system

Abstract

We provide the system. [Solution] A means for capturing audio data, Means for transmitting the aforementioned audio data to an edge computing server, The edge computing server includes means for converting the audio data into text data, A means for translating the aforementioned text data into a specified language, Means for transmitting the translated text data to the user terminal, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In modern globalized society, language barriers have become a major obstacle hindering smooth communication in various fields such as business, tourism, and education. Conventional translation services have difficulties in real-time response and have technical problems such as high communication latency and insufficient utilization of edge computing. In particular, there is a need for a system that can convert voice data into text in real time and translate it immediately. Against this background, a technical solution for providing real-time voice translation with ultra-low latency is required.

Means for Solving the Problems

[0005] The present invention provides a system that captures audio data, transmits it to an edge computing server, converts the audio data into text data, and further translates the text data into a specified language. This system includes means for capturing audio data, means for transmitting the audio data to an edge computing server, means for the edge computing server to convert the audio data into text data, means for translating the text data into a specified language, and means for transmitting the translated text data to a user terminal. As a result, real-time conversion and translation of audio data are achieved with ultra-low latency, effectively overcoming language barriers.

[0006] "Audio data" refers to digital data of audio signals generated by a user using an audio input device such as a microphone.

[0007] "Capture" refers to the process of taking in audio signals through a microphone device and saving them as digital data.

[0008] An "edge computing server" is a server located close to the user's terminal and is a device used for low-latency data processing.

[0009] "Transmission" refers to the process of transferring data from a user terminal to an edge computing server, or from an edge computing server to a user terminal, over a network.

[0010] "Text data" refers to string data of natural language converted by speech recognition processing.

[0011] "Translation" is the process of converting text expressed in one language into another language.

[0012] A "user terminal" is a device that is directly operated by the user and has functions for voice input, data transmission and reception, and result display. [Brief explanation of the drawing]

[0013] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2, which combines an emotion engine. [Modes for carrying out the invention]

[0014] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

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

[0016] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0017] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0018] In the following embodiments, the numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0019] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0020] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0021] [First Embodiment]

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

[0023] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0024] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0025] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0026] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0027] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0028] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

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

[0030] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0031] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0032] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0033] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0034] This invention is a system that captures audio data on a user terminal, transmits it to an edge computing server, converts the audio data into text data there, translates it into a specified language, and finally returns the translated text data to the user terminal. Embodiments of the present invention will be described in detail below with specific examples.

[0035] User terminal operation

[0036] 1. Audio Capture

[0037] When a user begins a conversation, the microphone device on the user's device captures the audio. This audio is stored on the device as digital data. Because this audio data is captured at short intervals, real-time functionality can be maintained.

[0038] 2. Sending to the edge server

[0039] The captured audio data is converted to binary format and sent to the edge server using an HTTP request. This allows the user terminal to send the audio data for immediate processing.

[0040] 3. Retrieving text data from edge servers

[0041] The edge server processes the audio data and generates text data, which the user terminal receives as an HTTP response. The user terminal then uses this text data in the next processing stage.

[0042] 4. Submitting a translation request

[0043] The system then sends a request to the edge server to translate the received text data into the target language specified by the user.

[0044] 5. Retrieving and displaying translated data

[0045] The system receives translated text data from the edge server and displays this data to the user. This allows the user to see what they said in a translated format in a different language.

[0046] Edge server operation

[0047] 1. Receiving audio data

[0048] The edge server receives audio data transmitted from the user's terminal. The edge server then prepares this data for analysis.

[0049] 2. Speech Recognition Processing

[0050] The edge server analyzes the received audio data using a speech recognition system and converts it into text data. This process employs advanced speech recognition algorithms.

[0051] 3. Return of text data

[0052] The generated text data is sent back to the user's terminal. This allows the user's terminal to begin the next translation process.

[0053] 4. Receiving a translation request

[0054] It receives translation requests for text data sent from the user's terminal. This is for converting text generated by speech recognition into a specified target language.

