification of confidential information are sent to the terminal via a secure communication channel. The input is the marked text, summary, and notification, and the output is the text and summary displayed on the user's terminal.

[1377] Step 9:

[1378] The user checks the call history screen. The user can operate the call history screen on their device to check the written text, summary, and notification of confidential information. The input is the text, summary, and notification sent to the device, and the output is the user's confirmation result.

[1379] 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.

[1380] This invention combines a system that automatically converts the contents of a telephone conversation into text and summarizes it so that the user can easily check it, with an emotion engine that recognizes the user's emotions. Specific embodiments of this invention are described below.

[1381] Overall system configuration

[1382] This system consists of a user's device, a server, an emotion engine, and related software components. The user's device records calls and transmits the data, while the server analyzes the call data, converts it to text, recognizes emotions, and generates summaries. The converted data, emotion information, and summaries are then sent to the user's device, where the user can view them on the call history screen.

[1383] User enables call transcription option

[1384] The user operates the device's settings screen and enables the "call transcription option." This setting causes all subsequent calls to be recorded and analyzed. The user's emotion recognition function is also enabled at the same time.

[1385] The device records the call

[1386] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records audio data in real time and temporarily stores it in a buffer. The recording data is encrypted for security reasons and temporarily stored.

[1387] The device sends the recording data to the server

[1388] After the call ends, the device compresses the recording and sends it over a secure channel to the server, where the transmission process includes encryption to protect the user's privacy.

[1389] The server receives and analyzes the recorded data.

[1390] When the server receives the recorded data, it first stores it in a database. Then, it converts the recorded data into text using a speech recognition AI engine. This AI engine uses advanced speech recognition technology and can handle noise and speaker characteristics.

[1391] Emotion recognition by emotion engine

[1392] The server analyzes the recorded data and recognizes the user's emotions using an emotion engine, which analyzes the tone of the voice and the words chosen to express emotions such as joy, anger, sadness, and happiness.

[1393] The server summarizes the converted text

[1394] The text-converted data, including emotional information, is then fed into a generative AI model to create a summary. This generative AI model is capable of quickly extracting key information and key points and summarizing them as emotionally relevant summary text. Specific keywords and context are also taken into account to improve the quality of the summary.

[1395] The server sends the summarized text back to the terminal

[1396] The generated text, summary, and user emotion information are then encrypted again and sent to the device via a secure channel. This process allows users to view the call content on their own devices, eliminating the risk of information leaking to the outside world.

[1397] Users can view the transcript and summary on the call history screen.

[1398] Users can view the written text, summaries, and recognized emotional information by operating the call history screen on their device. This screen displays a summary of each call, and users can view detailed text as needed. Emotional information is also displayed visually, allowing users to understand the emotional flow of the conversation.

[1399] Specific examples

[1400] Example: Customer support call

[1401] 1. The user (customer support representative) enables the call transcription option and emotion recognition feature.

[1402] 2. When a support call begins, the device will automatically record the call and collect emotional information.

[1403] 3. When the call ends, the device sends the recording data and emotional information to the server.

[1404] 4. The server receives the recording data and begins analyzing it.

[1405] 5. Voice recognition AI converts the call content into text, and the emotion engine analyzes emotional information.

[1406] 6. The generative AI model summarizes the call content and adjusts the quality of the summary based on emotional information.

[1407] 7. The server sends the generated text, summary, and emotion information to the terminal.

[1408] 8. The support agent opens the call history screen to see the main points of the call, detailed discussions, and the customer's emotional state.

[1409] In this way, the system of the present invention automatically converts telephone conversations into text and generates summaries with emotional information, allowing users to effectively manage and review important information. Furthermore, summaries that incorporate emotional information also allow users to grasp the emotional flow of the conversation, making them extremely useful in scenarios such as customer service.

[1410] The processing flow will be explained below.

[1411] Step 1:

[1412] The user enables the call transcription option and emotion recognition function. Specifically, they open the device's settings screen, select "Call transcription option" and "Emotion recognition function," and save the settings. This setting information is stored in the device's internal memory.

[1413] Step 2:

[1414] When a user makes or receives a call, the device detects the start of the call and automatically activates the recording and emotion recognition functions.

