System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving speech data; means for converting the received speech data into text data; means for analyzing the converted text data and inferring a context and an intention of a speaker; means for complementing the inferred intention and correcting an unclear portion; and means for visually displaying the complemented text data.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] Conventional telephone systems have the problem that accurate communication is difficult to hear when the other party has poor pronunciation, when there is ambient noise, or when the line conditions are poor. This can lead to having to repeat what is being said multiple times, which can make the other party feel rude or make them uncomfortable communicating over the phone. There is a need for technology that can solve this problem and make telephone communication smoother. [Means for solving the problem]

[0005] This invention provides a system including means for receiving voice data, means for converting the received voice data into text data, means for analyzing the converted text data and inferring the context and the speaker's intent, means for completing the inferred intent and correcting unclear parts, and means for visually displaying the completed text data. Furthermore, by adding means for receiving voice data in real time and transcribing it, means for completing emotions and intent based on the intonation and strength of the voice during analysis, and means for highlighting important keywords and phrases in the completed text data, it becomes easier to accurately and visually understand the contents of telephone conversations. As a result, it becomes possible to facilitate smooth telephone communication and eliminate difficulties in hearing.

[0006] "Audio data" refers to audio signals in analog or digital form received through an audio input device such as a telephone or microphone.

[0007] "Means for receiving" refers to a device or module for receiving and capturing audio data, including a telephone line, an internet connection, wireless communication, and the like.

[0008] "Text data" refers to data expressed in character format that is converted from voice data using voice recognition technology.

[0009] "Means for converting" refers to technology or devices for converting voice data into text data, and typical examples include voice recognition software and hardware.

[0010] "Means of analysis" refers to technologies and modules for analyzing text data to understand the context and the speaker's intent, such as natural language analysis and machine learning algorithms.

[0011] "Means of inference" refers to technologies or modules for inferring the speaker's intention based on analyzed data, such as inference models and generative AI.

[0012] "Completion methods" are technologies or modules that interpolate unclear parts and correct text to make it more natural and easier to understand.

[0013] "Visual display means" refers to the technology or device used to display the parsed and completed text data on a screen or display, in order to present the information in a way that is easily understandable to the user.

[0014] "Real-time" means that audio data is received, converted, analyzed, and displayed nearly simultaneously, with minimal or no time delay.

[0015] "Transcription" is the process of converting audio data into text, typically using speech recognition technology.

[0016] "Intonation" refers to changes in pitch and volume in a speaker's voice, and is an important element in conveying emotion and intention.

[0017] "Keywords" are words or phrases that are considered particularly important within text data and serve as indicators for understanding the context. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0026] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0039] This invention relates to a "visual telephone" system that analyzes telephone voices and visually complements them to eliminate the difficulty of hearing telephone calls. This system operates in cooperation with the server, terminals, and users, analyzing voice data and displaying it visually in real time.

[0040] 1. System Configuration

[0041] server

[0042] The server is the central device that receives the voice data from the phone and processes it in real time. The server first receives the voice data and sequentially transfers it to the voice recognition module. The voice recognition module converts the voice data into text data. This converted text data is analyzed by the generative AI to infer the speaker's intent and context. Finally, the inferred intent is complemented and corrected into a format that is easy to display visually.

[0043] Terminal

[0044] The terminal is a device that allows the user to receive information visually. It receives the completed text data sent from the server and displays it on the screen. The terminal is made up of a television or a dedicated display, and has the function of highlighting important keywords and phrases. The user looks at this display to understand the content of the call.

[0045] User

[0046] The user is the subject of communication via telephone, and receives visually supplemented information through the terminal. The user can read the displayed text data and make requests or follow-up inquiries as needed. This allows the user to visually confirm content that is difficult to understand through voice alone.

[0047] 2. Program Processing

[0048] Audio Receiving and Forwarding

[0049] The server receives the voice data and transfers it to the voice recognition module in real time. The server stores the voice data in a buffer and supplies the data to the voice recognition module as needed.

[0050] Speech-to-text conversion

[0051] The speech recognition module converts the received voice data into text data in real time, including phonemic analysis of the voice and converting it into words, and the generated text data is sent to the analysis module.

[0052] Contextual analysis and intent inference

[0053] The generative AI (analysis module) analyzes text data and performs analysis to understand the context and the speaker's intent. This includes grammatical analysis and natural language processing to understand the context and infer the speaker's intent.

[0054] Visual Indication

[0055] The server sends the completed text data to the device, which then displays it visually, highlighting important keywords and phrases to make it easier for the user to understand.

[0056] 3. Specific Examples

[0057] For example, consider the case where a user calls customer support to inquire about an internet connection problem. The server receives the voice data "The internet is not connected," and the speech recognition module converts it into text data "The internet is not connected." The generation AI analyzes this text data and completes it to reflect the user's intent: "I want to solve my internet connection problem." This completed text data is sent to the device, and the user can visually confirm the text "I want to solve my internet connection problem."

[0058] This system allows users to visually confirm content that is difficult to understand through audio alone, making telephone communication smoother.

[0059] The processing flow will be explained below.

[0060] Step 1:

[0061] The server receives the voice data from the phone, converts it into a digital format, and stores it in a buffer. At this stage, processing is performed to minimize degradation of the voice data.

[0062] Step 2:

[0063] The server transfers the buffered voice data to the voice recognition module in real time, ensuring efficient data transfer to avoid delays.

[0064] Step 3:

[0065] The speech recognition module analyzes the received voice data and breaks it down into phonemes. The phonemes are organized into words and converted into text data. This text data is temporarily stored in the speech recognition module.

[0066] Step 4:

[0067] The text data generated by the speech recognition module is sent to the analysis module, which receives the text data and begins analyzing the context. It uses grammatical analysis and natural language processing to analyze the structure and meaning of the sentence.

[0068] Step 5:

[0069] The generative AI (analysis module) understands the context of the text data and infers the speaker's intention. It derives a more accurate interpretation by referring to the intonation and stress of the voice and the history of past conversations. Any unclear parts are also filled in during this process.

[0070] Step 6:

[0071] The generative AI completes the inferred intent and corrects any ambiguities. The inferred results are formatted to make them natural and easy to understand. This text data is then generated as data for the final visual display.

[0072] Step 7:

[0073] The server then sends the completed text data to the device, formatted to make it easier for the user to read and highlighting important keywords and phrases.

[0074] Step 8:

[0075] The terminal receives the text data sent from the server and displays it visually to the user. The display screen is laid out in an easy-to-read format and is designed to allow the user to quickly grasp important information.

[0076] Step 9:

[0077] The user can read the text data displayed on the device to check the contents of the call. If necessary, they can make a request or ask again. This operation allows the user to obtain more detailed information.

[0078] In this way, the server, terminal, and user work together to analyze voice data in real time and provide visual assistance, making it easier to accurately understand the content of telephone conversations.

[0079] Example 1

[0080] 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."

[0081] Conventional telephone communication relies solely on the transmission of information through voice, which often leads to difficulty in hearing and misunderstandings. In particular, important content can be difficult to understand due to environmental noise or the speaker's unique accent or intonation. This makes telephone communication extremely difficult, especially for elderly people and those with hearing impairments. To solve this problem, a system that visually complements voice is needed.

[0082] 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.

[0083] In this invention, the server includes means for receiving voice data, means for sequentially supplying the voice data, and means for transmitting analyzed text data to the terminal. This enables the voice data to be analyzed quickly and efficiently, and helps the user's visual understanding. Also, by including means for converting the received voice data into text data, means for analyzing the converted text data and inferring the context and the speaker's intention, means for completing the inferred intention and correcting unclear parts, means for visually displaying the completed text data, and means for the terminal to highlight the text data, it becomes possible to comprehensively understand the voice data and quickly grasp important information.

[0084] "Voice data" means analog or digital audio signals obtained from telephone communications or other audio information.

[0085] A "means" is a device, system, software, or combination thereof used to achieve a particular purpose.

[0086] A "server" is a computer system that provides services to multiple terminals over a network.

[0087] "Serial supply" means sending data continuously in small amounts in order.

[0088] "Text data" refers to character string information generated from voice data through processing such as voice recognition.

[0089] "Analysis" means analyzing and interpreting the structure and meaning of data in order to understand its content.

[0090] "Context" refers to the circumstances and related information that form the background of a text or utterance.

[0091] The "speaker's intention" refers to the purpose, thoughts, and intentions behind what is said.

[0092] "Complement" means adding missing parts to make the whole complete.

[0093] An "ambiguous part" is one that is not clearly understood or is ambiguous.

[0094] "Visual display" means expressing information using visual elements such as letters and figures and displaying it on a display or monitor.

[0095] A "terminal" is a device such as a computer, smartphone, or tablet that a user uses to send, receive, and process information.

[0096] "Highlighting" refers to displaying specific information in a way that makes it stand out, such as by changing the color or font.

[0097] The present invention relates to a system that analyzes voice data and visually complements it to eliminate difficulties in hearing telephone conversations. This system is designed to convert voice data into text in real time, analyze the context and intent, and visualize the text. An embodiment of the system is described below in detail.

[0098] Hardware and Software Use

[0099] This system involves the cooperation of a server, a terminal, and a user. The server is a central device for receiving telephone voice data and processing it in real time. The server includes a means for receiving voice data and a means for buffering and sequentially supplying the voice data. For example, the server can be integrated with the Google Speech-to-Text API.

[0100] When the voice data is received by the server, it is converted into text data by a speech recognition module (such as the Google Speech-to-Text API). This text data is then analyzed by a generative AI (such as OpenAI's GPT-4). The generative AI analyzes the data to understand the context and the speaker's intent, and generates supplemental text data.

[0101] The completed text data is sent from the server to the terminal. The terminal is a device that allows the user to receive information visually, such as an iPad or a dedicated display. The terminal has the function of highlighting important keywords and phrases.

[0102] Specific examples

[0103] For example, consider a case where a user calls customer support to report a problem with their Internet connection. When the user says, "The Internet is down," the voice data is received by the server. This data is transferred to a speech recognition module and converted into text data that says, "The Internet is down."

[0104] This text data is analyzed by a generative AI to infer the context and user intent. The generative AI generates a completed text, such as "I want to solve my internet connection problem." The completed text data is sent from the server to the device and displayed on the device's display. The device highlights important keywords such as "internet" and "connection problem."

[0105] Prompt Sentence Examples

[0106] "A user is on the phone talking about an issue with their internet connection. Convert this audio data into text and infer what the user meant."

[0107] This system allows users to visually confirm information that is difficult to understand through voice alone, making telephone communication smoother and particularly useful for elderly people and those with hearing impairments.

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

[0109] Step 1:

[0110] The server receives the voice data and transfers it to the voice recognition module in real time.

[0111] Specifically, the server receives the voice data (input) spoken by the user over the communication line. The received voice data is temporarily stored in buffer memory. The data is then sequentially supplied from the buffer to the Google Speech-to-Text API (output).

[0112] Step 2:

[0113] The voice recognition module converts the received voice data into text data.

[0114] Specifically, the Google Speech-to-Text API performs phonemic analysis on the voice data (input) transferred from the server and converts it into words, and the resulting text data (output) is sent to the analysis module.

[0115] Step 3:

[0116] Generative AI analyzes text data and infers the context and speaker's intention.

[0117] Specifically, the server receives text data (input) from the Google Speech-to-Text API and sends it to OpenAI's GPT-4. GPT-4 uses grammar analysis and natural language processing techniques to understand the context of the text and infer the speaker's intent. During this process, relevant keywords and phrases are identified and a complementary text (output) is generated that clearly indicates the speaker's intent.

[0118] Step 4:

[0119] The server transmits the completed text data to the terminal.

[0120] Specifically, the server sends the completed text data (input) generated by GPT-4 to the device. This includes a mechanism for safely and reliably transferring data using a communication protocol. The completed text data (output) arrives at the device.

[0121] Step 5:

[0122] The terminal displays the completed text data.

[0123] Specifically, the device (e.g., iPad) receives the completed text data (input) sent from the server and displays it on the display. The device highlights important keywords and phrases so that the user can easily confirm them visually. The user then checks this highlighted completed text (output).

[0124] Step 6:

[0125] The user checks the text data displayed on the screen and makes requests or re-inquires as necessary.

[0126] Specifically, the user reads and understands the text data (input) displayed on the device's display. If necessary, the user can obtain more information by issuing a follow-up inquiry or request through the device. This interaction generates new voice data (output), and the system operates again.

[0127] (Application example 1)

[0128] 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."

[0129] Recent voice communication has the problem that the accuracy of information transmission is easily reduced, especially in noisy environments or when the speaker is difficult to hear. There is also a risk that users may miss important information while answering the phone. Furthermore, in customer support and customer service at physical stores, staff are required to accurately understand the content of phone calls in real time. To solve these issues, it is necessary to develop a system that visually complements voice data and allows users to obtain information efficiently.

[0130] 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.

[0131] In this invention, the server includes means for receiving voice data, means for converting the received voice data into text data, means for analyzing the converted text data and inferring the context and the speaker's intent, means for completing the inferred intent and correcting unclear parts, means for visually displaying the completed text data, and means for using a display device of a personal digital assistant or portable terminal as display means to allow the user to visually confirm the information. This improves the accuracy of voice communication, allowing the user to efficiently obtain important information without missing it. Furthermore, in customer support and customer service at physical stores, staff can visually understand the content of phone calls, enabling them to respond quickly and accurately.

[0132] "Audio data" refers to audio signal information captured through a microphone or other audio collection device.

[0133] "Text data" is character information converted from voice data using voice recognition technology.

[0134] "Context" is the context or background information needed to understand the meaning of spoken words or sentences.

[0135] "Speaker intention" is the main message or purpose that the speaker is trying to convey.

[0136] "Completion" means making something more complete by adding unclear or missing information.

[0137] An "unclear portion" is a portion of the audio data or text data that is unclear due to noise or difficulty in hearing.

[0138] "Visual display means" refers to a device or technology that uses a screen or display to display text data in a form that is visually recognizable to the user.

[0139] A "personal digital assistant" is a portable electronic device such as a smartphone or tablet.

[0140] A "portable terminal" is an electronic device that can be worn by a user, such as a glasses-type display or a smart watch.

[0141] A "display device" is a hardware device that can visually display characters, images, etc.

[0142] This invention relates to a "visual telephone" system that analyzes voice data and provides visual supplementation. This system converts voice data into text data in real time, analyzes it using a generative AI, infers the speaker's intention and context, and provides visual supplementation. Specific embodiments of this system are described below.

[0143] System configuration

[0144] server

[0145] The server is the central part of the system and includes the following functions:

[0146] 1. Means for receiving voice data: The server receives voice data from the telephone. This reception is performed in real time.

[0147] 2. Means for converting voice data to text data: The received voice data is converted to text data using a voice recognition module. Specifically, the Python SpeechRecognition library is used.

[0148] 3. Means of inferring context and speaker intent: Generative AI analyzes text data to understand the context and speaker intent. This process uses the Hugging Face transformers library.

[0149] 4. Means of completing inferred intent and correcting ambiguities: Complete inferred intent and make corrections as necessary.

[0150] 5. Means for visually displaying the completed text data: The completed text data is transmitted to the user's personal digital assistant or portable terminal.

[0151] Terminal

[0152] A terminal is a device through which a user receives visual information. It has the following functions:

[0153] 1. Display device: A display device such as a smartphone or smart glasses visually displays the text data sent from the server, highlighting important keywords and phrases.

[0154] User

[0155] Users are the main actors in communication via telephone and receive visually supplemented information through their devices. For example, customer support staff at a brick-and-mortar store can wear smart glasses and understand customer inquiries in real time.

[0156] Program processing and specific examples

[0157] The server receives the voice data and converts it into text data using a speech recognition module. The converted text data is then analyzed by a generative AI to infer the speaker's intent and context. This inferred intent and context is then augmented into a format that is easy to display visually and sent to the user's device.

[0158] Hardware and software used

[0159] Server: Receives voice data, converts it into text, and generates AI analysis

[0160] Software: SpeechRecognition, transformers (natural language processing)

[0161] Devices: smart glasses, smartphone

[0162] Display device: smart glasses display

[0163] Specific examples

[0164] Let's take the example of a user (a customer support staff member at a brick-and-mortar store) receiving a phone call from a customer asking, "Please check if the product is in stock." At this time, the server receives the voice data and converts it into text data saying, "Please check if the product is in stock." The generation AI analyzes this text data, infers that the product's stock information needs to be checked, and displays this information on the smart glasses' display.

[0165] Example prompt sentence:

[0166] I want to check if the product is in stock

[0167] This example allows the user (customer support staff) to immediately take appropriate action based on the visually supplemented information. This system eliminates ambiguity in voice data and enables efficient information transmission.

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

[0169] Step 1:

[0170] The server receives the voice data. Specifically, the server captures the telephone voice signal in real time and stores it in a buffer as voice data. The input is the telephone voice signal, and the output is the voice data stored in the buffer. This prepares the server to pass the received voice data to the next process.

[0171] Step 2:

[0172] The server converts the voice data into text data. Specifically, it uses a speech recognition module (SpeechRecognition library) to convert the voice data stored in the buffer into text data. The input is the voice data stored in the buffer, and the output is the converted text data. This data processing involves phonemic analysis and conversion into words.

[0173] Step 3:

[0174] The server uses a generative AI model to analyze text data and infer the context and speaker intent. Specifically, the text data is sent to a natural language processing module (the transformers library) to analyze the context and intent. The input is the transformed text data, and the output is information about the analyzed intent and context. This data processing includes grammatical analysis and context understanding.