[0055] 5. Text translation processing

[0056] The received text data is translated into the target language. The edge server uses a high-performance translation engine to perform this transfer.

[0057] 6. Return of translated data

[0058] The translated text data is sent back to the user's terminal. This allows the user to actually check the translation results.

[0059] Specific examples

[0060] For example, a Japanese-speaking user says the phrase "Hello, how are you?" to the user terminal. This audio data is immediately captured and sent to the edge server. The edge server uses a speech recognition engine to convert this audio data into text "Hello, how are you?" and sends it back to the user terminal. Next, the user terminal resends this text data to the edge server, specifying English as the target language. The edge server translates the text data into English "Hello, how are you?" and finally sends it back to the user terminal. The user terminal displays this result, and the user can confirm the translation.

[0061] Through the process described above, the system of the present invention can translate speech into different languages ​​in real time and with ultra-low latency, and provide it to the user. This system has high practical applications in various fields such as business, tourism, and education, and can effectively overcome language barriers.

[0062] The following describes the processing flow.

[0063] Step 1:

[0064] When a user begins voice input, the user's device captures the audio data using its microphone. Specifically, it converts the user's spoken voice into a digital format in real time and temporarily stores it on the device.

[0065] Step 2:

[0066] The device converts the captured audio data into binary format and sends it to a specified endpoint on the edge computing server (e.g., / transcribe) using an HTTP request. The audio data is attached to this request.

[0067] Step 3:

[0068] The edge server receives the audio data and the speech recognition engine performs the analysis. Specifically, it initiates the process of converting the audio data into text data.

[0069] Step 4:

[0070] The edge server sends the text data generated as a result of speech recognition back to the user terminal as an HTTP response. The user terminal receives this response and extracts the text data.

[0071] Step 5:

[0072] The user's terminal uses the received text data to resend the translation request to the edge computing server. This request includes information about the original text and the target language.

[0073] Step 6:

[0074] The edge server processes the received translation request and translates the text data into the specified target language. Specifically, it uses a translation engine on the server to translate the text.

[0075] Step 7:

[0076] The edge server sends the translated text data back to the user's terminal. The user terminal receives this response and extracts the translated text data.

[0077] Step 8:

[0078] The user's device displays the final translation result to the user. The user can check the translated text displayed on the screen.

[0079] (Example 1)

[0080] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0081] Conventional translation systems require real-time performance and accuracy in the process of capturing audio data, converting it to text, and then translating it. Especially in today's society where multilingual communication is essential, systems that perform these processes quickly and efficiently are indispensable. However, delays often occur between capturing audio data and obtaining translated text data, which is stressful for users. Furthermore, challenges exist regarding data transmission formats and security, making it difficult to balance system reliability and speed. Therefore, there is a need for the development of a faster and more accurate real-time speech translation system.

[0082] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0083] In this invention, the server includes means for capturing audio data, means for transmitting the audio data to an edge computing server, means for the edge computing server to convert the audio data into text data, means for translating the text data into a specified language, means for transmitting the translated text data to a user terminal, means for capturing and transmitting audio data at short intervals to maintain real-time performance, means for sending and receiving data between the edge computing server and the user terminal using HTTP requests and responses, and means for displaying the text data on the screen of the user terminal. This enables users to translate audio between multiple languages ​​in real time and accurately, allowing for rapid communication.

[0084] "Audio data" refers to data that records the voice spoken by a user in digital format.

[0085] "Means of capturing" refers to a device or method for acquiring a user's voice as digital data through a microphone device.

[0086] An "edge computing server" is a distributed computing resource located close to the user's terminal for processing voice data.

[0087] "Means of transmission" refers to methods or devices for transferring data from a user terminal to an edge computing server via communication.

[0088] "Means of converting to text data" refers to processes or systems that convert audio data into string data.

[0089] "Specified language" refers to the target language that the user wishes to have translated into.