[1415] Step 3:

[1416] The device records the voice data during a call in real time and temporarily stores it in a buffer. At the same time, the emotion recognition function analyzes the voice data and detects the user's emotions in real time.

[1417] Step 4:

[1418] The user ends the call. When the device detects the end of the call, it completes buffering the recorded data and begins preparing to save it as a file on a storage device. Emotional information is also saved at the same time.

[1419] Step 5:

[1420] The device compresses the recorded data and emotional information and prepares it for transmission. The data compression process is processed to ensure quality.

[1421] Step 6:

[1422] The device establishes a secure communication channel with the server, encrypts the recorded data and emotional information, and then transmits the encrypted data to the server, where a security protocol is applied.

[1423] Step 7:

[1424] The server receives the recordings and emotional information, stores them securely in a database, ensures the integrity of the received data, and prepares it for analysis.

[1425] Step 8:

[1426] The server starts the speech recognition AI engine and begins analyzing the recorded data, converting the audio data into text data.

[1427] Step 9:

[1428] The server inputs the text data into a generative AI model to generate a summary, which also takes emotional information into account. The AI ​​model extracts important information and key points and provides a view that responds to changes in emotional state.

[1429] Step 10:

[1430] The generated text, summary, and sentiment information are stored in a database on the server, and this storage process ensures that data inconsistencies do not occur.

[1431] Step 11:

[1432] The server encrypts the generated text, summary, and emotion information and sends it back to the device through a secure communication channel. The returned data is transmitted with security in mind.

[1433] Step 12:

[1434] The device receives and decrypts the data from the server, and the received data is stored in local storage.

[1435] Step 13:

[1436] The user can open the call history screen on their device and check the text and summary, as well as the recognized emotion information. The history screen allows them to easily refer to the necessary information and changes in emotion. This is useful for analyzing whether the problem has been gradually resolved or whether the customer is satisfied.

[1437] As a concrete example, during a customer support call, the user (support representative) enables the call transcription option and emotion recognition function. When the meeting begins, the call is recorded and emotion information is collected. After the call ends, the recording data and emotion information are sent to the server for analysis. The generated text, summary, and emotion information are returned to the terminal, and the support representative can check the call content and changes in the customer's emotion from the history screen. This process allows the representative to improve the quality of customer support.

[1438] Example 2

[1439] 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."

[1440] Efficiently recording phone call content and managing important information is a time-consuming and labor-intensive process. Furthermore, manual transcription and summarization of call content can lack accuracy and consistency. Furthermore, it is difficult to properly recognize emotional information during a call and use it to evaluate the conversation.

[1441] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1442] In this invention, the server includes means for utilizing a speech recognition engine to convert call data into text, means for utilizing an emotion recognition engine to analyze emotional information, and means for utilizing a generative AI model to summarize the converted text including the emotional information. This allows efficient text conversion of the call content, recognition of emotional information, and summarization to be performed automatically, enabling the user to quickly and accurately grasp important information.

[1443] A "means for recording a call" is a function or device that records the audio of a call in real time and saves it in a format that can be analyzed later.

[1444] The "means for transmitting recorded call data to a server" refers to a function or device for securely compressing recorded voice data and transmitting it to a server via an encrypted communication channel.

[1445] "Means utilizing a speech recognition engine at a server to convert call data to text" means a function or device that operates an engine with speech recognition technology used to convert recorded voice data into text format.

[1446] "Means for utilizing an emotion recognition engine to analyze emotion information" refers to a function or device that operates an engine used to recognize and analyze a user's emotion from the converted text data.

[1447] "Means utilizing a generative AI model to summarize transformed text containing emotional information" refers to a function or device that operates a generative model using artificial intelligence to summarize text data containing emotional information.

[1448] The "means for returning the summarized text to the terminal" is a function or device for securely encrypting the generated summarized text and transmitting it to the user's terminal.

[1449] The "means for the user to check the returned text and summary" is a function or device that allows the user to view and check the text data and summary received at the terminal.

[1450] This invention combines a system that automatically converts telephone conversation content into text and summarizes it for easy user confirmation with an emotion engine that recognizes the user's emotions. Specific embodiments of this invention are described below.