[0175] Step 4:

[0176] The server completes the inferred intent and corrects any unclear parts. Specifically, based on the analysis results of the generative AI, any unclear parts in the utterance are corrected and the final intent is determined. The input is the analyzed intent and context information, and the output is the completed text data. A completion algorithm is used in this process.

[0177] Step 5:

[0178] The server sends the completed text data to the device. Specifically, the completed text data is sent to the device (smart glasses or smartphone) via a network. The input is the completed text data, and the output is the state after transmission to the device is complete. Network communication technology is used for this data transfer.

[0179] Step 6:

[0180] The terminal visually displays the received text completion data. Specifically, the received text completion data is displayed on the display of smart glasses or a smartphone, highlighting important keywords and phrases. The input is the text completion data sent from the server, and the output is the visual information displayed on the display.

[0181] Step 7:

[0182] The user checks the visual information and responds accordingly. Specifically, the user (customer support staff) looks at the information displayed on the smart glasses display, accurately understands the content of the call, and responds appropriately to the customer. The input is the visual information displayed on the display, and the output is the user's appropriate response.

[0183] 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.

[0184] This invention relates to a "visual telephone" system that analyzes telephone voices and visually complements them to eliminate difficulties in hearing calls. Furthermore, this invention enables more intuitive and effective communication by incorporating an emotion engine that recognizes the user's emotions. This system operates in cooperation with the server, terminal, and user, analyzing voice data and displaying it visually in real time.

[0185] 1. System Configuration

[0186] server

[0187] The server is the central device that receives telephone voice data and processes it in real time. The server first receives the voice data and sequentially transfers it to the voice recognition module. The voice recognition module converts the voice data into text data. This converted text data is analyzed by the generative AI to infer the speaker's intention and context. In addition, the emotion engine analyzes the user's tone and intonation of voice to identify the user's emotions. Finally, the inferred intention and user emotion are complemented and converted into a format that is easy to display visually.

[0188] Terminal

[0189] The terminal is a device that allows the user to receive information visually. It receives the completed text data sent from the server and displays it on the screen. The terminal is made up of a television or a dedicated display, and has the function of highlighting important keywords and phrases. The user looks at this display to understand the content of the call.

[0190] User

[0191] The user is the subject of communication via telephone, and receives visually supplemented information through the terminal. The user can read the displayed text data and make requests or follow-up inquiries as needed. This allows the user to visually confirm content that is difficult to understand through voice alone, and also obtain information that reflects emotions.

[0192] 2. Program Processing

[0193] Audio Receiving and Forwarding

[0194] The server receives the voice data and transfers it to the voice recognition module in real time. The server stores the voice data in a buffer and supplies the data to the voice recognition module as needed.

[0195] Speech-to-text conversion

[0196] The speech recognition module converts the received voice data into text data in real time, including phonemic analysis of the voice and converting it into words, and the generated text data is sent to the analysis module.

[0197] Contextual analysis and intent inference

[0198] The generative AI (analysis module) analyzes text data and performs analysis to understand the context and the speaker's intent. This includes grammatical analysis and natural language processing to understand the context and infer the speaker's intent.

[0199] emotion recognition

[0200] The emotion engine analyzes the tone and intonation of the user's voice data to recognize the user's emotions. The emotion engine then classifies emotions using an emotion dictionary and acoustic model, and provides the identification results to the generation AI.

[0201] Visual Indication

[0202] The server sends the completed text data and emotion recognition results to the device, which then displays them visually. Important keywords and phrases are highlighted to make them easier for users to understand. Additionally, the user's emotions are also displayed, supporting a more intuitive understanding.

[0203] 3. Specific Examples

[0204] For example, consider the case where a user calls customer support to inquire about an internet connection problem. The server receives the voice data "The internet is not connected," and the speech recognition module converts it into text data "The internet is not connected." The generation AI analyzes this text data and complements it with the user's intent, "I want to solve my internet connection problem." The emotion engine identifies urgency and anxiety from the user's tone of voice and reflects this in the display. This complemented text data is sent to the device, and the user can visually confirm emotions such as "urgency" and "anxiety" along with the text "I want to solve my internet connection problem."

[0205] This system allows users to visually confirm content that is difficult to understand through audio alone, making telephone communication smoother and providing comprehensive information that takes emotions into consideration.

[0206] The processing flow will be explained below.

[0207] Step 1:

[0208] The server receives the voice data from the call. The voice data is captured digitally and stored in a buffer, where it is filtered to preserve the quality of the voice data.

[0209] Step 2:

[0210] The server transfers the buffered voice data to the speech recognition module in real time, using streaming technology to minimize data delays.

[0211] Step 3:

[0212] The speech recognition module analyzes the received speech data and breaks it down into phonemes, converts the phonemes into words, and organizes the generated words into text data.

[0213] Step 4:

[0214] The text data generated by the speech recognition module is sent to the analysis module, which receives the text data and analyzes the context of the text data using grammar analysis and natural language processing.

[0215] Step 5:

[0216] The generative AI (analysis module) understands the context of the text data and infers the speaker's intention. Based on the analysis results, it further examines the inferred intention and fills in any unclear parts.

[0217] Step 6:

[0218] The emotion engine analyzes the voice data received from the server and identifies emotions based on the tone and intonation of the user's voice. The emotion engine performs emotion classification using an acoustic model and an emotion dictionary.

[0219] Step 7:

[0220] The emotion engine passes the identified emotion information to the generation AI, which then complements the emotion information along with the inferred intent. This allows the user's emotions to be reflected in the text data.

[0221] Step 8:

[0222] The server sends the completed text data and emotion information to the device. The data is sent in a format that is easy for the user to understand, and is formatted to emphasize important keywords and emotions.

[0223] Step 9:

[0224] The terminal receives the data sent from the server and visually displays it to the user. The display screen combines the text data with the user's emotions, highlighting important keywords and emotions.

[0225] Step 10:

[0226] The user can read the text data and emotional information displayed on the device and check the contents of the call. If necessary, the user can make a request or ask again. This operation allows for more detailed information to be obtained.

[0227] Example 2

[0228] 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."

[0229] In conventional telephone communication, voice alone can make the content unclear, and the information conveyed can be inaccurate, especially in noisy environments or when the voice quality is poor. It is also difficult to accurately understand emotions and intentions through voice alone, making misunderstandings highly likely, especially in communication that involves emotions. This can lead to problems that reduce user satisfaction, especially in user support and customer service.

[0230] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving voice data, means for converting the received voice data into text data, means for analyzing the converted text data and inferring the context and the speaker's intention, means for analyzing the speaker's tone and intonation to identify the speaker's emotion, means for complementing the inferred intention and the identified emotion and correcting unclear parts, and means for visually displaying the complemented text data and the identified emotion. As a result, by converting voice data into text data in real time and performing context analysis and emotion identification, it becomes possible to visually complement information that is difficult to convey through voice alone.

[0231] The "means for receiving voice data" is a function that receives voice data in real time from a telephone line, VoIP, etc., and temporarily stores it in a buffer.

[0232] The "means for converting into text data" is a function that uses a voice recognition module to analyze voice data, perform phonemic analysis, and convert into words to generate text data.

[0233] "Means of inferring context and speaker intent" refers to a function that uses generative AI to analyze text data and understand the speaker's intent and context through grammatical analysis and natural language processing.

[0234] "Means for identifying emotions by analyzing voice tone and intonation" is a function that uses an emotion engine to analyze the intonation and strength of the voice in the audio data, and classify and identify emotions.

[0235] "Means to complement inferred intent and identified emotion and correct unclear parts" is a function that uses the analysis results obtained by generative AI and emotion engines to complement intent and emotion and make appropriate corrections to clarify unclear parts.

[0236] The "means for visually displaying the complemented text data and the identified emotion" is a function that uses the display device of the terminal to display the text data and the identified emotion on the screen in a form that is easy for the user to check.

[0237] This invention relates to a "visual telephone" system that eliminates the difficulty of hearing telephone calls by analyzing voice data and visually complementing it. This system analyzes voice data, infers the user's intentions and emotions, and visually displays them, enabling more intuitive and effective communication.

[0238] The specific system configuration mainly involves the collaboration of a server, a terminal, and a user.

[0239] server

[0240] The server is the central device that receives and processes telephone voice data in real time. First, the server buffers the voice data from the telephone line or VoIP. Next, it converts the voice data into text using a speech recognition module (e.g., Google Speech-to-Text API). The converted text data is analyzed by a generative AI (e.g., OpenAI GPT-4) to understand the speaker's intent and context. It also uses an emotion engine (e.g., Affectiva SDK) to analyze the user's tone and intonation to identify emotions. The server then integrates the inferred intent and identified emotions and presents them in a format that is easy to display visually.

[0241] Terminal

[0242] The terminal is a device that allows the user to visually receive information. It receives the supplemented text data and emotion recognition results sent from the server and displays them on the terminal's display device (e.g., tablet, PC, smartphone). This display also includes a function to highlight important keywords and phrases, making it easy for the user to grasp the information.

[0243] User

[0244] The user is the subject of communication via telephone, and receives visually supplemented information through the terminal. The user can read the displayed text data and make requests or follow-up inquiries as needed. This allows the user to visually confirm content that is difficult to understand through voice alone, and also obtain information that reflects emotions.

[0245] Specific examples

[0246] For example, consider the case where a user calls customer support to inquire about an internet connection problem. In this case, the server receives the voice data "The internet is not connecting," and the speech recognition module converts it into text data "The internet is not connecting." The generation AI analyzes this text data and infers the user's intention, which is "I want to solve my internet connection problem." Furthermore, the emotion engine identifies "urgency" and "anxiety" from the user's tone of voice and reflects these in the display. This complemented text data is sent to the device, and the user can visually confirm emotions such as "urgency" and "anxiety" along with the text data "I want to solve my internet connection problem."

[0247] Prompt Sentence Examples

[0248] User: "I can't connect to the internet, what should I do?"

[0249] Generation AI:

[0250] 1. Analyze user comments and convert them into text.

[0251] 2. Infer the user's intent from the text (e.g., I want to fix my internet connection problem).

[0252] 3. Identify urgency and anxiety from the user's tone of voice.

[0253] 4. The final results are compiled into a text and sent to the device.

[0254] This system allows users to visually confirm content that is difficult to understand through audio alone, making telephone communication smoother and providing comprehensive information that takes emotions into consideration.

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

[0256] Step 1:

[0257] Voice data reception

[0258] The server receives voice data in real time from a telephone line or VoIP.

[0259] Input: Voice data from telephone line or VoIP

[0260] Output: Audio data stored in the server's buffer

[0261] Specific operation: The server receives audio data via the API and stores it in a buffer.

[0262] Step 2:

[0263] Converting audio data to text

[0264] The server converts the voice data into text data using a voice recognition module.

[0265] Input: Audio data stored in the server's buffer

[0266] Output: Text data

[0267] Specific operation: The server inputs the voice data into a speech recognition API (e.g., Google Speech-to-Text API) and obtains the text data "Internet connection is down."

[0268] Step 3:

[0269] Contextual analysis and intent inference of text data

[0270] The server uses generative AI to analyze the text data and infer the context and speaker's intent.

[0271] Input: Text data obtained from the speech recognition module

[0272] Output: Inferred intent and contextually-based complementation data

[0273] Specific operation: The server inputs the text data into a generation AI (e.g., GPT-4) and obtains the intent "I want to solve the Internet connection problem" as the analysis result.

[0274] Step 4:

[0275] emotion recognition

[0276] The server uses an emotion engine to analyze the tone and intonation of the voice based on the voice data and identify the user's emotions.

[0277] Input: Audio data stored on the server

[0278] Output: Identified emotion data (e.g., urgency, anxiety)

[0279] Specific operation: The server uses an emotion engine (e.g., Affectiva SDK) to identify emotions such as "anxiety" from the voice data.

[0280] Step 5:

[0281] Visual display of complementary data

[0282] The server transmits the completed text data and the identified emotion data to the terminal, which then visually displays them.

[0283] Input: Complemented text data, identified emotion data

[0284] Output: Visually displayed data (text and sentiment)

[0285] Specific operation: The server sends the data to the terminal, and the terminal displays the text "I want to solve the problem of the Internet connection" and the emotion "anxiety" highlighted on the display device (e.g., tablet, PC).

[0286] The above is the specific processing flow of the program of this system.

[0287] (Application example 2)

[0288] 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."

[0289] In current telephone communications, it can be difficult to understand what is being said through voice alone. Furthermore, the user's emotions and intentions may not be conveyed accurately. Smooth communication is especially important in brick-and-mortar stores and customer support settings. However, existing systems lack a means to visually complement emotions and intentions, which reduces the efficiency and effectiveness of communication.

[0290] 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 receiving voice data, means for converting the received voice data into text data, means for analyzing the converted text data and inferring the context and the speaker's intention, means for completing the inferred intention and correcting unclear parts, means for analyzing the tone and intonation of the user's voice using an emotion engine to recognize emotions, and means for integrating the results of the emotion analysis into visually displayed data and displaying the results. This not only makes it possible to visually grasp content that is difficult to understand from voice alone, but also makes it possible to provide information that reflects the user's emotions, thereby enabling smooth communication between customers and staff, particularly in customer support situations in physical stores.

[0291] The "audio data receiving means" is a part of the system that has the function of receiving audio data transmitted via voice calls and the like.

[0292] The "text conversion means" is a part of the system that has the function of analyzing received voice data and converting it into text data.

[0293] The "intention inference means" is a part of the system that has the function of analyzing and understanding the context and the speaker's intention based on the converted text data.

[0294] The "intention completion means" is a part of the system that has the function of completing the speaker's inferred intention and correcting any unclear parts.

[0295] The "visual display means" is a part of the system that has the function of visually displaying the analyzed text data and emotion recognition results.

[0296] The "emotion engine" is a part of the system that has the ability to recognize the user's emotions by analyzing the tone and intonation of the voice.

[0297] The "emotion analysis result display means" is a part of the system that has the function of visually displaying the results of the emotion analysis by the emotion engine.

[0298] This invention relates to a customer support system for brick-and-mortar stores. This system receives and analyzes voice data and delivers information visually to users, thereby realizing smooth communication.

[0299] A specific example of the system is a customer support robot in a brick-and-mortar store that receives voice data from customers in real time, converts it into text, and then analyzes and visually displays the customer's intentions and emotions.

[0300] The server first receives the voice data. Using a voice data receiving means, it receives voice data transmitted via telephone or voice call in real time. Next, it converts the received voice data into text data using a text conversion means. For example, it uses a speech recognition library (speech_recognition). The converted text data is analyzed using an intent inference means and an intent completion means to infer the context and the speaker's intent. This process uses a generative AI model and natural language processing technology (TextBlob).

[0301] Furthermore, an emotion engine is used to analyze the tone and intonation of the voice to recognize the user's emotions. Specifically, an emotion analysis library (e.g., emotion dictionary and acoustic model) is used. This makes it possible to visually display the emotion analysis result together with the text data using an emotion analysis result display means. In-store displays and the screens of support robots are used as visual display means. The analyzed text data and the emotion results are displayed integrated on this screen.

[0302] As a concrete example, consider a scenario in which a customer asks a robot about a product. In this case, the server receives the voice data "Do you have this product in stock?" and converts it into text data "Do you have this product in stock?" using a speech recognition library. Next, a generative AI model analyzes this text data and infers the customer's intent in the form of "Check product availability." The emotion engine identifies urgency or anxiety from the customer's tone of voice and reflects this in the display.

[0303] Finally, as a visual display, emotions such as "urgent" or "anxiety" are displayed on in-store displays and on the screens of support robots along with the text "Check product inventory." This allows customers to visually confirm content that would be difficult to understand through voice alone, and enables store staff to accurately grasp the customer's intentions and emotions.

[0304] Examples of input prompts for generative AI models include:

[0305] "A customer is asking, 'Does the store have this item in stock?' Visually display this question in real time and use an emotion recognition engine to read the customer's emotions (urgency, anxiety, etc.)."

[0306] Using these prompts, the system visually displays customer questions based on text and emotion recognition, allowing store staff to accurately understand the customer's needs.

[0307] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0308] Step 1:

[0309] The server receives audio data sent via telephone or voice call. The input is audio data from a microphone or a call system. The output is to store the received audio data in a buffer. Specifically, the audio receiving module converts audio into digital data and processes it in a streaming format.

[0310] Step 2:

[0311] The server converts voice data into text data in real time. The input is the voice data stored in the buffer. The output is the converted text data. Specifically, it uses a speech recognition library (e.g., speech_recognition) to analyze the voice data into phonemes and output them as text.

[0312] Step 3:

[0313] The server analyzes the text data and infers the context and the speaker's intent. The input is the converted text data. The output is the inferred intent and its analysis results. Specific operations include using generative AI models and natural language processing techniques (e.g., TextBlob) to analyze the text grammatically and semantically.

[0314] Step 4:

[0315] The server completes the inferred intent and corrects any ambiguities. The input is the inferred intent and the text data. The output is the completed and corrected text data. Specifically, the analysis module reevaluates the context and intent and adds additional information as needed.

[0316] Step 5:

[0317] The server uses an emotion engine to analyze the emotions in the voice data. The input is the voice data and its text conversion results. The output is the emotion recognition results. Specifically, it analyzes the tone and intonation of the voice and classifies the emotion using an emotion dictionary and acoustic model.

[0318] Step 6:

[0319] The server sends the completed text data and the emotion analysis results to the terminal for visual display. The input is the completed text data and the emotion analysis results. The output is data for visual display that includes these data. Specifically, the data formatting module organizes this information and formats it into a format that is easy to display.