[0090] "Translation means" refers to the process or system of converting generated text data into a specified target language.

[0091] "Means of sending and receiving" refers to communication methods for transmitting and receiving data.

[0092] "Maintaining real-time performance" means minimizing the delay between capturing audio data and displaying translated text data.

[0093] "Short interval" means processing or transmitting data in the shortest possible time units, such as every 100 milliseconds.

[0094] An "HTTP request" is a request from a client to a server to send data, and it is a protocol primarily used for web communication.

[0095] An "HTTP response" is a data response that a server provides to a client, and it refers to the result of an HTTP request.

[0096] "Means of display" refers to methods or devices for displaying text data on the screen of a user's terminal.

[0097] This invention is a system that captures audio data on a user terminal, sends it to an edge computing server, converts the audio data into text data there, translates it into a specified language, and finally sends the translated text data back to the user terminal.

[0098] User terminal operation

[0099] 1. Audio Capture

[0100] When a user begins a conversation, the microphone device on the user's device captures the audio. This audio is stored on the device as digital data. Because this audio data is captured at short intervals, real-time functionality can be maintained. For example, a smartphone's microphone picks up sound and converts it into digital data through an analog-to-digital converter.

[0101] 2. Sending to the edge server

[0102] The captured audio data is converted to binary format and sent to the edge server using an HTTP request. This allows the user terminal to send the audio data for processing immediately. The terminal application converts the audio data to binary format and sends it to the edge server using an HTTP library (e.g., OkHttp).

[0103] 3. Retrieving text data from edge servers

[0104] The edge server processes the audio data and generates text data, which the user terminal receives as an HTTP response. The user terminal uses this text data in the next processing stage. To receive the HTTP response, the terminal uses asynchronous communication to wait for the response from the server and saves the received text data to its internal storage or memory.

[0105] 4. Submitting a translation request

[0106] The application sends a request back to the edge server to translate the received text data into the target language specified by the user. The application creates an HTTP POST request, including the text and target language as parameters.

[0107] 5. Retrieving and displaying translated data

[0108] The system receives translated text data from the edge server and displays this data to the user. This allows the user to see what they said in a translated form in a different language. It receives an HTTP response and displays the translated text in a UI component (e.g., TextView) on the user's device.

[0109] Edge server operation

[0110] 1. Receiving audio data

[0111] The edge server receives audio data transmitted from the user's terminal. The edge server prepares this data for analysis. It receives an HTTP request, decodes the audio data, and temporarily stores it in storage.

[0112] 2. Speech Recognition Processing

[0113] The edge server analyzes the received audio data using a speech recognition system and converts it into text data. This process employs advanced speech recognition algorithms. For example, the Google® Cloud Speech-to-Text API can be used to convert audio data into text data. The audio data is input into the speech recognition system, and text data is generated. The generated text is temporarily stored in memory.

[0114] 3. Return of text data

[0115] The generated text data is sent back to the user's terminal. This allows the user's terminal to begin the next translation process. The text data is serialized into JSON format and sent as an HTTP response.

[0116] 4. Receiving a translation request

[0117] It receives translation requests for text data sent from the user's terminal. This is for converting text generated by speech recognition into a specified target language. It receives HTTP requests and parses the request parameters.

[0118] 5. Text translation processing

[0119] The received text data is translated into the target language. The edge server uses a high-performance translation engine to perform this transfer. For example, the Google Cloud Translation API is used to translate the text data. A request is sent to the translation API and the result is received.

[0120] 6. Return of translated data

[0121] The translated text data is sent back to the user's terminal. This allows the user to actually check the translation result. The translation result is sent as an HTTP response, and whether the communication was successful is recorded in the log.