[1451] Overall system configuration

[1452] This system consists of a user's device, a server, an emotion recognition engine, and related software components. The user's device records calls and transmits the data, while the server analyzes the call data, converts it to text, recognizes emotions, and generates summaries. The converted data, emotion information, and summaries are then sent to the user's device, where the user can view them on the call history screen.

[1453] Hardware and Software Configuration

[1454] Device: This could be a mobile device such as a smartphone or tablet, which requires recording software and encryption capabilities.

[1455] Server: A high-performance cloud or on-premise server is preferred, including a database and an AI engine for analytics.

[1456] Speech recognition engine: Uses highly accurate speech recognition technology such as Google Cloud Speech-to-Text.

[1457] Emotion recognition engine: Can use IBM Watson Tone Analyzer, etc. to analyze emotions from voice and text.

[1458] Generative AI models: Use models capable of automatic summary generation and natural language processing, such as OpenAI GPT-4.

[1459] System Operation

[1460] The system operates as follows.

[1461] User enables call transcription option

[1462] The user operates the device's settings screen and enables the "call transcription option." This setting causes all subsequent calls to be recorded and analyzed. The user's emotion recognition function is also enabled at the same time.

[1463] The device records the call

[1464] When a user starts a phone call, the device automatically activates the recording function. During the call, the device records audio data in real time and temporarily stores it in a buffer. For security reasons, the recording data is encrypted using AES-256 encryption before being temporarily stored.

[1465] The device sends the recording data to the server

[1466] After the call ends, the device compresses the recording and sends it to the server via a secure channel (SSL / TLS protocol). This transmission process includes encryption, ensuring user privacy.

[1467] The server receives and analyzes the recorded data.

[1468] The server receives the recording and first stores it in a database. It then converts the recording into text using a speech recognition engine such as Google Cloud Speech-to-Text. This engine has advanced speech recognition technology and can handle noise and speaker characteristics.

[1469] Emotion recognition by emotion engine

[1470] The server inputs the converted text data into an emotion recognition engine such as IBM Watson Tone Analyzer to analyze emotions. The emotion engine analyzes the text content, tone of voice, and selected words to infer emotions such as joy, anger, sadness, and happiness.

[1471] The server summarizes the converted text

[1472] The converted text and sentiment information are then fed into a generative AI model, such as OpenAI GPT-4, to create a summary. This model quickly extracts key information and key points, and generates a mature summary that also reflects sentiment. Specific keywords and context are also taken into account, resulting in a high-quality summary.

[1473] The server sends the summarized text back to the terminal

[1474] The generated text, summary, and emotion information are then encrypted again and sent to the device via a secure channel. This process allows the user to review the call contents on their device, minimizing the risk of information leakage.

[1475] Users can view the transcript and summary on the call history screen.

[1476] Users can view the written text, summary, and recognized emotion information on the call history screen of their device. This screen displays a summary of each call and allows users to view detailed text. Emotion information is also displayed visually, allowing users to understand the flow of emotions.

[1477] Specific examples

[1478] Customer Support Calls

[1479] 1. The user (customer support representative) enables the call transcription option and emotion recognition feature.

[1480] 2. When a support call begins, the device will automatically record the call and collect emotional information.

[1481] 3. When the call ends, the device sends the recording data and emotional information to the server.

[1482] 4. The server receives the recording data and begins analyzing it.

[1483] 5. A voice recognition AI (e.g., Google Cloud Speech-to-Text) converts the call content into text, and an emotion engine (e.g., IBM Watson Tone Analyzer) analyzes emotions.

[1484] 6. A generative AI model (e.g., OpenAI GPT-4) summarizes the call and adjusts the quality of the summary based on emotional information.

[1485] 7. The server sends the generated text, summary, and emotion information to the terminal.

[1486] 8. The support representative opens the call history screen and checks the main points of the call, detailed discussions, and the customer's emotional state.

[1487] Prompt Sentence Examples

[1488] "Please transcribe and summarize this call and include the user's sentiment."

[1489] This system automatically converts phone conversations into text and generates summaries with emotional information, allowing users to efficiently manage and check important information. Furthermore, summaries that take emotional information into account also allow users to grasp the emotional flow of the conversation, making them extremely useful in customer service and other situations.