[0320] Step 7:

[0321] The terminal receives the data sent from the server and displays it visually. The input is the data for visual display sent from the server. The output is the text data and sentiment analysis results displayed on the display. Specifically, the terminal's display module highlights important keywords and phrases, providing information in a format that is easy for the user to review.

[0322] Step 8:

[0323] The user checks the visually displayed information and makes requests or re-inquires as necessary. The input is the text data displayed on the screen and the results of sentiment analysis. The output is the user's understanding and the next action. In concrete terms, the user checks the display and smoothly proceeds with the response via phone or chat.

[0324] 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.

[0325] 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.

[0326] 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.

[0327] [Second embodiment]

[0328] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0329] 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.

[0330] 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).

[0331] 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.

[0332] 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.

[0333] 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).

[0334] 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.

[0335] 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.

[0336] 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.

[0337] 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.

[0338] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0339] 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."

[0340] This invention relates to a "visual telephone" system that analyzes telephone voices and visually complements them to eliminate the difficulty of hearing telephone calls. This system operates in cooperation with the server, terminals, and users, analyzing voice data and displaying it visually in real time.

[0341] 1. System Configuration

[0342] server

[0343] The server is the central device that receives the voice data from the phone and processes it in real time. The server first receives the voice data and sequentially transfers it to the voice recognition module. The voice recognition module converts the voice data into text data. This converted text data is analyzed by the generative AI to infer the speaker's intent and context. Finally, the inferred intent is complemented and corrected into a format that is easy to display visually.

[0344] Terminal

[0345] The terminal is a device that allows the user to receive information visually. It receives the completed text data sent from the server and displays it on the screen. The terminal is made up of a television or a dedicated display, and has the function of highlighting important keywords and phrases. The user looks at this display to understand the content of the call.

[0346] User

[0347] The user is the subject of communication via telephone, and receives visually supplemented information through the terminal. The user can read the displayed text data and make requests or follow-up inquiries as needed. This allows the user to visually confirm content that is difficult to understand through voice alone.

[0348] 2. Program Processing

[0349] Audio Receiving and Forwarding

[0350] The server receives the voice data and transfers it to the voice recognition module in real time. The server stores the voice data in a buffer and supplies the data to the voice recognition module as needed.

[0351] Speech-to-text conversion

[0352] The speech recognition module converts the received voice data into text data in real time, including phonemic analysis of the voice and converting it into words, and the generated text data is sent to the analysis module.

[0353] Contextual analysis and intent inference

[0354] The generative AI (analysis module) analyzes text data and performs analysis to understand the context and the speaker's intent. This includes grammatical analysis and natural language processing to understand the context and infer the speaker's intent.

[0355] Visual Indication

[0356] The server sends the completed text data to the device, which then displays it visually, highlighting important keywords and phrases to make it easier for the user to understand.

[0357] 3. Specific Examples

[0358] For example, consider the case where a user calls customer support to inquire about an internet connection problem. The server receives the voice data "The internet is not connected," and the speech recognition module converts it into text data "The internet is not connected." The generation AI analyzes this text data and completes it to reflect the user's intent: "I want to solve my internet connection problem." This completed text data is sent to the device, and the user can visually confirm the text "I want to solve my internet connection problem."

[0359] This system allows users to visually confirm content that is difficult to understand through audio alone, making telephone communication smoother.

[0360] The processing flow will be explained below.

[0361] Step 1:

[0362] The server receives the voice data from the phone, converts it into a digital format, and stores it in a buffer. At this stage, processing is performed to minimize degradation of the voice data.

[0363] Step 2:

[0364] The server transfers the buffered voice data to the voice recognition module in real time, ensuring efficient data transfer to avoid delays.

[0365] Step 3:

[0366] The speech recognition module analyzes the received voice data and breaks it down into phonemes. The phonemes are organized into words and converted into text data. This text data is temporarily stored in the speech recognition module.

[0367] Step 4:

[0368] The text data generated by the speech recognition module is sent to the analysis module, which receives the text data and begins analyzing the context. It uses grammatical analysis and natural language processing to analyze the structure and meaning of the sentence.

[0369] Step 5:

[0370] The generative AI (analysis module) understands the context of the text data and infers the speaker's intention. It derives a more accurate interpretation by referring to the intonation and stress of the voice and the history of past conversations. Any unclear parts are also filled in during this process.

[0371] Step 6:

[0372] The generative AI completes the inferred intent and corrects any ambiguities. The inferred results are formatted to make them natural and easy to understand. This text data is then generated as data for the final visual display.

[0373] Step 7:

[0374] The server then sends the completed text data to the device, formatted to make it easier for the user to read and highlighting important keywords and phrases.

[0375] Step 8:

[0376] The terminal receives the text data sent from the server and displays it visually to the user. The display screen is laid out in an easy-to-read format and is designed to allow the user to quickly grasp important information.

[0377] Step 9:

[0378] The user can read the text data displayed on the device to check the contents of the call. If necessary, they can make a request or ask again. This operation allows the user to obtain more detailed information.

[0379] In this way, the server, terminal, and user work together to analyze voice data in real time and provide visual assistance, making it easier to accurately understand the content of telephone conversations.

[0380] Example 1

[0381] 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."

[0382] Conventional telephone communication relies solely on the transmission of information through voice, which often leads to difficulty in hearing and misunderstandings. In particular, important content can be difficult to understand due to environmental noise or the speaker's unique accent or intonation. This makes telephone communication extremely difficult, especially for elderly people and those with hearing impairments. To solve this problem, a system that visually complements voice is needed.

[0383] 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.

[0384] In this invention, the server includes means for receiving voice data, means for sequentially supplying the voice data, and means for transmitting analyzed text data to the terminal. This enables the voice data to be analyzed quickly and efficiently, and helps the user's visual understanding. Also, by including means for converting the received voice data into text data, means for analyzing the converted text data and inferring the context and the speaker's intention, means for completing the inferred intention and correcting unclear parts, means for visually displaying the completed text data, and means for the terminal to highlight the text data, it becomes possible to comprehensively understand the voice data and quickly grasp important information.

[0385] "Voice data" means analog or digital audio signals obtained from telephone communications or other audio information.

[0386] A "means" is a device, system, software, or combination thereof used to achieve a particular purpose.

[0387] A "server" is a computer system that provides services to multiple terminals over a network.

[0388] "Serial supply" means sending data continuously in small amounts in order.

[0389] "Text data" refers to character string information generated from voice data through processing such as voice recognition.

[0390] "Analysis" means analyzing and interpreting the structure and meaning of data in order to understand its content.

[0391] "Context" refers to the circumstances and related information that form the background of a text or utterance.

[0392] The "speaker's intention" refers to the purpose, thoughts, and intentions behind what is said.

[0393] "Complement" means adding missing parts to make the whole complete.

[0394] An "ambiguous part" is one that is not clearly understood or is ambiguous.

[0395] "Visual display" means expressing information using visual elements such as letters and figures and displaying it on a display or monitor.

[0396] A "terminal" is a device such as a computer, smartphone, or tablet that a user uses to send, receive, and process information.

[0397] "Highlighting" refers to displaying specific information in a way that makes it stand out, such as by changing the color or font.

[0398] The present invention relates to a system that analyzes voice data and visually complements it to eliminate difficulties in hearing telephone conversations. This system is designed to convert voice data into text in real time, analyze the context and intent, and visualize the text. An embodiment of the system is described below in detail.

[0399] Hardware and Software Use

[0400] This system involves the cooperation of a server, a terminal, and a user. The server is a central device for receiving telephone voice data and processing it in real time. The server includes a means for receiving voice data and a means for buffering and sequentially supplying the voice data. For example, the server can be integrated with the Google Speech-to-Text API.

[0401] When the voice data is received by the server, it is converted into text data by a speech recognition module (such as the Google Speech-to-Text API). This text data is then analyzed by a generative AI (such as OpenAI's GPT-4). The generative AI analyzes the data to understand the context and the speaker's intent, and generates supplemental text data.

[0402] The completed text data is sent from the server to the terminal. The terminal is a device that allows the user to receive information visually, such as an iPad or a dedicated display. The terminal has the function of highlighting important keywords and phrases.

[0403] Specific examples

[0404] For example, consider a case where a user calls customer support to report a problem with their Internet connection. When the user says, "The Internet is down," the voice data is received by the server. This data is transferred to a speech recognition module and converted into text data that says, "The Internet is down."

[0405] This text data is analyzed by a generative AI to infer the context and user intent. The generative AI generates a completed text, such as "I want to solve my internet connection problem." The completed text data is sent from the server to the device and displayed on the device's display. The device highlights important keywords such as "internet" and "connection problem."

[0406] Prompt Sentence Examples

[0407] "A user is on the phone talking about an issue with their internet connection. Convert this audio data into text and infer what the user meant."

[0408] This system allows users to visually confirm information that is difficult to understand through voice alone, making telephone communication smoother and particularly useful for elderly people and those with hearing impairments.

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

[0410] Step 1:

[0411] The server receives the voice data and transfers it to the voice recognition module in real time.

[0412] Specifically, the server receives the voice data (input) spoken by the user over the communication line. The received voice data is temporarily stored in buffer memory. The data is then sequentially supplied from the buffer to the Google Speech-to-Text API (output).

[0413] Step 2:

[0414] The voice recognition module converts the received voice data into text data.

[0415] Specifically, the Google Speech-to-Text API performs phonemic analysis on the voice data (input) transferred from the server and converts it into words, and the resulting text data (output) is sent to the analysis module.

[0416] Step 3:

[0417] Generative AI analyzes text data and infers the context and speaker's intention.

[0418] Specifically, the server receives text data (input) from the Google Speech-to-Text API and sends it to OpenAI's GPT-4. GPT-4 uses grammar analysis and natural language processing techniques to understand the context of the text and infer the speaker's intent. During this process, relevant keywords and phrases are identified and a complementary text (output) is generated that clearly indicates the speaker's intent.

[0419] Step 4:

[0420] The server transmits the completed text data to the terminal.

[0421] Specifically, the server sends the completed text data (input) generated by GPT-4 to the device. This includes a mechanism for safely and reliably transferring data using a communication protocol. The completed text data (output) arrives at the device.

[0422] Step 5:

[0423] The terminal displays the completed text data.

[0424] Specifically, the device (e.g., iPad) receives the completed text data (input) sent from the server and displays it on the display. The device highlights important keywords and phrases so that the user can easily confirm them visually. The user then checks this highlighted completed text (output).

[0425] Step 6:

[0426] The user checks the text data displayed on the screen and makes requests or re-inquires as necessary.

[0427] Specifically, the user reads and understands the text data (input) displayed on the device's display. If necessary, the user can obtain more information by issuing a follow-up inquiry or request through the device. This interaction generates new voice data (output), and the system operates again.

[0428] (Application example 1)

[0429] 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."

[0430] Recent voice communication has the problem that the accuracy of information transmission is easily reduced, especially in noisy environments or when the speaker is difficult to hear. There is also a risk that users may miss important information while answering the phone. Furthermore, in customer support and customer service at physical stores, staff are required to accurately understand the content of phone calls in real time. To solve these issues, it is necessary to develop a system that visually complements voice data and allows users to obtain information efficiently.

[0431] 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.

[0432] In this invention, the server includes means for receiving voice data, means for converting the received voice data into text data, means for analyzing the converted text data and inferring the context and the speaker's intent, means for completing the inferred intent and correcting unclear parts, means for visually displaying the completed text data, and means for using a display device of a personal digital assistant or portable terminal as display means to allow the user to visually confirm the information. This improves the accuracy of voice communication, allowing the user to efficiently obtain important information without missing it. Furthermore, in customer support and customer service at physical stores, staff can visually understand the content of phone calls, enabling them to respond quickly and accurately.

[0433] "Audio data" refers to audio signal information captured through a microphone or other audio collection device.

[0434] "Text data" is character information converted from voice data using voice recognition technology.

[0435] "Context" is the context or background information needed to understand the meaning of spoken words or sentences.

[0436] "Speaker intention" is the main message or purpose that the speaker is trying to convey.

[0437] "Completion" means making something more complete by adding unclear or missing information.

[0438] An "unclear portion" is a portion of the audio data or text data that is unclear due to noise or difficulty in hearing.

[0439] "Visual display means" refers to a device or technology that uses a screen or display to display text data in a form that is visually recognizable to the user.

[0440] A "personal digital assistant" is a portable electronic device such as a smartphone or tablet.

[0441] A "portable terminal" is an electronic device that can be worn by a user, such as a glasses-type display or a smart watch.

[0442] A "display device" is a hardware device that can visually display characters, images, etc.

[0443] This invention relates to a "visual telephone" system that analyzes voice data and provides visual supplementation. This system converts voice data into text data in real time, analyzes it using a generative AI, infers the speaker's intention and context, and provides visual supplementation. Specific embodiments of this system are described below.

[0444] System configuration

[0445] server

[0446] The server is the central part of the system and includes the following functions:

[0447] 1. Means for receiving voice data: The server receives voice data from the telephone. This reception is performed in real time.

[0448] 2. Means for converting voice data to text data: The received voice data is converted to text data using a voice recognition module. Specifically, the Python SpeechRecognition library is used.

[0449] 3. Means of inferring context and speaker intent: Generative AI analyzes text data to understand the context and speaker intent. This process uses the Hugging Face transformers library.

[0450] 4. Means of completing inferred intent and correcting ambiguities: Complete inferred intent and make corrections as necessary.

[0451] 5. Means for visually displaying the completed text data: The completed text data is transmitted to the user's personal digital assistant or portable terminal.

[0452] Terminal

[0453] A terminal is a device through which a user receives visual information. It has the following functions:

[0454] 1. Display device: A display device such as a smartphone or smart glasses visually displays the text data sent from the server, highlighting important keywords and phrases.

[0455] User

[0456] Users are the main actors in communication via telephone and receive visually supplemented information through their devices. For example, customer support staff at a brick-and-mortar store can wear smart glasses and understand customer inquiries in real time.

[0457] Program processing and specific examples

[0458] The server receives the voice data and converts it into text data using a speech recognition module. The converted text data is then analyzed by a generative AI to infer the speaker's intent and context. This inferred intent and context is then augmented into a format that is easy to display visually and sent to the user's device.

[0459] Hardware and software used

[0460] Server: Receives voice data, converts it into text, and generates AI analysis

[0461] Software: SpeechRecognition, transformers (natural language processing)

[0462] Devices: smart glasses, smartphone

[0463] Display device: smart glasses display

[0464] Specific examples

[0465] Let's take the example of a user (a customer support staff member at a brick-and-mortar store) receiving a phone call from a customer asking, "Please check if the product is in stock." At this time, the server receives the voice data and converts it into text data saying, "Please check if the product is in stock." The generation AI analyzes this text data, infers that the product's stock information needs to be checked, and displays this information on the smart glasses' display.

[0466] Example prompt sentence:

[0467] I want to check if the product is in stock

[0468] This example allows the user (customer support staff) to immediately take appropriate action based on the visually supplemented information. This system eliminates ambiguity in voice data and enables efficient information transmission.

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

[0470] Step 1:

[0471] The server receives the voice data. Specifically, the server captures the telephone voice signal in real time and stores it in a buffer as voice data. The input is the telephone voice signal, and the output is the voice data stored in the buffer. This prepares the server to pass the received voice data to the next process.

[0472] Step 2:

[0473] The server converts the voice data into text data. Specifically, it uses a speech recognition module (SpeechRecognition library) to convert the voice data stored in the buffer into text data. The input is the voice data stored in the buffer, and the output is the converted text data. This data processing involves phonemic analysis and conversion into words.

[0474] Step 3:

[0475] The server uses a generative AI model to analyze text data and infer the context and speaker intent. Specifically, the text data is sent to a natural language processing module (the transformers library) to analyze the context and intent. The input is the transformed text data, and the output is information about the analyzed intent and context. This data processing includes grammatical analysis and context understanding.

[0476] Step 4:

[0477] The server completes the inferred intent and corrects any unclear parts. Specifically, based on the analysis results of the generative AI, any unclear parts in the utterance are corrected and the final intent is determined. The input is the analyzed intent and context information, and the output is the completed text data. A completion algorithm is used in this process.

[0478] Step 5:

[0479] The server sends the completed text data to the device. Specifically, the completed text data is sent to the device (smart glasses or smartphone) via a network. The input is the completed text data, and the output is the state after transmission to the device is complete. Network communication technology is used for this data transfer.

[0480] Step 6:

[0481] The terminal visually displays the received text completion data. Specifically, the received text completion data is displayed on the display of smart glasses or a smartphone, highlighting important keywords and phrases. The input is the text completion data sent from the server, and the output is the visual information displayed on the display.

[0482] Step 7:

[0483] The user checks the visual information and responds accordingly. Specifically, the user (customer support staff) looks at the information displayed on the smart glasses display, accurately understands the content of the call, and responds appropriately to the customer. The input is the visual information displayed on the display, and the output is the user's appropriate response.

[0484] 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.

[0485] This invention relates to a "visual telephone" system that analyzes telephone voices and visually complements them to eliminate difficulties in hearing calls. Furthermore, this invention enables more intuitive and effective communication by incorporating an emotion engine that recognizes the user's emotions. This system operates in cooperation with the server, terminal, and user, analyzing voice data and displaying it visually in real time.

[0486] 1. System Configuration

[0487] server

[0488] The server is the central device that receives telephone voice data and processes it in real time. The server first receives the voice data and sequentially transfers it to the voice recognition module. The voice recognition module converts the voice data into text data. This converted text data is analyzed by the generative AI to infer the speaker's intention and context. In addition, the emotion engine analyzes the user's tone and intonation of voice to identify the user's emotions. Finally, the inferred intention and user emotion are complemented and converted into a format that is easy to display visually.