[0122] Specific example

[0123] For example, a Japanese-speaking user says the phrase "Hello, how are you?" to the user terminal. This audio data is immediately captured and sent to the edge server. The edge server uses a speech recognition engine to convert this audio data into text "Hello, how are you?" and sends it back to the user terminal. Next, the user terminal resends this text data to the edge server, specifying English as the target language. The edge server translates the text data into English "Hello, how are you?" and finally sends it back to the user terminal. The user terminal displays this result, and the user can confirm the translation.

[0124] The following are specific examples of prompt statements to input to a generative AI model.

[0125] Example: "Convert the following Japanese audio to text, and then translate it into English. The audio says 'Hello, how are you?'"

[0126] This system can translate users' voices in real time and accurately, converting them into different languages. This makes it highly practical in diverse fields such as business, tourism, and education, effectively overcoming language barriers.

[0127] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0128] Step 1:

[0129] When a user begins a conversation, the microphone device on the user's device captures the audio. The input is an audio signal, and the output is digital data. Specifically, the smartphone's microphone picks up the audio and converts it into digital data through an ADC (analog-to-digital converter). This data is temporarily stored on the device.

[0130] Step 2:

[0131] The user terminal converts the captured audio data into binary format and sends it to the edge server using an HTTP POST request. The input is digital audio data, and the output is binary data. Specifically, the terminal's application converts the audio data into binary format and sends it to the edge server using a network library (e.g., OkHttp).

[0132] Step 3:

[0133] The edge server receives audio data sent from the user's terminal. The input is binary data sent in an HTTP request, and the output is audio data ready for analysis. Specifically, the server receives the HTTP request, decodes the binary data, and temporarily stores it in storage.

[0134] Step 4:

[0135] The edge server inputs the received audio data into a speech recognition engine and converts it into text data. The input is digital audio data, and the output is text data. Specifically, the edge server uses a speech recognition API (e.g., Google Cloud Speech-to-Text API) to analyze the audio and generate text data.

[0136] Step 5:

[0137] The edge server sends the generated text data to the user's terminal as an HTTP response. The input is text data, and the output is an HTTP response. Specifically, the server serializes the text data into JSON format and sends it to the user's terminal as an HTTP response.

[0138] Step 6:

[0139] The user terminal receives text data from the edge server and uses this data for the next processing step. The input is the text data sent in the HTTP response, and the output is the text data used in the request. Specifically, it receives the HTTP response and saves the text data to memory or internal storage.

[0140] Step 7:

[0141] The user terminal sends a request to the edge server to translate the received text data into the target language. The input is the text data and the specified target language, and the output is the translation request. Specifically, the application creates an HTTP POST request, including the text and target language as parameters.

[0142] Step 8:

[0143] The edge server receives translation requests from user terminals and inputs them into the translation engine. The input consists of text and the target language, and the output is translated text data. Specifically, the server calls a translation API (e.g., Google Cloud Translation API) to translate the text into the specified language.

[0144] Step 9:

[0145] The translated text data is sent back from the edge server to the user terminal. The input is the translated text data, and the output is the translated data as an HTTP response. Specifically, the server serializes the translated text into JSON format and sends it to the user terminal as an HTTP response.

[0146] Step 10:

[0147] The user terminal receives translated text data from the edge server and displays it to the user. The input is the translated data sent via HTTP response, and the output is the text data displayed on the user screen. Specifically, it receives the HTTP response and displays the translated text in a UI component (e.g., TextView). This allows the user to see the translated content in real time.

[0148] (Application Example 1)

[0149] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0150] In multilingual food delivery services, a challenge exists: users often find it difficult to place orders smoothly in their native language. Therefore, it is necessary to provide a system that allows users who speak different languages ​​to comfortably use the service without experiencing language barriers. Furthermore, utilizing voice input is required to provide a more intuitive and convenient ordering experience.

[0151] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0152] In this invention, the server includes means for capturing voice data, means for transmitting the voice data to an edge computing server, means for the edge computing server to convert the voice data into text data, means for translating the text data into a specified language, means for transmitting the translated text data to a user terminal and for the user terminal to display the translated text data, means for the user terminal to place an order based on voice input, and means for the edge computing server to process the order details and notify the delivery service. This enables users to order food delivery using voice input in different languages, providing a comfortable ordering experience that transcends language barriers.