[1490] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1491] Step 1:

[1492] The user enables the call transcription option. In this step, the user operates the device's settings screen and turns on the "call transcription option." This setting will record all future calls and also enable the user's emotion recognition function. The input is the user's operation, and the output is the confirmation of the setting change.

[1493] Step 2:

[1494] A user initiates a phone call. When the call begins, the device automatically activates the recording function. During the call, the device records audio data in real time and temporarily stores it in a buffer. The hardware used is the device's built-in microphone, and the software used is a recording application. For security reasons, the recorded data is encrypted using the AES-256 encryption method and temporarily stored. The input is the start of the call, and the output is the encrypted recorded data.

[1495] Step 3:

[1496] After the call ends, the device compresses the recording and sends it to the server using a secure communication protocol (SSL / TLS). During this transmission process, the compressed recording is further encrypted with AES-256. The input is the encrypted recording, and the output is the data sent to the server.

[1497] Step 4:

[1498] The server receives the recording and stores it in a database. It then converts the recording into text using a speech recognition engine (e.g., Google Cloud Speech-to-Text). This engine uses noise filtering and speech enhancement techniques to convert the recording into highly accurate text. The input is the recording, and the output is text.

[1499] Step 5:

[1500] The server sends the text data to an emotion recognition engine (e.g., IBM Watson Tone Analyzer) for emotion recognition. The emotion recognition engine analyzes the content of the text, the tone of the voice, and the selected words to infer emotions such as joy, anger, sadness, and happiness. The input is text data, and the output is text data with emotional information.

[1501] Step 6:

[1502] The server inputs text data containing emotional information into a generative AI model (e.g., OpenAI GPT-4) to create a summary. This generative AI model quickly extracts important information and key points and generates a summary that reflects emotional information. The input is text data with emotional information, and the output is a summary text.

[1503] Step 7:

[1504] The generated text, summary, and sentiment information are again encrypted with AES-256 and sent to the device via a secure communication protocol, with the summary text and sentiment information as input and the encrypted data as output.

[1505] Step 8:

[1506] Users can view written text, summaries, and emotion information on the call history screen on their device. This screen displays a summary of each call and allows users to view detailed text. Emotion information is displayed visually, allowing users to understand the emotional flow of the conversation. The input is data returned from the server, and the output is text and emotion information displayed on the user's device.

[1507] (Application example 2)

[1508] 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."

[1509] In modern security services, it is difficult to quickly and accurately grasp the content of customer telephone conversations and extract and manage important information. Furthermore, if emotional information during a call could be grasped, it would be possible to respond based on the customer's psychological state, but this is also difficult to do using conventional means. The present invention aims to provide a system that automatically converts the content of a call into text, summarizes it, analyzes emotional information, and utilizes this information in customer service.

[1510] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for converting call data into text, means for analyzing the call data using an emotion engine and generating emotion information, and means for summarizing the text using the generated emotion information. This automatically converts the call content into text and summarizes it, and also analyzes the customer's emotion information, making it possible to quickly and accurately grasp important information in customer service and to respond appropriately based on the customer's psychological state.

[1511] "Call Data" refers to audio information recorded during a telephone call.

[1512] "Text conversion means" refers to software or hardware that performs the process of converting audio data into written information.

[1513] "Server" refers to a computer system that performs processes such as receiving call data, analyzing it, converting it into text, generating summaries, and collecting and transmitting emotional information.

[1514] An "emotion engine" refers to software that has algorithms or functions for analyzing voice and text data to estimate the speaker's emotional state.

[1515] "Summarization tool" refers to software or algorithms that extract key information from long text data and provide it in a shortened form.

[1516] "Means for returning" refers to the communication protocol or software for retransmitting data processed by the server to the user's terminal.

[1517] "Means for user confirmation" refers to an interface or software that allows a user to view the text, summary, and emotional information of the call using a terminal.

[1518] "Transcription options" refers to the settings that allow a user to enable a feature on the device that automatically transcribes phone conversations into text.

[1519] The "call history screen" refers to the display screen of the terminal on which the user can check the contents of past calls, their summaries, and emotional information.

[1520] "Generative AI models" refer to algorithms or software equipped with artificial intelligence technology used to summarize text data and analyze emotional information.