[0489] Terminal

[0490] The terminal is a device that allows the user to receive information visually. It receives the completed text data sent from the server and displays it on the screen. The terminal is made up of a television or a dedicated display, and has the function of highlighting important keywords and phrases. The user looks at this display to understand the content of the call.

[0491] User

[0492] The user is the subject of communication via telephone, and receives visually supplemented information through the terminal. The user can read the displayed text data and make requests or follow-up inquiries as needed. This allows the user to visually confirm content that is difficult to understand through voice alone, and also obtain information that reflects emotions.

[0493] 2. Program Processing

[0494] Audio Receiving and Forwarding

[0495] The server receives the voice data and transfers it to the voice recognition module in real time. The server stores the voice data in a buffer and supplies the data to the voice recognition module as needed.

[0496] Speech-to-text conversion

[0497] The speech recognition module converts the received voice data into text data in real time, including phonemic analysis of the voice and converting it into words, and the generated text data is sent to the analysis module.

[0498] Contextual analysis and intent inference

[0499] The generative AI (analysis module) analyzes text data and performs analysis to understand the context and the speaker's intent. This includes grammatical analysis and natural language processing to understand the context and infer the speaker's intent.

[0500] emotion recognition

[0501] The emotion engine analyzes the tone and intonation of the user's voice data to recognize the user's emotions. The emotion engine then classifies emotions using an emotion dictionary and acoustic model, and provides the identification results to the generation AI.

[0502] Visual Indication

[0503] The server sends the completed text data and emotion recognition results to the device, which then displays them visually. Important keywords and phrases are highlighted to make them easier for users to understand. Additionally, the user's emotions are also displayed, supporting a more intuitive understanding.

[0504] 3. Specific Examples

[0505] For example, consider the case where a user calls customer support to inquire about an internet connection problem. The server receives the voice data "The internet is not connected," and the speech recognition module converts it into text data "The internet is not connected." The generation AI analyzes this text data and complements it with the user's intent, "I want to solve my internet connection problem." The emotion engine identifies urgency and anxiety from the user's tone of voice and reflects this in the display. This complemented text data is sent to the device, and the user can visually confirm emotions such as "urgency" and "anxiety" along with the text "I want to solve my internet connection problem."

[0506] This system allows users to visually confirm content that is difficult to understand through audio alone, making telephone communication smoother and providing comprehensive information that takes emotions into consideration.

[0507] The processing flow will be explained below.

[0508] Step 1:

[0509] The server receives the voice data from the call. The voice data is captured digitally and stored in a buffer, where it is filtered to preserve the quality of the voice data.

[0510] Step 2:

[0511] The server transfers the buffered voice data to the speech recognition module in real time, using streaming technology to minimize data delays.

[0512] Step 3:

[0513] The speech recognition module analyzes the received speech data and breaks it down into phonemes, converts the phonemes into words, and organizes the generated words into text data.

[0514] Step 4:

[0515] The text data generated by the speech recognition module is sent to the analysis module, which receives the text data and analyzes the context of the text data using grammar analysis and natural language processing.

[0516] Step 5:

[0517] The generative AI (analysis module) understands the context of the text data and infers the speaker's intention. Based on the analysis results, it further examines the inferred intention and fills in any unclear parts.

[0518] Step 6:

[0519] The emotion engine analyzes the voice data received from the server and identifies emotions based on the tone and intonation of the user's voice. The emotion engine performs emotion classification using an acoustic model and an emotion dictionary.

[0520] Step 7:

[0521] The emotion engine passes the identified emotion information to the generation AI, which then complements the emotion information along with the inferred intent. This allows the user's emotions to be reflected in the text data.

[0522] Step 8:

[0523] The server sends the completed text data and emotion information to the device. The data is sent in a format that is easy for the user to understand, and is formatted to emphasize important keywords and emotions.

[0524] Step 9:

[0525] The terminal receives the data sent from the server and visually displays it to the user. The display screen combines the text data with the user's emotions, highlighting important keywords and emotions.

[0526] Step 10:

[0527] The user can read the text data and emotional information displayed on the device and check the contents of the call. If necessary, the user can make a request or ask again. This operation allows for more detailed information to be obtained.

[0528] Example 2

[0529] 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."

[0530] In conventional telephone communication, voice alone can make the content unclear, and the information conveyed can be inaccurate, especially in noisy environments or when the voice quality is poor. It is also difficult to accurately understand emotions and intentions through voice alone, making misunderstandings highly likely, especially in communication that involves emotions. This can lead to problems that reduce user satisfaction, especially in user support and customer service.

[0531] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving voice data, means for converting the received voice data into text data, means for analyzing the converted text data and inferring the context and the speaker's intention, means for analyzing the speaker's tone and intonation to identify the speaker's emotion, means for complementing the inferred intention and the identified emotion and correcting unclear parts, and means for visually displaying the complemented text data and the identified emotion. As a result, by converting voice data into text data in real time and performing context analysis and emotion identification, it becomes possible to visually complement information that is difficult to convey through voice alone.

[0532] The "means for receiving voice data" is a function that receives voice data in real time from a telephone line, VoIP, etc., and temporarily stores it in a buffer.

[0533] The "means for converting into text data" is a function that uses a voice recognition module to analyze voice data, perform phonemic analysis, and convert into words to generate text data.

[0534] "Means of inferring context and speaker intent" refers to a function that uses generative AI to analyze text data and understand the speaker's intent and context through grammatical analysis and natural language processing.

[0535] "Means for identifying emotions by analyzing voice tone and intonation" is a function that uses an emotion engine to analyze the intonation and strength of the voice in the audio data, and classify and identify emotions.

[0536] "Means to complement inferred intent and identified emotion and correct unclear parts" is a function that uses the analysis results obtained by generative AI and emotion engines to complement intent and emotion and make appropriate corrections to clarify unclear parts.

[0537] The "means for visually displaying the complemented text data and the identified emotion" is a function that uses the display device of the terminal to display the text data and the identified emotion on the screen in a form that is easy for the user to check.

[0538] This invention relates to a "visual telephone" system that eliminates the difficulty of hearing telephone calls by analyzing voice data and visually complementing it. This system analyzes voice data, infers the user's intentions and emotions, and visually displays them, enabling more intuitive and effective communication.

[0539] The specific system configuration mainly involves the collaboration of a server, a terminal, and a user.

[0540] server

[0541] The server is the central device that receives and processes telephone voice data in real time. First, the server buffers the voice data from the telephone line or VoIP. Next, it converts the voice data into text using a speech recognition module (e.g., Google Speech-to-Text API). The converted text data is analyzed by a generative AI (e.g., OpenAI GPT-4) to understand the speaker's intent and context. It also uses an emotion engine (e.g., Affectiva SDK) to analyze the user's tone and intonation to identify emotions. The server then integrates the inferred intent and identified emotions and presents them in a format that is easy to display visually.

[0542] Terminal

[0543] The terminal is a device that allows the user to visually receive information. It receives the supplemented text data and emotion recognition results sent from the server and displays them on the terminal's display device (e.g., tablet, PC, smartphone). This display also includes a function to highlight important keywords and phrases, making it easy for the user to grasp the information.

[0544] User

[0545] The user is the subject of communication via telephone, and receives visually supplemented information through the terminal. The user can read the displayed text data and make requests or follow-up inquiries as needed. This allows the user to visually confirm content that is difficult to understand through voice alone, and also obtain information that reflects emotions.

[0546] Specific examples

[0547] For example, consider the case where a user calls customer support to inquire about an internet connection problem. In this case, the server receives the voice data "The internet is not connecting," and the speech recognition module converts it into text data "The internet is not connecting." The generation AI analyzes this text data and infers the user's intention, which is "I want to solve my internet connection problem." Furthermore, the emotion engine identifies "urgency" and "anxiety" from the user's tone of voice and reflects these in the display. This complemented text data is sent to the device, and the user can visually confirm emotions such as "urgency" and "anxiety" along with the text data "I want to solve my internet connection problem."

[0548] Prompt Sentence Examples

[0549] User: "I can't connect to the internet, what should I do?"

[0550] Generation AI:

[0551] 1. Analyze user comments and convert them into text.

[0552] 2. Infer the user's intent from the text (e.g., I want to fix my internet connection problem).

[0553] 3. Identify urgency and anxiety from the user's tone of voice.

[0554] 4. The final results are compiled into a text and sent to the device.

[0555] This system allows users to visually confirm content that is difficult to understand through audio alone, making telephone communication smoother and providing comprehensive information that takes emotions into consideration.

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

[0557] Step 1:

[0558] Voice data reception

[0559] The server receives voice data in real time from a telephone line or VoIP.

[0560] Input: Voice data from telephone line or VoIP

[0561] Output: Audio data stored in the server's buffer

[0562] Specific operation: The server receives audio data via the API and stores it in a buffer.

[0563] Step 2:

[0564] Converting audio data to text

[0565] The server converts the voice data into text data using a voice recognition module.

[0566] Input: Audio data stored in the server's buffer

[0567] Output: Text data

[0568] Specific operation: The server inputs the voice data into a speech recognition API (e.g., Google Speech-to-Text API) and obtains the text data "Internet connection is down."

[0569] Step 3:

[0570] Contextual analysis and intent inference of text data

[0571] The server uses generative AI to analyze the text data and infer the context and speaker's intent.

[0572] Input: Text data obtained from the speech recognition module

[0573] Output: Inferred intent and contextually-based complementation data

[0574] Specific operation: The server inputs the text data into a generation AI (e.g., GPT-4) and obtains the intent "I want to solve the Internet connection problem" as the analysis result.

[0575] Step 4:

[0576] emotion recognition

[0577] The server uses an emotion engine to analyze the tone and intonation of the voice based on the voice data and identify the user's emotions.

[0578] Input: Audio data stored on the server

[0579] Output: Identified emotion data (e.g., urgency, anxiety)

[0580] Specific operation: The server uses an emotion engine (e.g., Affectiva SDK) to identify emotions such as "anxiety" from the voice data.

[0581] Step 5:

[0582] Visual display of complementary data

[0583] The server transmits the completed text data and the identified emotion data to the terminal, which then visually displays them.

[0584] Input: Complemented text data, identified emotion data

[0585] Output: Visually displayed data (text and sentiment)

[0586] Specific operation: The server sends the data to the terminal, and the terminal displays the text "I want to solve the problem of the Internet connection" and the emotion "anxiety" highlighted on the display device (e.g., tablet, PC).

[0587] The above is the specific processing flow of the program of this system.

[0588] (Application example 2)

[0589] 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."

[0590] In current telephone communications, it can be difficult to understand what is being said through voice alone. Furthermore, the user's emotions and intentions may not be conveyed accurately. Smooth communication is especially important in brick-and-mortar stores and customer support settings. However, existing systems lack a means to visually complement emotions and intentions, which reduces the efficiency and effectiveness of communication.

[0591] 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 receiving voice data, means for converting the received voice data into text data, means for analyzing the converted text data and inferring the context and the speaker's intention, means for completing the inferred intention and correcting unclear parts, means for analyzing the tone and intonation of the user's voice using an emotion engine to recognize emotions, and means for integrating the results of the emotion analysis into visually displayed data and displaying the results. This not only makes it possible to visually grasp content that is difficult to understand from voice alone, but also makes it possible to provide information that reflects the user's emotions, thereby enabling smooth communication between customers and staff, particularly in customer support situations in physical stores.

[0592] The "audio data receiving means" is a part of the system that has the function of receiving audio data transmitted via voice calls and the like.

[0593] The "text conversion means" is a part of the system that has the function of analyzing received voice data and converting it into text data.

[0594] The "intention inference means" is a part of the system that has the function of analyzing and understanding the context and the speaker's intention based on the converted text data.

[0595] The "intention completion means" is a part of the system that has the function of completing the speaker's inferred intention and correcting any unclear parts.

[0596] The "visual display means" is a part of the system that has the function of visually displaying the analyzed text data and emotion recognition results.

[0597] The "emotion engine" is a part of the system that has the ability to recognize the user's emotions by analyzing the tone and intonation of the voice.

[0598] The "emotion analysis result display means" is a part of the system that has the function of visually displaying the results of the emotion analysis by the emotion engine.

[0599] This invention relates to a customer support system for brick-and-mortar stores. This system receives and analyzes voice data and delivers information visually to users, thereby realizing smooth communication.

[0600] A specific example of the system is a customer support robot in a brick-and-mortar store that receives voice data from customers in real time, converts it into text, and then analyzes and visually displays the customer's intentions and emotions.

[0601] The server first receives the voice data. Using a voice data receiving means, it receives voice data transmitted via telephone or voice call in real time. Next, it converts the received voice data into text data using a text conversion means. For example, it uses a speech recognition library (speech_recognition). The converted text data is analyzed using an intent inference means and an intent completion means to infer the context and the speaker's intent. This process uses a generative AI model and natural language processing technology (TextBlob).

[0602] Furthermore, an emotion engine is used to analyze the tone and intonation of the voice to recognize the user's emotions. Specifically, an emotion analysis library (e.g., emotion dictionary and acoustic model) is used. This makes it possible to visually display the emotion analysis result together with the text data using an emotion analysis result display means. In-store displays and the screens of support robots are used as visual display means. The analyzed text data and the emotion results are displayed integrated on this screen.

[0603] As a concrete example, consider a scenario in which a customer asks a robot about a product. In this case, the server receives the voice data "Do you have this product in stock?" and converts it into text data "Do you have this product in stock?" using a speech recognition library. Next, a generative AI model analyzes this text data and infers the customer's intent in the form of "Check product availability." The emotion engine identifies urgency or anxiety from the customer's tone of voice and reflects this in the display.

[0604] Finally, as a visual display, emotions such as "urgent" or "anxiety" are displayed on in-store displays and on the screens of support robots along with the text "Check product inventory." This allows customers to visually confirm content that would be difficult to understand through voice alone, and enables store staff to accurately grasp the customer's intentions and emotions.

[0605] Examples of input prompts for generative AI models include:

[0606] "A customer is asking, 'Does the store have this item in stock?' Visually display this question in real time and use an emotion recognition engine to read the customer's emotions (urgency, anxiety, etc.)."

[0607] Using these prompts, the system visually displays customer questions based on text and emotion recognition, allowing store staff to accurately understand the customer's needs.

[0608] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0609] Step 1:

[0610] The server receives audio data sent via telephone or voice call. The input is audio data from a microphone or a call system. The output is to store the received audio data in a buffer. Specifically, the audio receiving module converts audio into digital data and processes it in a streaming format.

[0611] Step 2:

[0612] The server converts voice data into text data in real time. The input is the voice data stored in the buffer. The output is the converted text data. Specifically, it uses a speech recognition library (e.g., speech_recognition) to analyze the voice data into phonemes and output them as text.

[0613] Step 3:

[0614] The server analyzes the text data and infers the context and the speaker's intent. The input is the converted text data. The output is the inferred intent and its analysis results. Specific operations include using generative AI models and natural language processing techniques (e.g., TextBlob) to analyze the text grammatically and semantically.

[0615] Step 4:

[0616] The server completes the inferred intent and corrects any ambiguities. The input is the inferred intent and the text data. The output is the completed and corrected text data. Specifically, the analysis module reevaluates the context and intent and adds additional information as needed.

[0617] Step 5:

[0618] The server uses an emotion engine to analyze the emotions in the voice data. The input is the voice data and its text conversion results. The output is the emotion recognition results. Specifically, it analyzes the tone and intonation of the voice and classifies the emotion using an emotion dictionary and acoustic model.

[0619] Step 6:

[0620] The server sends the completed text data and the emotion analysis results to the terminal for visual display. The input is the completed text data and the emotion analysis results. The output is data for visual display that includes these data. Specifically, the data formatting module organizes this information and formats it into a format that is easy to display.

[0621] Step 7:

[0622] The terminal receives the data sent from the server and displays it visually. The input is the data for visual display sent from the server. The output is the text data and sentiment analysis results displayed on the display. Specifically, the terminal's display module highlights important keywords and phrases, providing information in a format that is easy for the user to review.

[0623] Step 8:

[0624] The user checks the visually displayed information and makes requests or re-inquires as necessary. The input is the text data displayed on the screen and the results of sentiment analysis. The output is the user's understanding and the next action. In concrete terms, the user checks the display and smoothly proceeds with the response via phone or chat.

[0625] 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.

[0626] 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.

[0627] 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.

[0628] [Third embodiment]

[0629] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0630] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0631] 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).

[0632] 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.

[0633] 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.

[0634] 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).

[0635] 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.

[0636] 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.

[0637] 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.

[0638] 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.

[0639] 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.

[0640] 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."

[0641] This invention relates to a "visual telephone" system that analyzes telephone voices and visually complements them to eliminate the difficulty of hearing telephone calls. This system operates in cooperation with the server, terminals, and users, analyzing voice data and displaying it visually in real time.

[0642] 1. System Configuration

[0643] server

[0644] The server is the central device that receives the voice data from the phone and processes it in real time. The server first receives the voice data and sequentially transfers it to the voice recognition module. The voice recognition module converts the voice data into text data. This converted text data is analyzed by the generative AI to infer the speaker's intent and context. Finally, the inferred intent is complemented and corrected into a format that is easy to display visually.

[0645] Terminal

[0646] The terminal is a device that allows the user to receive information visually. It receives the completed text data sent from the server and displays it on the screen. The terminal is made up of a television or a dedicated display, and has the function of highlighting important keywords and phrases. The user looks at this display to understand the content of the call.