[0153] "Audio data" refers to data obtained by converting audio signals input by a user through a microphone device into a digital format.

[0154] "Capture method" refers to a function that captures and records audio data using the microphone device on the user's terminal.

[0155] An "edge computing server" is a server that receives voice and text data sent from user terminals and processes it in real time.

[0156] "Text data" refers to data in string format generated from audio data through speech recognition.

[0157] A "means of translation" is a function for converting text data written in one language into another specified language.

[0158] A "user terminal" refers to a device used by a user, such as a smartphone or tablet, and is a device used for voice capture and data transmission / reception.

[0159] "Means of placing an order" refers to a function that allows users to confirm food delivery orders via voice input or other means and transmit the details to the system.

[0160] "Means of notifying delivery services" refers to a function that transmits order details processed on edge computing servers to food delivery personnel and operations systems.

[0161] "Means of display" refers to functions that visually present translated text data and order information to the user on the user's terminal.

[0162] This invention is a system that captures voice data and translates it into different languages ​​in real time via an edge computing server. It is particularly applicable to food delivery services, enabling a multilingual ordering process.

[0163] The user terminal first has the function to capture audio data. This includes devices such as smartphones and tablets, which use microphone devices to capture audio as digital data. The captured audio data is converted to binary format and sent to the edge computing server via an HTTP request.

[0164] The edge computing server converts received audio data into text data through speech recognition processing. This speech recognition uses speech recognition libraries such as Google Cloud Speech-to-Text or Amazon Transcribe. The converted text data is then translated into the language specified by the user. High-performance translation engines such as the Google Cloud Translation API or Microsoft® Translator Text API are used for the translation process.

[0165] The translated text data is sent back to the user's terminal and displayed on the user's screen. This allows the user to see the results of their voice input being translated into different languages. Furthermore, users can use voice input to place food delivery orders, which are processed on an edge computing server and notified to the delivery service.

[0166] For example, when a user says "I want to order curry" into their smartphone, this voice data is captured and sent to an edge server. The edge server converts the voice into text "I want to order curry," and then translates it into English as "I want to order curry." This translation is then displayed on the user's smartphone. The order details are also processed by the edge server and notified to the food delivery service.

[0167] An example of a prompt message is as follows:

[0168] User: "I want to order curry."

[0169] Method:

[0170] 1. Audio signal capture.

[0171] 2. Sent audio signal to edge server.

[0172] 3. Received text: "I want to order curry."

[0173] 4. Sent text for translation.

[0174] 5. Received translated text: "I want to order curry"

[0175] 6. Displaying translated text: "I want to order curry"

[0176] This allows users to order food delivery through voice input in different languages, without being hindered by language barriers. This system can also be used for a variety of purposes, including business, tourism, and education.

[0177] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0178] Step 1:

[0179] The user captures audio data using the microphone device on their smartphone or tablet. As input, the user's voice signal is converted into a digital format. As output, digital audio data is generated. This data is captured at short time intervals and immediately prepared to proceed to the next processing step.

[0180] Step 2:

[0181] The terminal converts the captured audio data into binary format and sends it to the edge computing server as an HTTP request. Digital audio data exists as input. The output is the audio data converted to binary format and configured as the payload for the HTTP request. A POST request is sent to the URL of the edge computing server.

[0182] Step 3:

[0183] The server receives an HTTP request and retrieves audio data. The input is audio data in binary format, delivered via the HTTP request. The output is decoded audio data generated on the server side. This is then passed to the speech recognition process.

[0184] Step 4:

[0185] The server performs speech recognition processing and converts the audio data into text data. The input is the decoded audio data. The output is the text data generated by speech recognition. This process utilizes speech recognition libraries such as Google Cloud Speech-to-Text and Amazon Transcribe.