[1521] The system of the present invention is for smooth customer service in the field of security services. The system includes a means for recording calls, a means for transmitting the recorded data to a server, a means for analyzing the data on the server, a means for converting the data into text and generating a summary, and a means for analyzing emotional information. This makes it possible to improve the quality of customer service.

[1522] Hardware and software used

[1523] Terminal

[1524] This is a device such as a smartphone or smart glasses that users use when making phone calls. The device automatically records the call and temporarily stores the data. The recorded data is encrypted for security reasons.

[1525] server

[1526] The server is a computer system that receives and analyzes the recorded data. The software used includes a speech recognition AI engine, an emotion recognition engine, and a generative AI model.

[1527] Data processing and calculation procedures

[1528] 1. Record and send calls

[1529] When a user starts a phone call, the device automatically starts recording and temporarily saves the call. After the call ends, the recording data is sent to the server via a secure communication channel.

[1530] 2. Converting audio data into text

[1531] Once the server receives the recording, it uses a speech recognition AI engine (for example, Google Speech-to-Text API) to convert the audio data into text, which is then passed on to the next analysis step.

[1532] 3. Emotion Analysis

[1533] The text data is analyzed by an emotion engine to estimate the emotional state of the customer and the operator. Software used includes EmotionRecognition. This information is stored along with the text data.

[1534] 4. Summary Generation

[1535] It uses generative AI models (for example, Hugging Face's T5 model) to extract key information from long text data and provide it in a shortened form, which also reflects emotional information.

[1536] 5. Return and confirmation of results

[1537] The generated text, summary, and emotion information are encrypted and sent to the terminal via a secure communication channel. The user can then view this information on the call history screen.

[1538] Specific examples

[1539] For example, this system works in the background when a security company operator is talking to a customer. At the end of the call, the operator can view a summary of the call and sentiment information within the application, and if the customer is feeling anxious, they can quickly provide additional support.

[1540] Prompt Sentence Examples

[1541] An example of a prompt is as follows:

[1542] Please summarize the following Japanese conversation text and detect any emotions present:

[1543] <conversation_text>

[1544] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1545] Step 1:

[1546] The user enables the call transcription option and emotion recognition feature, which automatically records calls and collects emotional information. At this point, the device is ready to record and transmit data from the next call.

[1547] Step 2:

[1548] The device records the call. When a user initiates a phone call, the device automatically starts recording. During the call, audio data is recorded in real time and the recording data is temporarily stored in a buffer. This recording data is encrypted for security purposes.

[1549] Step 3:

[1550] The device sends the recording to the server. After the call ends, the recording is compressed and sent to the server over a secure channel. The data is re-encrypted during this process, protecting the user's privacy.

[1551] Step 4:

[1552] The server receives and analyzes the recorded data. When the server receives the recorded data, it first stores it in a database. Next, it converts the recorded data into text using a speech recognition AI engine. One example of the software used is the Google Speech-to-Text API. The input at this stage is the recorded data, and the output is text information.

[1553] Step 5:

[1554] The server analyzes the emotions in the voice data. An emotion engine is used to analyze the text data and generate emotional information for the customer and the operator. The software used is EmotionRecognition. In this process, the text data is the input and the output is emotional information.

[1555] Step 6:

[1556] The server summarizes the converted text. The data converted into text is then summarized by a generative AI model. The software used is the Hugging Face T5 model. Specifically, the text data and generated emotion information are input, the necessary information is extracted, and a summary of one or several sentences is output.

[1557] Step 7:

[1558] The server sends the summarized text and emotion information to the terminal. The generated text and emotion information are encrypted again and sent to the terminal via a secure communication channel. Here, the input is the summarized text and emotion information, and the output is the data sent back to the terminal.

[1559] Step 8:

[1560] The user checks the text, summary, and emotional information on the call history screen. The user opens the call history screen on the device and views the returned text, summary, and emotional information. This allows the user to grasp the main points of the call and the emotional flow of the conversation.

[1561] 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.

[1562] 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.

[1563] 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.

[1564] 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.

[1565] 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.

[1566] 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.

[1567] 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).

[1568] 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.

[1569] 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."

[1570] 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.

[1571] 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).

[1572] 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.

[1573] 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.

[1574] 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.

[1575] 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.