[0647] User

[0648] The user is the subject of communication via telephone, and receives visually supplemented information through the terminal. The user can read the displayed text data and make requests or follow-up inquiries as needed. This allows the user to visually confirm content that is difficult to understand through voice alone.

[0649] 2. Program Processing

[0650] Audio Receiving and Forwarding

[0651] The server receives the voice data and transfers it to the voice recognition module in real time. The server stores the voice data in a buffer and supplies the data to the voice recognition module as needed.

[0652] Speech-to-text conversion

[0653] The speech recognition module converts the received voice data into text data in real time, including phonemic analysis of the voice and converting it into words, and the generated text data is sent to the analysis module.

[0654] Contextual analysis and intent inference

[0655] The generative AI (analysis module) analyzes text data and performs analysis to understand the context and the speaker's intent. This includes grammatical analysis and natural language processing to understand the context and infer the speaker's intent.

[0656] Visual Indication

[0657] The server sends the completed text data to the device, which then displays it visually, highlighting important keywords and phrases to make it easier for the user to understand.

[0658] 3. Specific Examples

[0659] For example, consider the case where a user calls customer support to inquire about an internet connection problem. The server receives the voice data "The internet is not connected," and the speech recognition module converts it into text data "The internet is not connected." The generation AI analyzes this text data and completes it to reflect the user's intent: "I want to solve my internet connection problem." This completed text data is sent to the device, and the user can visually confirm the text "I want to solve my internet connection problem."

[0660] This system allows users to visually confirm content that is difficult to understand through audio alone, making telephone communication smoother.

[0661] The processing flow will be explained below.

[0662] Step 1:

[0663] The server receives the voice data from the phone, converts it into a digital format, and stores it in a buffer. At this stage, processing is performed to minimize degradation of the voice data.

[0664] Step 2:

[0665] The server transfers the buffered voice data to the voice recognition module in real time, ensuring efficient data transfer to avoid delays.

[0666] Step 3:

[0667] The speech recognition module analyzes the received voice data and breaks it down into phonemes. The phonemes are organized into words and converted into text data. This text data is temporarily stored in the speech recognition module.

[0668] Step 4:

[0669] The text data generated by the speech recognition module is sent to the analysis module, which receives the text data and begins analyzing the context. It uses grammatical analysis and natural language processing to analyze the structure and meaning of the sentence.

[0670] Step 5:

[0671] The generative AI (analysis module) understands the context of the text data and infers the speaker's intention. It derives a more accurate interpretation by referring to the intonation and stress of the voice and the history of past conversations. Any unclear parts are also filled in during this process.

[0672] Step 6:

[0673] The generative AI completes the inferred intent and corrects any ambiguities. The inferred results are formatted to make them natural and easy to understand. This text data is then generated as data for the final visual display.

[0674] Step 7:

[0675] The server then sends the completed text data to the device, formatted to make it easier for the user to read and highlighting important keywords and phrases.

[0676] Step 8:

[0677] The terminal receives the text data sent from the server and displays it visually to the user. The display screen is laid out in an easy-to-read format and is designed to allow the user to quickly grasp important information.

[0678] Step 9:

[0679] The user can read the text data displayed on the device to check the contents of the call. If necessary, they can make a request or ask again. This operation allows the user to obtain more detailed information.

[0680] In this way, the server, terminal, and user work together to analyze voice data in real time and provide visual assistance, making it easier to accurately understand the content of telephone conversations.

[0681] Example 1

[0682] 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."

[0683] Conventional telephone communication relies solely on the transmission of information through voice, which often leads to difficulty in hearing and misunderstandings. In particular, important content can be difficult to understand due to environmental noise or the speaker's unique accent or intonation. This makes telephone communication extremely difficult, especially for elderly people and those with hearing impairments. To solve this problem, a system that visually complements voice is needed.

[0684] 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.

[0685] In this invention, the server includes means for receiving voice data, means for sequentially supplying the voice data, and means for transmitting analyzed text data to the terminal. This enables the voice data to be analyzed quickly and efficiently, and helps the user's visual understanding. Also, by including means for converting the received voice data into text data, means for analyzing the converted text data and inferring the context and the speaker's intention, means for completing the inferred intention and correcting unclear parts, means for visually displaying the completed text data, and means for the terminal to highlight the text data, it becomes possible to comprehensively understand the voice data and quickly grasp important information.

[0686] "Voice data" means analog or digital audio signals obtained from telephone communications or other audio information.

[0687] A "means" is a device, system, software, or combination thereof used to achieve a particular purpose.

[0688] A "server" is a computer system that provides services to multiple terminals over a network.

[0689] "Serial supply" means sending data continuously in small amounts in order.

[0690] "Text data" refers to character string information generated from voice data through processing such as voice recognition.

[0691] "Analysis" means analyzing and interpreting the structure and meaning of data in order to understand its content.

[0692] "Context" refers to the circumstances and related information that form the background of a text or utterance.

[0693] The "speaker's intention" refers to the purpose, thoughts, and intentions behind what is said.

[0694] "Complement" means adding missing parts to make the whole complete.

[0695] An "ambiguous part" is one that is not clearly understood or is ambiguous.

[0696] "Visual display" means expressing information using visual elements such as letters and figures and displaying it on a display or monitor.

[0697] A "terminal" is a device such as a computer, smartphone, or tablet that a user uses to send, receive, and process information.

[0698] "Highlighting" refers to displaying specific information in a way that makes it stand out, such as by changing the color or font.

[0699] The present invention relates to a system that analyzes voice data and visually complements it to eliminate difficulties in hearing telephone conversations. This system is designed to convert voice data into text in real time, analyze the context and intent, and visualize the text. An embodiment of the system is described below in detail.

[0700] Hardware and Software Use

[0701] This system involves the cooperation of a server, a terminal, and a user. The server is a central device for receiving telephone voice data and processing it in real time. The server includes a means for receiving voice data and a means for buffering and sequentially supplying the voice data. For example, the server can be integrated with the Google Speech-to-Text API.

[0702] When the voice data is received by the server, it is converted into text data by a speech recognition module (such as the Google Speech-to-Text API). This text data is then analyzed by a generative AI (such as OpenAI's GPT-4). The generative AI analyzes the data to understand the context and the speaker's intent, and generates supplemental text data.

[0703] The completed text data is sent from the server to the terminal. The terminal is a device that allows the user to receive information visually, such as an iPad or a dedicated display. The terminal has the function of highlighting important keywords and phrases.

[0704] Specific examples

[0705] For example, consider a case where a user calls customer support to report a problem with their Internet connection. When the user says, "The Internet is down," the voice data is received by the server. This data is transferred to a speech recognition module and converted into text data that says, "The Internet is down."

[0706] This text data is analyzed by a generative AI to infer the context and user intent. The generative AI generates a completed text, such as "I want to solve my internet connection problem." The completed text data is sent from the server to the device and displayed on the device's display. The device highlights important keywords such as "internet" and "connection problem."

[0707] Prompt Sentence Examples

[0708] "A user is on the phone talking about an issue with their internet connection. Convert this audio data into text and infer what the user meant."

[0709] This system allows users to visually confirm information that is difficult to understand through voice alone, making telephone communication smoother and particularly useful for elderly people and those with hearing impairments.

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

[0711] Step 1:

[0712] The server receives the voice data and transfers it to the voice recognition module in real time.

[0713] Specifically, the server receives the voice data (input) spoken by the user over the communication line. The received voice data is temporarily stored in buffer memory. The data is then sequentially supplied from the buffer to the Google Speech-to-Text API (output).

[0714] Step 2:

[0715] The voice recognition module converts the received voice data into text data.

[0716] Specifically, the Google Speech-to-Text API performs phonemic analysis on the voice data (input) transferred from the server and converts it into words, and the resulting text data (output) is sent to the analysis module.

[0717] Step 3:

[0718] Generative AI analyzes text data and infers the context and speaker's intention.

[0719] Specifically, the server receives text data (input) from the Google Speech-to-Text API and sends it to OpenAI's GPT-4. GPT-4 uses grammar analysis and natural language processing techniques to understand the context of the text and infer the speaker's intent. During this process, relevant keywords and phrases are identified and a complementary text (output) is generated that clearly indicates the speaker's intent.

[0720] Step 4:

[0721] The server transmits the completed text data to the terminal.

[0722] Specifically, the server sends the completed text data (input) generated by GPT-4 to the device. This includes a mechanism for safely and reliably transferring data using a communication protocol. The completed text data (output) arrives at the device.

[0723] Step 5:

[0724] The terminal displays the completed text data.

[0725] Specifically, the device (e.g., iPad) receives the completed text data (input) sent from the server and displays it on the display. The device highlights important keywords and phrases so that the user can easily confirm them visually. The user then checks this highlighted completed text (output).

[0726] Step 6:

[0727] The user checks the text data displayed on the screen and makes requests or re-inquires as necessary.

[0728] Specifically, the user reads and understands the text data (input) displayed on the device's display. If necessary, the user can obtain more information by issuing a follow-up inquiry or request through the device. This interaction generates new voice data (output), and the system operates again.

[0729] (Application example 1)

[0730] 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."

[0731] Recent voice communication has the problem that the accuracy of information transmission is easily reduced, especially in noisy environments or when the speaker is difficult to hear. There is also a risk that users may miss important information while answering the phone. Furthermore, in customer support and customer service at physical stores, staff are required to accurately understand the content of phone calls in real time. To solve these issues, it is necessary to develop a system that visually complements voice data and allows users to obtain information efficiently.

[0732] 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.

[0733] In this invention, the server includes means for receiving voice data, means for converting the received voice data into text data, means for analyzing the converted text data and inferring the context and the speaker's intent, means for completing the inferred intent and correcting unclear parts, means for visually displaying the completed text data, and means for using a display device of a personal digital assistant or portable terminal as display means to allow the user to visually confirm the information. This improves the accuracy of voice communication, allowing the user to efficiently obtain important information without missing it. Furthermore, in customer support and customer service at physical stores, staff can visually understand the content of phone calls, enabling them to respond quickly and accurately.

[0734] "Audio data" refers to audio signal information captured through a microphone or other audio collection device.

[0735] "Text data" is character information converted from voice data using voice recognition technology.

[0736] "Context" is the context or background information needed to understand the meaning of spoken words or sentences.

[0737] "Speaker intention" is the main message or purpose that the speaker is trying to convey.

[0738] "Completion" means making something more complete by adding unclear or missing information.

[0739] An "unclear portion" is a portion of the audio data or text data that is unclear due to noise or difficulty in hearing.

[0740] "Visual display means" refers to a device or technology that uses a screen or display to display text data in a form that is visually recognizable to the user.

[0741] A "personal digital assistant" is a portable electronic device such as a smartphone or tablet.

[0742] A "portable terminal" is an electronic device that can be worn by a user, such as a glasses-type display or a smart watch.

[0743] A "display device" is a hardware device that can visually display characters, images, etc.

[0744] This invention relates to a "visual telephone" system that analyzes voice data and provides visual supplementation. This system converts voice data into text data in real time, analyzes it using a generative AI, infers the speaker's intention and context, and provides visual supplementation. Specific embodiments of this system are described below.

[0745] System configuration

[0746] server

[0747] The server is the central part of the system and includes the following functions:

[0748] 1. Means for receiving voice data: The server receives voice data from the telephone. This reception is performed in real time.

[0749] 2. Means for converting voice data to text data: The received voice data is converted to text data using a voice recognition module. Specifically, the Python SpeechRecognition library is used.

[0750] 3. Means of inferring context and speaker intent: Generative AI analyzes text data to understand the context and speaker intent. This process uses the Hugging Face transformers library.

[0751] 4. Means of completing inferred intent and correcting ambiguities: Complete inferred intent and make corrections as necessary.

[0752] 5. Means for visually displaying the completed text data: The completed text data is transmitted to the user's personal digital assistant or portable terminal.

[0753] Terminal

[0754] A terminal is a device through which a user receives visual information. It has the following functions:

[0755] 1. Display device: A display device such as a smartphone or smart glasses visually displays the text data sent from the server, highlighting important keywords and phrases.

[0756] User

[0757] Users are the main actors in communication via telephone and receive visually supplemented information through their devices. For example, customer support staff at a brick-and-mortar store can wear smart glasses and understand customer inquiries in real time.

[0758] Program processing and specific examples

[0759] The server receives the voice data and converts it into text data using a speech recognition module. The converted text data is then analyzed by a generative AI to infer the speaker's intent and context. This inferred intent and context is then augmented into a format that is easy to display visually and sent to the user's device.

[0760] Hardware and software used

[0761] Server: Receives voice data, converts it into text, and generates AI analysis

[0762] Software: SpeechRecognition, transformers (natural language processing)

[0763] Devices: smart glasses, smartphone

[0764] Display device: smart glasses display

[0765] Specific examples

[0766] Let's take the example of a user (a customer support staff member at a brick-and-mortar store) receiving a phone call from a customer asking, "Please check if the product is in stock." At this time, the server receives the voice data and converts it into text data saying, "Please check if the product is in stock." The generation AI analyzes this text data, infers that the product's stock information needs to be checked, and displays this information on the smart glasses' display.

[0767] Example prompt sentence:

[0768] I want to check if the product is in stock

[0769] This example allows the user (customer support staff) to immediately take appropriate action based on the visually supplemented information. This system eliminates ambiguity in voice data and enables efficient information transmission.

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

[0771] Step 1:

[0772] The server receives the voice data. Specifically, the server captures the telephone voice signal in real time and stores it in a buffer as voice data. The input is the telephone voice signal, and the output is the voice data stored in the buffer. This prepares the server to pass the received voice data to the next process.

[0773] Step 2:

[0774] The server converts the voice data into text data. Specifically, it uses a speech recognition module (SpeechRecognition library) to convert the voice data stored in the buffer into text data. The input is the voice data stored in the buffer, and the output is the converted text data. This data processing involves phonemic analysis and conversion into words.

[0775] Step 3:

[0776] The server uses a generative AI model to analyze text data and infer the context and speaker intent. Specifically, the text data is sent to a natural language processing module (the transformers library) to analyze the context and intent. The input is the transformed text data, and the output is information about the analyzed intent and context. This data processing includes grammatical analysis and context understanding.

[0777] Step 4:

[0778] The server completes the inferred intent and corrects any unclear parts. Specifically, based on the analysis results of the generative AI, any unclear parts in the utterance are corrected and the final intent is determined. The input is the analyzed intent and context information, and the output is the completed text data. A completion algorithm is used in this process.

[0779] Step 5:

[0780] The server sends the completed text data to the device. Specifically, the completed text data is sent to the device (smart glasses or smartphone) via a network. The input is the completed text data, and the output is the state after transmission to the device is complete. Network communication technology is used for this data transfer.

[0781] Step 6:

[0782] The terminal visually displays the received text completion data. Specifically, the received text completion data is displayed on the display of smart glasses or a smartphone, highlighting important keywords and phrases. The input is the text completion data sent from the server, and the output is the visual information displayed on the display.

[0783] Step 7:

[0784] The user checks the visual information and responds accordingly. Specifically, the user (customer support staff) looks at the information displayed on the smart glasses display, accurately understands the content of the call, and responds appropriately to the customer. The input is the visual information displayed on the display, and the output is the user's appropriate response.

[0785] 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.

[0786] This invention relates to a "visual telephone" system that analyzes telephone voices and visually complements them to eliminate difficulties in hearing calls. Furthermore, this invention enables more intuitive and effective communication by incorporating an emotion engine that recognizes the user's emotions. This system operates in cooperation with the server, terminal, and user, analyzing voice data and displaying it visually in real time.

[0787] 1. System Configuration

[0788] server

[0789] The server is the central device that receives telephone voice data and processes it in real time. The server first receives the voice data and sequentially transfers it to the voice recognition module. The voice recognition module converts the voice data into text data. This converted text data is analyzed by the generative AI to infer the speaker's intention and context. In addition, the emotion engine analyzes the user's tone and intonation of voice to identify the user's emotions. Finally, the inferred intention and user emotion are complemented and converted into a format that is easy to display visually.

[0790] Terminal

[0791] The terminal is a device that allows the user to receive information visually. It receives the completed text data sent from the server and displays it on the screen. The terminal is made up of a television or a dedicated display, and has the function of highlighting important keywords and phrases. The user looks at this display to understand the content of the call.

[0792] User

[0793] The user is the subject of communication via telephone, and receives visually supplemented information through the terminal. The user can read the displayed text data and make requests or follow-up inquiries as needed. This allows the user to visually confirm content that is difficult to understand through voice alone, and also obtain information that reflects emotions.

[0794] 2. Program Processing

[0795] Audio Receiving and Forwarding

[0796] The server receives the voice data and transfers it to the voice recognition module in real time. The server stores the voice data in a buffer and supplies the data to the voice recognition module as needed.

[0797] Speech-to-text conversion

[0798] The speech recognition module converts the received voice data into text data in real time, including phonemic analysis of the voice and converting it into words, and the generated text data is sent to the analysis module.

[0799] Contextual analysis and intent inference

[0800] The generative AI (analysis module) analyzes text data and performs analysis to understand the context and the speaker's intent. This includes grammatical analysis and natural language processing to understand the context and infer the speaker's intent.

[0801] emotion recognition

[0802] The emotion engine analyzes the tone and intonation of the user's voice data to recognize the user's emotions. The emotion engine then classifies emotions using an emotion dictionary and acoustic model, and provides the identification results to the generation AI.

[0803] Visual Indication

[0804] The server sends the completed text data and emotion recognition results to the device, which then displays them visually. Important keywords and phrases are highlighted to make them easier for users to understand. Additionally, the user's emotions are also displayed, supporting a more intuitive understanding.