[0186] Step 5:

[0187] The server returns the generated text data to the user's terminal. Text data exists as input. Text data is returned as an HTTP response as output. The user's terminal receives this data and proceeds to the next translation processing step.

[0188] Step 6:

[0189] The terminal sends a request back to the edge computing server to translate the received text data into the specified language. The input includes the text data and information about the target language. The output is an HTTP request sent back to the edge computing server.

[0190] Step 7:

[0191] The server receives a translation request and translates the text data into the target language. The input consists of the text data to be translated and information about the target language. The output is translated text data generated using a high-performance translation engine (e.g., Google Cloud Translation API or Microsoft Translator Text API).

[0192] Step 8:

[0193] The server returns the translated text data to the user's terminal. The input is the translated text data. The output is the translated text data returned as an HTTP response. The user's terminal receives this data and displays it visually to the user.

[0194] Step 9:

[0195] The device displays the translated text data on the screen for the user to review. The input is the translated text data. The output is the translation result displayed on the device screen. This allows the user to immediately see how their voice input has been translated.

[0196] Step 10:

[0197] When a user places an order using voice input, the terminal sends the order details to an edge computing server. The input is text data containing the user's order. The output is an HTTP request sent to the edge computing server. This allows the server to process the order information and notify the appropriate food delivery service.

[0198] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0199] The present invention is a system that captures voice data on a user terminal, transmits it to an edge computing server, converts the voice data into text data there, translates it into a specified language, performs emotion recognition using an emotion engine, and finally adds emotion information to the translated text data before returning it to the user terminal. Embodiments of the present invention will be described in detail below with specific examples.

[0200] User terminal operation

[0201] 1. Audio Capture

[0202] When a user begins a conversation, the microphone device on the user's device captures the audio. This audio is stored on the device as digital data. Because this audio data is captured at short intervals, real-time functionality can be maintained.

[0203] 2. Sending to the edge server

[0204] The captured audio data is converted to binary format and sent to a specified endpoint on the edge server (e.g., / transcribe) using an HTTP request. This allows the user terminal to send the audio data for immediate processing.

[0205] 3. Retrieving text data from edge servers

[0206] The edge server processes the audio data and generates text data, which the user terminal receives as an HTTP response. The user terminal then uses this text data in the next processing stage.

[0207] 4. Submitting a translation request

[0208] The system then sends a request to the edge server to translate the received text data into the target language specified by the user.

[0209] 5. Retrieving and displaying translated data

[0210] The system receives translated text data from the edge server and displays this data to the user. This allows the user to see what they said in a translated format in a different language.

[0211] Edge server operation

[0212] 1. Receiving audio data

[0213] The edge server receives audio data transmitted from the user's terminal. The edge server then prepares this data for analysis.

[0214] 2. Speech Recognition Processing

[0215] The edge server analyzes the received audio data using a speech recognition system and converts it into text data. This process employs advanced speech recognition algorithms.

[0216] 3. Return of text data

[0217] The generated text data is sent back to the user's terminal. This allows the user's terminal to begin the next translation process.

[0218] 4. Receiving a text translation request

[0219] It receives translation requests for text data sent from the user's terminal. This is for converting text generated by speech recognition into a specified target language.

[0220] 5. Text translation processing

[0221] The received text data is translated into the target language. The edge server uses a high-performance translation engine to perform this transfer.

[0222] 6. Performing emotion recognition

[0223] After translation, an emotion engine is used to analyze the emotions contained in the text data. Based on the emotion information obtained from the audio data, emotion tags are added to the text data.

[0224] 7. Return of translated data and sentiment information

[0225] The translated text data and added sentiment information are sent back to the user's terminal. This allows the user's terminal to display the translation result with the sentiment information included.