[1576] 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.

[1577] 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.

[1578] 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.

[1579] 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.

[1580] 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.

[1581] 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.

[1582] The following is further disclosed regarding the above embodiment.

[1583] (Claim 1)

[1584] A means of recording calls;

[1585] means for transmitting the recorded call data to a server;

[1586] means for converting the call data to text at a server;

[1587] a means for summarizing the converted text;

[1588] means for transmitting the summarized text back to the terminal;

[1589] a means for the user to review the returned text and summary;

[1590] A system including:

[1591] (Claim 2)

[1592] 10. The system of claim 1, wherein the user enables a call transcription option.

[1593] (Claim 3)

[1594] 10. The system of claim 1, wherein the call history screen allows the user to view the written text and summary.

[1595] "Example 1"

[1596] (Claim 1)

[1597] a means by which a user can enable a call transcription option;

[1598] A means of recording calls;

[1599] means for compressing the recorded call data and transmitting it to a server via a secure communication channel;

[1600] A means for converting the call data into text using voice recognition technology in a server;

[1601] A means for summarizing the converted text using a generative AI model; and

[1602] means for returning the encrypted summary text to the terminal;

[1603] a means for the user to view the returned text and summary on the call history screen;

[1604] A system including:

[1605] (Claim 2)

[1606] 10. The system of claim 1, wherein the system enables a call transcription option.

[1607] (Claim 3)

[1608] 10. The system of claim 1, wherein the generative AI model is used to summarize the call data and the summarized text is encrypted and returned.

[1609] "Application Example 1"

[1610] (Claim 1)

[1611] A means of recording calls;

[1612] means for transmitting the recorded call data to a server;

[1613] means for converting the call data to text at a server;

[1614] a means for summarizing the converted text;

[1615] means for transmitting the summarized text back to the terminal;

[1616] a means for the user to review the returned text and summary;

[1617] A means for marking call content containing confidential or important information based on the results of text analysis;

[1618] A means of notifying administrators when sensitive information is detected; and

[1619] A system including:

[1620] (Claim 2)

[1621] 10. The system of claim 1, wherein the user enables a call transcription option.

[1622] (Claim 3)

[1623] 10. The system of claim 1, wherein the call history screen allows the user to view the written text and summary.

[1624] "Example 2: Combining Emotion Engines"

[1625] (Claim 1)

[1626] A means of recording calls;

[1627] means for transmitting the recorded call data to a server;

[1628] a means for utilizing a speech recognition engine in a server to convert the call data into text;

[1629] utilizing an emotion recognition engine to analyze emotion information from the converted text;

[1630] a means for utilizing a generative AI model to summarize the transformed text including sentiment information;

[1631] means for transmitting the summarized text back to the terminal;

[1632] a means for the user to review the returned text and summary;

[1633] A system including:

[1634] (Claim 2)

[1635] 10. The system of claim 1, wherein the user enables a call transcription option.

[1636] (Claim 3)

[1637] 10. The system of claim 1, wherein the call history screen allows the user to view written text, emotional information, and summaries.

[1638] "Application example 2 when combining emotion engines"

[1639] (Claim 1)

[1640] A means of recording calls;

[1641] means for transmitting the recorded call data to a server;

[1642] means for converting the call data to text at a server;

[1643] means for analyzing call data by an emotion engine and generating emotion information;

[1644] means for summarizing the text using the generated emotion information;

[1645] means for transmitting the summarized text and emotion information back to the terminal;

[1646] a means for the user to review the returned text and summary and sentiment information;

[1647] A system including:

[1648] (Claim 2)

[1649] 10. The system of claim 1, wherein the user enables the call transcription option and emotion recognition feature.

[1650] (Claim 3)

[1651] 10. The system of claim 1, wherein the call history screen allows the user to view written text, summaries, and emotional information. [Explanation of symbols]

[1652] 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. A means of recording calls; means for transmitting the recorded call data to a server; means for converting the call data to text at a server; a means for summarizing the converted text; means for transmitting the summarized text back to the terminal; a means for the user to review the returned text and summary; A system including:

2. The system of claim 1 , wherein the user enables a call transcription option.

3. 10. The system of claim 1, wherein a user can view the written text and summary on a call history screen.

Images