[0805] 3. Specific Examples

[0806] For example, consider the case where a user calls customer support to inquire about an internet connection problem. The server receives the voice data "The internet is not connected," and the speech recognition module converts it into text data "The internet is not connected." The generation AI analyzes this text data and complements it with the user's intent, "I want to solve my internet connection problem." The emotion engine identifies urgency and anxiety from the user's tone of voice and reflects this in the display. This complemented text data is sent to the device, and the user can visually confirm emotions such as "urgency" and "anxiety" along with the text "I want to solve my internet connection problem."

[0807] This system allows users to visually confirm content that is difficult to understand through audio alone, making telephone communication smoother and providing comprehensive information that takes emotions into consideration.

[0808] The processing flow will be explained below.

[0809] Step 1:

[0810] The server receives the voice data from the call. The voice data is captured digitally and stored in a buffer, where it is filtered to preserve the quality of the voice data.

[0811] Step 2:

[0812] The server transfers the buffered voice data to the speech recognition module in real time, using streaming technology to minimize data delays.

[0813] Step 3:

[0814] The speech recognition module analyzes the received speech data and breaks it down into phonemes, converts the phonemes into words, and organizes the generated words into text data.

[0815] Step 4:

[0816] The text data generated by the speech recognition module is sent to the analysis module, which receives the text data and analyzes the context of the text data using grammar analysis and natural language processing.

[0817] Step 5:

[0818] The generative AI (analysis module) understands the context of the text data and infers the speaker's intention. Based on the analysis results, it further examines the inferred intention and fills in any unclear parts.

[0819] Step 6:

[0820] The emotion engine analyzes the voice data received from the server and identifies emotions based on the tone and intonation of the user's voice. The emotion engine performs emotion classification using an acoustic model and an emotion dictionary.

[0821] Step 7:

[0822] The emotion engine passes the identified emotion information to the generation AI, which then complements the emotion information along with the inferred intent. This allows the user's emotions to be reflected in the text data.

[0823] Step 8:

[0824] The server sends the completed text data and emotion information to the device. The data is sent in a format that is easy for the user to understand, and is formatted to emphasize important keywords and emotions.

[0825] Step 9:

[0826] The terminal receives the data sent from the server and visually displays it to the user. The display screen combines the text data with the user's emotions, highlighting important keywords and emotions.

[0827] Step 10:

[0828] The user can read the text data and emotional information displayed on the device and check the contents of the call. If necessary, the user can make a request or ask again. This operation allows for more detailed information to be obtained.

[0829] Example 2

[0830] 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."

[0831] In conventional telephone communication, voice alone can make the content unclear, and the information conveyed can be inaccurate, especially in noisy environments or when the voice quality is poor. It is also difficult to accurately understand emotions and intentions through voice alone, making misunderstandings highly likely, especially in communication that involves emotions. This can lead to problems that reduce user satisfaction, especially in user support and customer service.

[0832] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving voice data, means for converting the received voice data into text data, means for analyzing the converted text data and inferring the context and the speaker's intention, means for analyzing the speaker's tone and intonation to identify the speaker's emotion, means for complementing the inferred intention and the identified emotion and correcting unclear parts, and means for visually displaying the complemented text data and the identified emotion. As a result, by converting voice data into text data in real time and performing context analysis and emotion identification, it becomes possible to visually complement information that is difficult to convey through voice alone.

[0833] The "means for receiving voice data" is a function that receives voice data in real time from a telephone line, VoIP, etc., and temporarily stores it in a buffer.

[0834] The "means for converting into text data" is a function that uses a voice recognition module to analyze voice data, perform phonemic analysis, and convert into words to generate text data.

[0835] "Means of inferring context and speaker intent" refers to a function that uses generative AI to analyze text data and understand the speaker's intent and context through grammatical analysis and natural language processing.

[0836] "Means for identifying emotions by analyzing voice tone and intonation" is a function that uses an emotion engine to analyze the intonation and strength of the voice in the audio data, and classify and identify emotions.

[0837] "Means to complement inferred intent and identified emotion and correct unclear parts" is a function that uses the analysis results obtained by generative AI and emotion engines to complement intent and emotion and make appropriate corrections to clarify unclear parts.

[0838] The "means for visually displaying the complemented text data and the identified emotion" is a function that uses the display device of the terminal to display the text data and the identified emotion on the screen in a form that is easy for the user to check.

[0839] This invention relates to a "visual telephone" system that eliminates the difficulty of hearing telephone calls by analyzing voice data and visually complementing it. This system analyzes voice data, infers the user's intentions and emotions, and visually displays them, enabling more intuitive and effective communication.

[0840] The specific system configuration mainly involves the collaboration of a server, a terminal, and a user.

[0841] server

[0842] The server is the central device that receives and processes telephone voice data in real time. First, the server buffers the voice data from the telephone line or VoIP. Next, it converts the voice data into text using a speech recognition module (e.g., Google Speech-to-Text API). The converted text data is analyzed by a generative AI (e.g., OpenAI GPT-4) to understand the speaker's intent and context. It also uses an emotion engine (e.g., Affectiva SDK) to analyze the user's tone and intonation to identify emotions. The server then integrates the inferred intent and identified emotions and presents them in a format that is easy to display visually.

[0843] Terminal

[0844] The terminal is a device that allows the user to visually receive information. It receives the supplemented text data and emotion recognition results sent from the server and displays them on the terminal's display device (e.g., tablet, PC, smartphone). This display also includes a function to highlight important keywords and phrases, making it easy for the user to grasp the information.

[0845] User

[0846] The user is the subject of communication via telephone, and receives visually supplemented information through the terminal. The user can read the displayed text data and make requests or follow-up inquiries as needed. This allows the user to visually confirm content that is difficult to understand through voice alone, and also obtain information that reflects emotions.

[0847] Specific examples

[0848] For example, consider the case where a user calls customer support to inquire about an internet connection problem. In this case, the server receives the voice data "The internet is not connecting," and the speech recognition module converts it into text data "The internet is not connecting." The generation AI analyzes this text data and infers the user's intention, which is "I want to solve my internet connection problem." Furthermore, the emotion engine identifies "urgency" and "anxiety" from the user's tone of voice and reflects these in the display. This complemented text data is sent to the device, and the user can visually confirm emotions such as "urgency" and "anxiety" along with the text data "I want to solve my internet connection problem."

[0849] Prompt Sentence Examples

[0850] User: "I can't connect to the internet, what should I do?"

[0851] Generation AI:

[0852] 1. Analyze user comments and convert them into text.

[0853] 2. Infer the user's intent from the text (e.g., I want to fix my internet connection problem).

[0854] 3. Identify urgency and anxiety from the user's tone of voice.

[0855] 4. The final results are compiled into a text and sent to the device.

[0856] This system allows users to visually confirm content that is difficult to understand through audio alone, making telephone communication smoother and providing comprehensive information that takes emotions into consideration.

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

[0858] Step 1:

[0859] Voice data reception

[0860] The server receives voice data in real time from a telephone line or VoIP.

[0861] Input: Voice data from telephone line or VoIP

[0862] Output: Audio data stored in the server's buffer

[0863] Specific operation: The server receives audio data via the API and stores it in a buffer.

[0864] Step 2:

[0865] Converting audio data to text

[0866] The server converts the voice data into text data using a voice recognition module.

[0867] Input: Audio data stored in the server's buffer

[0868] Output: Text data

[0869] Specific operation: The server inputs the voice data into a speech recognition API (e.g., Google Speech-to-Text API) and obtains the text data "Internet connection is down."

[0870] Step 3:

[0871] Contextual analysis and intent inference of text data

[0872] The server uses generative AI to analyze the text data and infer the context and speaker's intent.

[0873] Input: Text data obtained from the speech recognition module

[0874] Output: Inferred intent and contextually-based complementation data

[0875] Specific operation: The server inputs the text data into a generation AI (e.g., GPT-4) and obtains the intent "I want to solve the Internet connection problem" as the analysis result.

[0876] Step 4:

[0877] emotion recognition

[0878] The server uses an emotion engine to analyze the tone and intonation of the voice based on the voice data and identify the user's emotions.

[0879] Input: Audio data stored on the server

[0880] Output: Identified emotion data (e.g., urgency, anxiety)

[0881] Specific operation: The server uses an emotion engine (e.g., Affectiva SDK) to identify emotions such as "anxiety" from the voice data.

[0882] Step 5:

[0883] Visual display of complementary data

[0884] The server transmits the completed text data and the identified emotion data to the terminal, which then visually displays them.

[0885] Input: Complemented text data, identified emotion data

[0886] Output: Visually displayed data (text and sentiment)

[0887] Specific operation: The server sends the data to the terminal, and the terminal displays the text "I want to solve the problem of the Internet connection" and the emotion "anxiety" highlighted on the display device (e.g., tablet, PC).

[0888] The above is the specific processing flow of the program of this system.

[0889] (Application example 2)

[0890] 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."

[0891] In current telephone communications, it can be difficult to understand what is being said through voice alone. Furthermore, the user's emotions and intentions may not be conveyed accurately. Smooth communication is especially important in brick-and-mortar stores and customer support settings. However, existing systems lack a means to visually complement emotions and intentions, which reduces the efficiency and effectiveness of communication.

[0892] 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 receiving voice data, means for converting the received voice data into text data, means for analyzing the converted text data and inferring the context and the speaker's intention, means for completing the inferred intention and correcting unclear parts, means for analyzing the tone and intonation of the user's voice using an emotion engine to recognize emotions, and means for integrating the results of the emotion analysis into visually displayed data and displaying the results. This not only makes it possible to visually grasp content that is difficult to understand from voice alone, but also makes it possible to provide information that reflects the user's emotions, thereby enabling smooth communication between customers and staff, particularly in customer support situations in physical stores.

[0893] The "audio data receiving means" is a part of the system that has the function of receiving audio data transmitted via voice calls and the like.

[0894] The "text conversion means" is a part of the system that has the function of analyzing received voice data and converting it into text data.

[0895] The "intention inference means" is a part of the system that has the function of analyzing and understanding the context and the speaker's intention based on the converted text data.

[0896] The "intention completion means" is a part of the system that has the function of completing the speaker's inferred intention and correcting any unclear parts.

[0897] The "visual display means" is a part of the system that has the function of visually displaying the analyzed text data and emotion recognition results.

[0898] The "emotion engine" is a part of the system that has the ability to recognize the user's emotions by analyzing the tone and intonation of the voice.

[0899] The "emotion analysis result display means" is a part of the system that has the function of visually displaying the results of the emotion analysis by the emotion engine.

[0900] This invention relates to a customer support system for brick-and-mortar stores. This system receives and analyzes voice data and delivers information visually to users, thereby realizing smooth communication.

[0901] A specific example of the system is a customer support robot in a brick-and-mortar store that receives voice data from customers in real time, converts it into text, and then analyzes and visually displays the customer's intentions and emotions.

[0902] The server first receives the voice data. Using a voice data receiving means, it receives voice data transmitted via telephone or voice call in real time. Next, it converts the received voice data into text data using a text conversion means. For example, it uses a speech recognition library (speech_recognition). The converted text data is analyzed using an intent inference means and an intent completion means to infer the context and the speaker's intent. This process uses a generative AI model and natural language processing technology (TextBlob).

[0903] Furthermore, an emotion engine is used to analyze the tone and intonation of the voice to recognize the user's emotions. Specifically, an emotion analysis library (e.g., emotion dictionary and acoustic model) is used. This makes it possible to visually display the emotion analysis result together with the text data using an emotion analysis result display means. In-store displays and the screens of support robots are used as visual display means. The analyzed text data and the emotion results are displayed integrated on this screen.

[0904] As a concrete example, consider a scenario in which a customer asks a robot about a product. In this case, the server receives the voice data "Do you have this product in stock?" and converts it into text data "Do you have this product in stock?" using a speech recognition library. Next, a generative AI model analyzes this text data and infers the customer's intent in the form of "Check product availability." The emotion engine identifies urgency or anxiety from the customer's tone of voice and reflects this in the display.

[0905] Finally, as a visual display, emotions such as "urgent" or "anxiety" are displayed on in-store displays and on the screens of support robots along with the text "Check product inventory." This allows customers to visually confirm content that would be difficult to understand through voice alone, and enables store staff to accurately grasp the customer's intentions and emotions.

[0906] Examples of input prompts for generative AI models include:

[0907] "A customer is asking, 'Does the store have this item in stock?' Visually display this question in real time and use an emotion recognition engine to read the customer's emotions (urgency, anxiety, etc.)."

[0908] Using these prompts, the system visually displays customer questions based on text and emotion recognition, allowing store staff to accurately understand the customer's needs.

[0909] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0910] Step 1:

[0911] The server receives audio data sent via telephone or voice call. The input is audio data from a microphone or a call system. The output is to store the received audio data in a buffer. Specifically, the audio receiving module converts audio into digital data and processes it in a streaming format.

[0912] Step 2:

[0913] The server converts voice data into text data in real time. The input is the voice data stored in the buffer. The output is the converted text data. Specifically, it uses a speech recognition library (e.g., speech_recognition) to analyze the voice data into phonemes and output them as text.

[0914] Step 3:

[0915] The server analyzes the text data and infers the context and the speaker's intent. The input is the converted text data. The output is the inferred intent and its analysis results. Specific operations include using generative AI models and natural language processing techniques (e.g., TextBlob) to analyze the text grammatically and semantically.

[0916] Step 4:

[0917] The server completes the inferred intent and corrects any ambiguities. The input is the inferred intent and the text data. The output is the completed and corrected text data. Specifically, the analysis module reevaluates the context and intent and adds additional information as needed.

[0918] Step 5:

[0919] The server uses an emotion engine to analyze the emotions in the voice data. The input is the voice data and its text conversion results. The output is the emotion recognition results. Specifically, it analyzes the tone and intonation of the voice and classifies the emotion using an emotion dictionary and acoustic model.

[0920] Step 6:

[0921] The server sends the completed text data and the emotion analysis results to the terminal for visual display. The input is the completed text data and the emotion analysis results. The output is data for visual display that includes these data. Specifically, the data formatting module organizes this information and formats it into a format that is easy to display.

[0922] Step 7:

[0923] The terminal receives the data sent from the server and displays it visually. The input is the data for visual display sent from the server. The output is the text data and sentiment analysis results displayed on the display. Specifically, the terminal's display module highlights important keywords and phrases, providing information in a format that is easy for the user to review.

[0924] Step 8:

[0925] The user checks the visually displayed information and makes requests or re-inquires as necessary. The input is the text data displayed on the screen and the results of sentiment analysis. The output is the user's understanding and the next action. In concrete terms, the user checks the display and smoothly proceeds with the response via phone or chat.

[0926] 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.

[0927] 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.

[0928] 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.

[0929] [Fourth embodiment]

[0930] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0931] 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.

[0932] 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).

[0933] 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.

[0934] 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.

[0935] 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).

[0936] 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.

[0937] 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.

[0938] 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.

[0939] 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.

[0940] 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.

[0941] 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.

[0942] 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."

[0943] This invention relates to a "visual telephone" system that analyzes telephone voices and visually complements them to eliminate the difficulty of hearing telephone calls. This system operates in cooperation with the server, terminals, and users, analyzing voice data and displaying it visually in real time.

[0944] 1. System Configuration

[0945] server

[0946] The server is the central device that receives the voice data from the phone and processes it in real time. The server first receives the voice data and sequentially transfers it to the voice recognition module. The voice recognition module converts the voice data into text data. This converted text data is analyzed by the generative AI to infer the speaker's intent and context. Finally, the inferred intent is complemented and corrected into a format that is easy to display visually.

[0947] Terminal

[0948] The terminal is a device that allows the user to receive information visually. It receives the completed text data sent from the server and displays it on the screen. The terminal is made up of a television or a dedicated display, and has the function of highlighting important keywords and phrases. The user looks at this display to understand the content of the call.

[0949] User

[0950] The user is the subject of communication via telephone, and receives visually supplemented information through the terminal. The user can read the displayed text data and make requests or follow-up inquiries as needed. This allows the user to visually confirm content that is difficult to understand through voice alone.

[0951] 2. Program Processing

[0952] Audio Receiving and Forwarding

[0953] The server receives the voice data and transfers it to the voice recognition module in real time. The server stores the voice data in a buffer and supplies the data to the voice recognition module as needed.

[0954] Speech-to-text conversion

[0955] The speech recognition module converts the received voice data into text data in real time, including phonemic analysis of the voice and converting it into words, and the generated text data is sent to the analysis module.

[0956] Contextual analysis and intent inference

[0957] The generative AI (analysis module) analyzes text data and performs analysis to understand the context and the speaker's intent. This includes grammatical analysis and natural language processing to understand the context and infer the speaker's intent.

[0958] Visual Indication

[0959] The server sends the completed text data to the device, which then displays it visually, highlighting important keywords and phrases to make it easier for the user to understand.

[0960] 3. Specific Examples

[0961] For example, consider the case where a user calls customer support to inquire about an internet connection problem. The server receives the voice data "The internet is not connected," and the speech recognition module converts it into text data "The internet is not connected." The generation AI analyzes this text data and completes it to reflect the user's intent: "I want to solve my internet connection problem." This completed text data is sent to the device, and the user can visually confirm the text "I want to solve my internet connection problem."

[0962] This system allows users to visually confirm content that is difficult to understand through audio alone, making telephone communication smoother.

[0963] The processing flow will be explained below.

[0964] Step 1:

[0965] The server receives the voice data from the phone, converts it into a digital format, and stores it in a buffer. At this stage, processing is performed to minimize degradation of the voice data.

[0966] Step 2:

[0967] The server transfers the buffered voice data to the voice recognition module in real time, ensuring efficient data transfer to avoid delays.

[0968] Step 3:

[0969] The speech recognition module analyzes the received voice data and breaks it down into phonemes. The phonemes are organized into words and converted into text data. This text data is temporarily stored in the speech recognition module.

[0970] Step 4:

[0971] The text data generated by the speech recognition module is sent to the analysis module, which receives the text data and begins analyzing the context. It uses grammatical analysis and natural language processing to analyze the structure and meaning of the sentence.

[0972] Step 5:

[0973] The generative AI (analysis module) understands the context of the text data and infers the speaker's intention. It derives a more accurate interpretation by referring to the intonation and stress of the voice and the history of past conversations. Any unclear parts are also filled in during this process.

[0974] Step 6:

[0975] The generative AI completes the inferred intent and corrects any ambiguities. The inferred results are formatted to make them natural and easy to understand. This text data is then generated as data for the final visual display.

[0976] Step 7:

[0977] The server then sends the completed text data to the device, formatted to make it easier for the user to read and highlighting important keywords and phrases.

[0978] Step 8:

[0979] The terminal receives the text data sent from the server and displays it visually to the user. The display screen is laid out in an easy-to-read format and is designed to allow the user to quickly grasp important information.

[0980] Step 9:

[0981] The user can read the text data displayed on the device to check the contents of the call. If necessary, they can make a request or ask again. This operation allows the user to obtain more detailed information.

[0982] In this way, the server, terminal, and user work together to analyze voice data in real time and provide visual assistance, making it easier to accurately understand the content of telephone conversations.

[0983] Example 1

[0984] 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."

[0985] Conventional telephone communication relies solely on the transmission of information through voice, which often leads to difficulty in hearing and misunderstandings. In particular, important content can be difficult to understand due to environmental noise or the speaker's unique accent or intonation. This makes telephone communication extremely difficult, especially for elderly people and those with hearing impairments. To solve this problem, a system that visually complements voice is needed.

[0986] 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.

[0987] In this invention, the server includes means for receiving voice data, means for sequentially supplying the voice data, and means for transmitting analyzed text data to the terminal. This enables the voice data to be analyzed quickly and efficiently, and helps the user's visual understanding. Also, by including means for converting the received voice data into text data, means for analyzing the converted text data and inferring the context and the speaker's intention, means for completing the inferred intention and correcting unclear parts, means for visually displaying the completed text data, and means for the terminal to highlight the text data, it becomes possible to comprehensively understand the voice data and quickly grasp important information.

[0988] "Voice data" means analog or digital audio signals obtained from telephone communications or other audio information.

[0989] A "means" is a device, system, software, or combination thereof used to achieve a particular purpose.

[0990] A "server" is a computer system that provides services to multiple terminals over a network.

[0991] "Serial supply" means sending data continuously in small amounts in order.

[0992] "Text data" refers to character string information generated from voice data through processing such as voice recognition.

[0993] "Analysis" means analyzing and interpreting the structure and meaning of data in order to understand its content.

[0994] "Context" refers to the circumstances and related information that form the background of a text or utterance.

[0995] The "speaker's intention" refers to the purpose, thoughts, and intentions behind what is said.

[0996] "Complement" means adding missing parts to make the whole complete.

[0997] An "ambiguous part" is one that is not clearly understood or is ambiguous.

[0998] "Visual display" means expressing information using visual elements such as letters and figures and displaying it on a display or monitor.

[0999] A "terminal" is a device such as a computer, smartphone, or tablet that a user uses to send, receive, and process information.

[1000] "Highlighting" refers to displaying specific information in a way that makes it stand out, such as by changing the color or font.

[1001] The present invention relates to a system that analyzes voice data and visually complements it to eliminate difficulties in hearing telephone conversations. This system is designed to convert voice data into text in real time, analyze the context and intent, and visualize the text. An embodiment of the system is described below in detail.

[1002] Hardware and Software Use

[1003] This system involves the cooperation of a server, a terminal, and a user. The server is a central device for receiving telephone voice data and processing it in real time. The server includes a means for receiving voice data and a means for buffering and sequentially supplying the voice data. For example, the server can be integrated with the Google Speech-to-Text API.

[1004] When the voice data is received by the server, it is converted into text data by a speech recognition module (such as the Google Speech-to-Text API). This text data is then analyzed by a generative AI (such as OpenAI's GPT-4). The generative AI analyzes the data to understand the context and the speaker's intent, and generates supplemental text data.

[1005] The completed text data is sent from the server to the terminal. The terminal is a device that allows the user to receive information visually, such as an iPad or a dedicated display. The terminal has the function of highlighting important keywords and phrases.

[1006] Specific examples

[1007] For example, consider a case where a user calls customer support to report a problem with their Internet connection. When the user says, "The Internet is down," the voice data is received by the server. This data is transferred to a speech recognition module and converted into text data that says, "The Internet is down."

[1008] This text data is analyzed by a generative AI to infer the context and user intent. The generative AI generates a completed text, such as "I want to solve my internet connection problem." The completed text data is sent from the server to the device and displayed on the device's display. The device highlights important keywords such as "internet" and "connection problem."

[1009] Prompt Sentence Examples

[1010] "A user is on the phone talking about an issue with their internet connection. Convert this audio data into text and infer what the user meant."

[1011] This system allows users to visually confirm information that is difficult to understand through voice alone, making telephone communication smoother and particularly useful for elderly people and those with hearing impairments.

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

[1013] Step 1:

[1014] The server receives the voice data and transfers it to the voice recognition module in real time.

[1015] Specifically, the server receives the voice data (input) spoken by the user over the communication line. The received voice data is temporarily stored in buffer memory. The data is then sequentially supplied from the buffer to the Google Speech-to-Text API (output).

[1016] Step 2:

[1017] The voice recognition module converts the received voice data into text data.

[1018] Specifically, the Google Speech-to-Text API performs phonemic analysis on the voice data (input) transferred from the server and converts it into words, and the resulting text data (output) is sent to the analysis module.

[1019] Step 3:

[1020] Generative AI analyzes text data and infers the context and speaker's intention.

[1021] Specifically, the server receives text data (input) from the Google Speech-to-Text API and sends it to OpenAI's GPT-4. GPT-4 uses grammar analysis and natural language processing techniques to understand the context of the text and infer the speaker's intent. During this process, relevant keywords and phrases are identified and a complementary text (output) is generated that clearly indicates the speaker's intent.

[1022] Step 4:

[1023] The server transmits the completed text data to the terminal.

[1024] Specifically, the server sends the completed text data (input) generated by GPT-4 to the device. This includes a mechanism for safely and reliably transferring data using a communication protocol. The completed text data (output) arrives at the device.

[1025] Step 5:

[1026] The terminal displays the completed text data.

[1027] Specifically, the device (e.g., iPad) receives the completed text data (input) sent from the server and displays it on the display. The device highlights important keywords and phrases so that the user can easily confirm them visually. The user then checks this highlighted completed text (output).

[1028] Step 6:

[1029] The user checks the text data displayed on the screen and makes requests or re-inquires as necessary.

[1030] Specifically, the user reads and understands the text data (input) displayed on the device's display. If necessary, the user can obtain more information by issuing a follow-up inquiry or request through the device. This interaction generates new voice data (output), and the system operates again.

[1031] (Application example 1)

[1032] 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."

[1033] Recent voice communication has the problem that the accuracy of information transmission is easily reduced, especially in noisy environments or when the speaker is difficult to hear. There is also a risk that users may miss important information while answering the phone. Furthermore, in customer support and customer service at physical stores, staff are required to accurately understand the content of phone calls in real time. To solve these issues, it is necessary to develop a system that visually complements voice data and allows users to obtain information efficiently.

[1034] 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.

[1035] In this invention, the server includes means for receiving voice data, means for converting the received voice data into text data, means for analyzing the converted text data and inferring the context and the speaker's intent, means for completing the inferred intent and correcting unclear parts, means for visually displaying the completed text data, and means for using a display device of a personal digital assistant or portable terminal as display means to allow the user to visually confirm the information. This improves the accuracy of voice communication, allowing the user to efficiently obtain important information without missing it. Furthermore, in customer support and customer service at physical stores, staff can visually understand the content of phone calls, enabling them to respond quickly and accurately.

[1036] "Audio data" refers to audio signal information captured through a microphone or other audio collection device.

[1037] "Text data" is character information converted from voice data using voice recognition technology.

[1038] "Context" is the context or background information needed to understand the meaning of spoken words or sentences.

[1039] "Speaker intention" is the main message or purpose that the speaker is trying to convey.

[1040] "Completion" means making something more complete by adding unclear or missing information.

[1041] An "unclear portion" is a portion of the audio data or text data that is unclear due to noise or difficulty in hearing.

[1042] "Visual display means" refers to a device or technology that uses a screen or display to display text data in a form that is visually recognizable to the user.

[1043] A "personal digital assistant" is a portable electronic device such as a smartphone or tablet.

[1044] A "portable terminal" is an electronic device that can be worn by a user, such as a glasses-type display or a smart watch.

[1045] A "display device" is a hardware device that can visually display characters, images, etc.

[1046] This invention relates to a "visual telephone" system that analyzes voice data and provides visual supplementation. This system converts voice data into text data in real time, analyzes it using a generative AI, infers the speaker's intention and context, and provides visual supplementation. Specific embodiments of this system are described below.

[1047] System configuration

[1048] server

[1049] The server is the central part of the system and includes the following functions:

[1050] 1. Means for receiving voice data: The server receives voice data from the telephone. This reception is performed in real time.

[1051] 2. Means for converting voice data to text data: The received voice data is converted to text data using a voice recognition module. Specifically, the Python SpeechRecognition library is used.

[1052] 3. Means of inferring context and speaker intent: Generative AI analyzes text data to understand the context and speaker intent. This process uses the Hugging Face transformers library.

[1053] 4. Means of completing inferred intent and correcting ambiguities: Complete inferred intent and make corrections as necessary.

[1054] 5. Means for visually displaying the completed text data: The completed text data is transmitted to the user's personal digital assistant or portable terminal.

[1055] Terminal

[1056] A terminal is a device through which a user receives visual information. It has the following functions:

[1057] 1. Display device: A display device such as a smartphone or smart glasses visually displays the text data sent from the server, highlighting important keywords and phrases.

[1058] User

[1059] Users are the main actors in communication via telephone and receive visually supplemented information through their devices. For example, customer support staff at a brick-and-mortar store can wear smart glasses and understand customer inquiries in real time.

[1060] Program processing and specific examples

[1061] The server receives the voice data and converts it into text data using a speech recognition module. The converted text data is then analyzed by a generative AI to infer the speaker's intent and context. This inferred intent and context is then augmented into a format that is easy to display visually and sent to the user's device.

[1062] Hardware and software used

[1063] Server: Receives voice data, converts it into text, and generates AI analysis

[1064] Software: SpeechRecognition, transformers (natural language processing)

[1065] Devices: smart glasses, smartphone

[1066] Display device: smart glasses display

[1067] Specific examples

[1068] Let's take the example of a user (a customer support staff member at a brick-and-mortar store) receiving a phone call from a customer asking, "Please check if the product is in stock." At this time, the server receives the voice data and converts it into text data saying, "Please check if the product is in stock." The generation AI analyzes this text data, infers that the product's stock information needs to be checked, and displays this information on the smart glasses' display.

[1069] Example prompt sentence:

[1070] I want to check if the product is in stock

[1071] This example allows the user (customer support staff) to immediately take appropriate action based on the visually supplemented information. This system eliminates ambiguity in voice data and enables efficient information transmission.

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

[1073] Step 1:

[1074] The server receives the voice data. Specifically, the server captures the telephone voice signal in real time and stores it in a buffer as voice data. The input is the telephone voice signal, and the output is the voice data stored in the buffer. This prepares the server to pass the received voice data to the next process.

[1075] Step 2:

[1076] The server converts the voice data into text data. Specifically, it uses a speech recognition module (SpeechRecognition library) to convert the voice data stored in the buffer into text data. The input is the voice data stored in the buffer, and the output is the converted text data. This data processing involves phonemic analysis and conversion into words.

[1077] Step 3:

[1078] The server uses a generative AI model to analyze text data and infer the context and speaker intent. Specifically, the text data is sent to a natural language processing module (the transformers library) to analyze the context and intent. The input is the transformed text data, and the output is information about the analyzed intent and context. This data processing includes grammatical analysis and context understanding.

[1079] Step 4:

[1080] The server completes the inferred intent and corrects any unclear parts. Specifically, based on the analysis results of the generative AI, any unclear parts in the utterance are corrected and the final intent is determined. The input is the analyzed intent and context information, and the output is the completed text data. A completion algorithm is used in this process.

[1081] Step 5:

[1082] The server sends the completed text data to the device. Specifically, the completed text data is sent to the device (smart glasses or smartphone) via a network. The input is the completed text data, and the output is the state after transmission to the device is complete. Network communication technology is used for this data transfer.

[1083] Step 6:

[1084] The terminal visually displays the received text completion data. Specifically, the received text completion data is displayed on the display of smart glasses or a smartphone, highlighting important keywords and phrases. The input is the text completion data sent from the server, and the output is the visual information displayed on the display.

[1085] Step 7:

[1086] The user checks the visual information and responds accordingly. Specifically, the user (customer support staff) looks at the information displayed on the smart glasses display, accurately understands the content of the call, and responds appropriately to the customer. The input is the visual information displayed on the display, and the output is the user's appropriate response.

[1087] 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.

[1088] This invention relates to a "visual telephone" system that analyzes telephone voices and visually complements them to eliminate difficulties in hearing calls. Furthermore, this invention enables more intuitive and effective communication by incorporating an emotion engine that recognizes the user's emotions. This system operates in cooperation with the server, terminal, and user, analyzing voice data and displaying it visually in real time.

[1089] 1. System Configuration

[1090] server

[1091] The server is the central device that receives telephone voice data and processes it in real time. The server first receives the voice data and sequentially transfers it to the voice recognition module. The voice recognition module converts the voice data into text data. This converted text data is analyzed by the generative AI to infer the speaker's intention and context. In addition, the emotion engine analyzes the user's tone and intonation of voice to identify the user's emotions. Finally, the inferred intention and user emotion are complemented and converted into a format that is easy to display visually.

[1092] Terminal

[1093] The terminal is a device that allows the user to receive information visually. It receives the completed text data sent from the server and displays it on the screen. The terminal is made up of a television or a dedicated display, and has the function of highlighting important keywords and phrases. The user looks at this display to understand the content of the call.

[1094] User

[1095] The user is the subject of communication via telephone, and receives visually supplemented information through the terminal. The user can read the displayed text data and make requests or follow-up inquiries as needed. This allows the user to visually confirm content that is difficult to understand through voice alone, and also obtain information that reflects emotions.

[1096] 2. Program Processing

[1097] Audio Receiving and Forwarding

[1098] The server receives the voice data and transfers it to the voice recognition module in real time. The server stores the voice data in a buffer and supplies the data to the voice recognition module as needed.

[1099] Speech-to-text conversion

[1100] The speech recognition module converts the received voice data into text data in real time, including phonemic analysis of the voice and converting it into words, and the generated text data is sent to the analysis module.

[1101] Contextual analysis and intent inference

[1102] The generative AI (analysis module) analyzes text data and performs analysis to understand the context and the speaker's intent. This includes grammatical analysis and natural language processing to understand the context and infer the speaker's intent.

[1103] emotion recognition

[1104] The emotion engine analyzes the tone and intonation of the user's voice data to recognize the user's emotions. The emotion engine then classifies emotions using an emotion dictionary and acoustic model, and provides the identification results to the generation AI.

[1105] Visual Indication

[1106] The server sends the completed text data and emotion recognition results to the device, which then displays them visually. Important keywords and phrases are highlighted to make them easier for users to understand. Additionally, the user's emotions are also displayed, supporting a more intuitive understanding.

[1107] 3. Specific Examples

[1108] For example, consider the case where a user calls customer support to inquire about an internet connection problem. The server receives the voice data "The internet is not connected," and the speech recognition module converts it into text data "The internet is not connected." The generation AI analyzes this text data and complements it with the user's intent, "I want to solve my internet connection problem." The emotion engine identifies urgency and anxiety from the user's tone of voice and reflects this in the display. This complemented text data is sent to the device, and the user can visually confirm emotions such as "urgency" and "anxiety" along with the text "I want to solve my internet connection problem."

[1109] This system allows users to visually confirm content that is difficult to understand through audio alone, making telephone communication smoother and providing comprehensive information that takes emotions into consideration.

[1110] The processing flow will be explained below.

[1111] Step 1:

[1112] The server receives the voice data from the call. The voice data is captured digitally and stored in a buffer, where it is filtered to preserve the quality of the voice data.

[1113] Step 2:

[1114] The server transfers the buffered voice data to the speech recognition module in real time, using streaming technology to minimize data delays.

[1115] Step 3:

[1116] The speech recognition module analyzes the received speech data and breaks it down into phonemes, converts the phonemes into words, and organizes the generated words into text data.

[1117] Step 4:

[1118] The text data generated by the speech recognition module is sent to the analysis module, which receives the text data and analyzes the context of the text data using grammar analysis and natural language processing.

[1119] Step 5:

[1120] The generative AI (analysis module) understands the context of the text data and infers the speaker's intention. Based on the analysis results, it further examines the inferred intention and fills in any unclear parts.

[1121] Step 6:

[1122] The emotion engine analyzes the voice data received from the server and identifies emotions based on the tone and intonation of the user's voice. The emotion engine performs emotion classification using an acoustic model and an emotion dictionary.

[1123] Step 7:

[1124] The emotion engine passes the identified emotion information to the generation AI, which then complements the emotion information along with the inferred intent. This allows the user's emotions to be reflected in the text data.

[1125] Step 8:

[1126] The server sends the completed text data and emotion information to the device. The data is sent in a format that is easy for the user to understand, and is formatted to emphasize important keywords and emotions.

[1127] Step 9:

[1128] The terminal receives the data sent from the server and visually displays it to the user. The display screen combines the text data with the user's emotions, highlighting important keywords and emotions.

[1129] Step 10:

[1130] The user can read the text data and emotional information displayed on the device and check the contents of the call. If necessary, the user can make a request or ask again. This operation allows for more detailed information to be obtained.

[1131] Example 2

[1132] 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."

[1133] In conventional telephone communication, voice alone can make the content unclear, and the information conveyed can be inaccurate, especially in noisy environments or when the voice quality is poor. It is also difficult to accurately understand emotions and intentions through voice alone, making misunderstandings highly likely, especially in communication that involves emotions. This can lead to problems that reduce user satisfaction, especially in user support and customer service.

[1134] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving voice data, means for converting the received voice data into text data, means for analyzing the converted text data and inferring the context and the speaker's intention, means for analyzing the speaker's tone and intonation to identify the speaker's emotion, means for complementing the inferred intention and the identified emotion and correcting unclear parts, and means for visually displaying the complemented text data and the identified emotion. As a result, by converting voice data into text data in real time and performing context analysis and emotion identification, it becomes possible to visually complement information that is difficult to convey through voice alone.

[1135] The "means for receiving voice data" is a function that receives voice data in real time from a telephone line, VoIP, etc., and temporarily stores it in a buffer.

[1136] The "means for converting into text data" is a function that uses a voice recognition module to analyze voice data, perform phonemic analysis, and convert into words to generate text data.

[1137] "Means of inferring context and speaker intent" refers to a function that uses generative AI to analyze text data and understand the speaker's intent and context through grammatical analysis and natural language processing.

[1138] "Means for identifying emotions by analyzing voice tone and intonation" is a function that uses an emotion engine to analyze the intonation and strength of the voice in the audio data, and classify and identify emotions.

[1139] "Means to complement inferred intent and identified emotion and correct unclear parts" is a function that uses the analysis results obtained by generative AI and emotion engines to complement intent and emotion and make appropriate corrections to clarify unclear parts.

[1140] The "means for visually displaying the complemented text data and the identified emotion" is a function that uses the display device of the terminal to display the text data and the identified emotion on the screen in a form that is easy for the user to check.

[1141] This invention relates to a "visual telephone" system that eliminates the difficulty of hearing telephone calls by analyzing voice data and visually complementing it. This system analyzes voice data, infers the user's intentions and emotions, and visually displays them, enabling more intuitive and effective communication.

[1142] The specific system configuration mainly involves the collaboration of a server, a terminal, and a user.

[1143] server

[1144] The server is the central device that receives and processes telephone voice data in real time. First, the server buffers the voice data from the telephone line or VoIP. Next, it converts the voice data into text using a speech recognition module (e.g., Google Speech-to-Text API). The converted text data is analyzed by a generative AI (e.g., OpenAI GPT-4) to understand the speaker's intent and context. It also uses an emotion engine (e.g., Affectiva SDK) to analyze the user's tone and intonation to identify emotions. The server then integrates the inferred intent and identified emotions and presents them in a format that is easy to display visually.

[1145] Terminal

[1146] The terminal is a device that allows the user to visually receive information. It receives the supplemented text data and emotion recognition results sent from the server and displays them on the terminal's display device (e.g., tablet, PC, smartphone). This display also includes a function to highlight important keywords and phrases, making it easy for the user to grasp the information.

[1147] User

[1148] The user is the subject of communication via telephone, and receives visually supplemented information through the terminal. The user can read the displayed text data and make requests or follow-up inquiries as needed. This allows the user to visually confirm content that is difficult to understand through voice alone, and also obtain information that reflects emotions.

[1149] Specific examples

[1150] For example, consider the case where a user calls customer support to inquire about an internet connection problem. In this case, the server receives the voice data "The internet is not connecting," and the speech recognition module converts it into text data "The internet is not connecting." The generation AI analyzes this text data and infers the user's intention, which is "I want to solve my internet connection problem." Furthermore, the emotion engine identifies "urgency" and "anxiety" from the user's tone of voice and reflects these in the display. This complemented text data is sent to the device, and the user can visually confirm emotions such as "urgency" and "anxiety" along with the text data "I want to solve my internet connection problem."

[1151] Prompt Sentence Examples

[1152] User: "I can't connect to the internet, what should I do?"

[1153] Generation AI:

[1154] 1. Analyze user comments and convert them into text.

[1155] 2. Infer the user's intent from the text (e.g., I want to fix my internet connection problem).

[1156] 3. Identify urgency and anxiety from the user's tone of voice.

[1157] 4. The final results are compiled into a text and sent to the device.

[1158] This system allows users to visually confirm content that is difficult to understand through audio alone, making telephone communication smoother and providing comprehensive information that takes emotions into consideration.

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

[1160] Step 1:

[1161] Voice data reception

[1162] The server receives voice data in real time from a telephone line or VoIP.

[1163] Input: Voice data from telephone line or VoIP

[1164] Output: Audio data stored in the server's buffer

[1165] Specific operation: The server receives audio data via the API and stores it in a buffer.

[1166] Step 2:

[1167] Converting audio data to text

[1168] The server converts the voice data into text data using a voice recognition module.

[1169] Input: Audio data stored in the server's buffer

[1170] Output: Text data

[1171] Specific operation: The server inputs the voice data into a speech recognition API (e.g., Google Speech-to-Text API) and obtains the text data "Internet connection is down."

[1172] Step 3:

[1173] Contextual analysis and intent inference of text data

[1174] The server uses generative AI to analyze the text data and infer the context and speaker's intent.

[1175] Input: Text data obtained from the speech recognition module

[1176] Output: Inferred intent and contextually-based complementation data

[1177] Specific operation: The server inputs the text data into a generation AI (e.g., GPT-4) and obtains the intent "I want to solve the Internet connection problem" as the analysis result.

[1178] Step 4:

[1179] emotion recognition

[1180] The server uses an emotion engine to analyze the tone and intonation of the voice based on the voice data and identify the user's emotions.

[1181] Input: Audio data stored on the server

[1182] Output: Identified emotion data (e.g., urgency, anxiety)

[1183] Specific operation: The server uses an emotion engine (e.g., Affectiva SDK) to identify emotions such as "anxiety" from the voice data.

[1184] Step 5:

[1185] Visual display of complementary data

[1186] The server transmits the completed text data and the identified emotion data to the terminal, which then visually displays them.

[1187] Input: Complemented text data, identified emotion data

[1188] Output: Visually displayed data (text and sentiment)

[1189] Specific operation: The server sends the data to the terminal, and the terminal displays the text "I want to solve the problem of the Internet connection" and the emotion "anxiety" highlighted on the display device (e.g., tablet, PC).

[1190] The above is the specific processing flow of the program of this system.

[1191] (Application example 2)

[1192] 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."

[1193] In current telephone communications, it can be difficult to understand what is being said through voice alone. Furthermore, the user's emotions and intentions may not be conveyed accurately. Smooth communication is especially important in brick-and-mortar stores and customer support settings. However, existing systems lack a means to visually complement emotions and intentions, which reduces the efficiency and effectiveness of communication.

[1194] 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 receiving voice data, means for converting the received voice data into text data, means for analyzing the converted text data and inferring the context and the speaker's intention, means for completing the inferred intention and correcting unclear parts, means for analyzing the tone and intonation of the user's voice using an emotion engine to recognize emotions, and means for integrating the results of the emotion analysis into visually displayed data and displaying the results. This not only makes it possible to visually grasp content that is difficult to understand from voice alone, but also makes it possible to provide information that reflects the user's emotions, thereby enabling smooth communication between customers and staff, particularly in customer support situations in physical stores.

[1195] The "audio data receiving means" is a part of the system that has the function of receiving audio data transmitted via voice calls and the like.

[1196] The "text conversion means" is a part of the system that has the function of analyzing received voice data and converting it into text data.

[1197] The "intention inference means" is a part of the system that has the function of analyzing and understanding the context and the speaker's intention based on the converted text data.

[1198] The "intention completion means" is a part of the system that has the function of completing the speaker's inferred intention and correcting any unclear parts.

[1199] The "visual display means" is a part of the system that has the function of visually displaying the analyzed text data and emotion recognition results.

[1200] The "emotion engine" is a part of the system that has the ability to recognize the user's emotions by analyzing the tone and intonation of the voice.

[1201] The "emotion analysis result display means" is a part of the system that has the function of visually displaying the results of the emotion analysis by the emotion engine.

[1202] This invention relates to a customer support system for brick-and-mortar stores. This system receives and analyzes voice data and delivers information visually to users, thereby realizing smooth communication.

[1203] A specific example of the system is a customer support robot in a brick-and-mortar store that receives voice data from customers in real time, converts it into text, and then analyzes and visually displays the customer's intentions and emotions.

[1204] The server first receives the voice data. Using a voice data receiving means, it receives voice data transmitted via telephone or voice call in real time. Next, it converts the received voice data into text data using a text conversion means. For example, it uses a speech recognition library (speech_recognition). The converted text data is analyzed using an intent inference means and an intent completion means to infer the context and the speaker's intent. This process uses a generative AI model and natural language processing technology (TextBlob).

[1205] Furthermore, an emotion engine is used to analyze the tone and intonation of the voice to recognize the user's emotions. Specifically, an emotion analysis library (e.g., emotion dictionary and acoustic model) is used. This makes it possible to visually display the emotion analysis result together with the text data using an emotion analysis result display means. In-store displays and the screens of support robots are used as visual display means. The analyzed text data and the emotion results are displayed integrated on this screen.

[1206] As a concrete example, consider a scenario in which a customer asks a robot about a product. In this case, the server receives the voice data "Do you have this product in stock?" and converts it into text data "Do you have this product in stock?" using a speech recognition library. Next, a generative AI model analyzes this text data and infers the customer's intent in the form of "Check product availability." The emotion engine identifies urgency or anxiety from the customer's tone of voice and reflects this in the display.

[1207] Finally, as a visual display, emotions such as "urgent" or "anxiety" are displayed on in-store displays and on the screens of support robots along with the text "Check product inventory." This allows customers to visually confirm content that would be difficult to understand through voice alone, and enables store staff to accurately grasp the customer's intentions and emotions.

[1208] Examples of input prompts for generative AI models include:

[1209] "A customer is asking, 'Does the store have this item in stock?' Visually display this question in real time and use an emotion recognition engine to read the customer's emotions (urgency, anxiety, etc.)."

[1210] Using these prompts, the system visually displays customer questions based on text and emotion recognition, allowing store staff to accurately understand the customer's needs.

[1211] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1212] Step 1:

[1213] The server receives audio data sent via telephone or voice call. The input is audio data from a microphone or a call system. The output is to store the received audio data in a buffer. Specifically, the audio receiving module converts audio into digital data and processes it in a streaming format.

[1214] Step 2:

[1215] The server converts voice data into text data in real time. The input is the voice data stored in the buffer. The output is the converted text data. Specifically, it uses a speech recognition library (e.g., speech_recognition) to analyze the voice data into phonemes and output them as text.

[1216] Step 3:

[1217] The server analyzes the text data and infers the context and the speaker's intent. The input is the converted text data. The output is the inferred intent and its analysis results. Specific operations include using generative AI models and natural language processing techniques (e.g., TextBlob) to analyze the text grammatically and semantically.

[1218] Step 4:

[1219] The server completes the inferred intent and corrects any ambiguities. The input is the inferred intent and the text data. The output is the completed and corrected text data. Specifically, the analysis module reevaluates the context and intent and adds additional information as needed.

[1220] Step 5:

[1221] The server uses an emotion engine to analyze the emotions in the voice data. The input is the voice data and its text conversion results. The output is the emotion recognition results. Specifically, it analyzes the tone and intonation of the voice and classifies the emotion using an emotion dictionary and acoustic model.

[1222] Step 6:

[1223] The server sends the completed text data and the emotion analysis results to the terminal for visual display. The input is the completed text data and the emotion analysis results. The output is data for visual display that includes these data. Specifically, the data formatting module organizes this information and formats it into a format that is easy to display.

[1224] Step 7:

[1225] The terminal receives the data sent from the server and displays it visually. The input is the data for visual display sent from the server. The output is the text data and sentiment analysis results displayed on the display. Specifically, the terminal's display module highlights important keywords and phrases, providing information in a format that is easy for the user to review.

[1226] Step 8:

[1227] The user checks the visually displayed information and makes requests or re-inquires as necessary. The input is the text data displayed on the screen and the results of sentiment analysis. The output is the user's understanding and the next action. In concrete terms, the user checks the display and smoothly proceeds with the response via phone or chat.

[1228] 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.

[1229] 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.

[1230] 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.

[1231] 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.

[1232] 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.

[1233] 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.

[1234] 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).

[1235] 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.

[1236] 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."

[1237] 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.

[1238] 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).

[1239] 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.

[1240] 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.

[1241] 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.

[1242] 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.

[1243] 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.

[1244] 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.

[1245] 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.

[1246] 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.

[1247] 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.

[1248] 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.

[1249] The following is further disclosed regarding the above embodiment.

[1250] (Claim 1)

[1251] means for receiving audio data;

[1252] means for converting received voice data into text data;

[1253] A means for analyzing the converted text data and inferring the context and the speaker's intent;

[1254] a means of completing inferred intent and correcting ambiguities;

[1255] a means for visually displaying the completed text data;

[1256] A system including:

[1257] (Claim 2)

[1258] 10. The system of claim 1, further comprising means for receiving and transcribing audio data in real time.

[1259] (Claim 3)

[1260] 10. The system of claim 1, further comprising means for complementing emotion and intent based on intonation and stress of the voice during analysis.

[1261] (Claim 4)

[1262] 10. The system of claim 1, further comprising means for highlighting important keywords or phrases in the completed text data.

[1263] "Example 1"

[1264] (Claim 1)

[1265] means for receiving audio data;

[1266] means for converting received voice data into text data;

[1267] A means for analyzing the converted text data and inferring the context and the speaker's intent;

[1268] a means of completing inferred intent and correcting ambiguities;

[1269] a means for visually displaying the completed text data;

[1270] A means for the server to sequentially supply audio data;

[1271] means for transmitting the analyzed text data to a terminal;

[1272] means for the terminal to highlight text data;

[1273] A system including:

[1274] (Claim 2)

[1275] 10. The system of claim 1, further comprising means for receiving and transcribing audio data in real time.

[1276] (Claim 3)

[1277] 10. The system of claim 1, further comprising means for complementing emotion and intent based on intonation and stress of the voice during analysis.

[1278] "Application Example 1"

[1279] (Claim 1)

[1280] means for receiving audio data;

[1281] means for converting received voice data into text data;

[1282] A means for analyzing the converted text data and inferring the context and the speaker's intent;

[1283] a means of completing inferred intent and correcting ambiguities;

[1284] a means for visually displaying the completed text data;

[1285] A means for using a display device of a portable information terminal or a portable terminal as a display means, so that a user can visually confirm information;

[1286] A system including:

[1287] (Claim 2)

[1288] 10. The system of claim 1, further comprising means for receiving and transcribing audio data in real time.

[1289] (Claim 3)

[1290] 10. The system of claim 1, further comprising means for complementing emotion and intent based on intonation and stress of the voice during analysis.

[1291] "Example 2: Combining Emotion Engines"

[1292] (Claim 1)

[1293] means for receiving audio data;

[1294] means for converting received voice data into text data;

[1295] A means for analyzing the converted text data and inferring the context and the speaker's intent;

[1296] A means for identifying emotions by analyzing the tone and intonation of a speaker's voice;

[1297] A means of completing the inferred intent and identified emotion and correcting any ambiguities;

[1298] a means for visually displaying the completed text data and the identified emotions;

[1299] A system including:

[1300] (Claim 2)

[1301] 10. The system of claim 1, further comprising means for receiving and transcribing audio data in real time.

[1302] (Claim 3)

[1303] 10. The system of claim 1, further comprising means for complementing emotion and intent based on intonation and stress of the voice during analysis.

[1304] "Application example 2 when combining emotion engines"

[1305] (Claim 1)

[1306] means for receiving audio data;

[1307] means for converting received voice data into text data;

[1308] A means for analyzing the converted text data and inferring the context and the speaker's intent;

[1309] a means of completing inferred intent and correcting ambiguities;

[1310] a means for visually displaying the completed text data;

[1311] a means for analyzing the tone and intonation of a user's voice using an emotion engine to recognize emotions;

[1312] a means for integrating and displaying the results of sentiment analysis into the visually displayed data;

[1313] A system including:

[1314] (Claim 2)

[1315] 10. The system of claim 1, further comprising means for receiving and transcribing audio data in real time.

[1316] (Claim 3)

[1317] 10. The system of claim 1, further comprising means for complementing emotion and intent based on intonation and stress of the voice during analysis. [Explanation of symbols]

[1318] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving audio data; means for converting received voice data into text data; A means for analyzing the converted text data and inferring the context and the speaker's intent; a means of completing inferred intent and correcting ambiguities; a means for visually displaying the completed text data; A system including:

2. The system of claim 1 , further comprising means for receiving and transcribing audio data in real time.

3. The system according to claim 1, further comprising means for complementing emotions and intentions based on intonation and stress of the voice during analysis.

4. 10. The system of claim 1, further comprising means for highlighting important keywords and phrases in the completed text data.

Images