[0226] Specific examples

[0227] For example, a Japanese-speaking user says the phrase "Hello, how are you?" to the user terminal. This audio data is immediately captured and sent to the edge server. The edge server uses a speech recognition engine to convert this audio data into text "Hello, how are you?" and sends it back to the user terminal. Next, the user terminal resends this text data to the edge server, specifying English as the target language. The edge server translates the text data into English "Hello, how are you?" and uses an emotion engine to recognize the user's emotion (e.g., positive emotion) from the original Japanese audio data. This emotion information is added to the translation result and finally sent back to the user terminal. As a result, the user terminal displays the translated text with emotion information, such as "Hello, how are you? [Positive]".

[0228] Through the process described above, the system of the present invention can translate speech into different languages ​​in real time and with ultra-low latency, and can also add emotional information. This system has high practical applications in various fields such as business, tourism, and education, and helps to effectively overcome language barriers.

[0229] The following describes the processing flow.

[0230] Step 1:

[0231] When a user starts a conversation, their device captures the audio. Specifically, it uses a microphone to pick up the user's voice and converts it to a digital format in real time.

[0232] Step 2:

[0233] The device captures audio data, converts it to binary format, and sends it to the edge computing server using an HTTP request. This request includes the audio data.

[0234] Step 3:

[0235] The edge server receives the audio data and the speech recognition engine begins analysis. Specifically, the process of converting the audio data into text data is executed.

[0236] Step 4:

[0237] The edge server sends the text data generated as a result of speech recognition back to the user terminal as an HTTP response. The user terminal receives this response and extracts the text data.

[0238] Step 5:

[0239] The user's terminal uses the received text data to resend the translation request to the edge computing server. This request includes information about the original text and the target language.

[0240] Step 6:

[0241] The edge server receives the text translation request and uses a high-performance translation engine to translate the text data into the specified target language.

[0242] Step 7:

[0243] The edge server performs sentiment recognition on the translated text data. Specifically, the sentiment engine analyzes the audio data and text to evaluate relevant sentiment information.

[0244] Step 8:

[0245] The edge server adds sentiment information to the translation result and sends it to the user terminal as an HTTP response. The user terminal receives this response and extracts the translated text data and sentiment information.

[0246] Step 9:

[0247] The user's device displays the final translation result and sentiment information to the user. Specifically, it visually displays the translated text and its sentiment on the screen, making it easy for the user to understand.

[0248] (Example 2)

[0249] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0250] Conventional speech recognition systems could convert speech data to text and translate languages, but they could not add emotional information, making it difficult to achieve communication that reflected the user's emotions. Furthermore, they often lacked processing speed and real-time capabilities, limiting their practicality. As a result, they were unable to effectively overcome language barriers in fields such as business, tourism, and education.

[0251] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0252] In this invention, the server includes means for capturing audio data, means for transmitting the audio data to an edge computing server, means for the edge computing server to convert the audio data into text data, means for translating the text data into a specified language, means for adding sentiment information to the translated text data, and means for transmitting the translated text data and sentiment information to a user terminal. This enables the addition of sentiment information in addition to the conversion and translation of audio data into text. As a result, users can use text with sentiment information that has been translated in real time, enabling richer communication. Furthermore, high-speed edge computing improves processing speed and real-time capabilities, increasing its practicality in various fields.

[0253] "Audio data" refers to a digital audio signal acquired using a microphone device.

[0254] An "edge computing server" refers to a server located close to the user's terminal that performs data processing with an emphasis on real-time performance.

[0255] "Text data" refers to string data generated by analyzing audio data.

[0256] "Translation" refers to the process of converting text data expressed in one language into another language.

[0257] "Emotional information" refers to information that indicates a user's emotions (e.g., positive, negative, neutral) as recognized from text data or audio data.

[0258] An "HTTP request" refers to a request made by a client to a server using a protocol for sending data.

[0259] This invention relates to a system that captures audio data, converts it into text data in real time, translates it into a specified language, and adds emotional information. This system operates based on a user terminal and an edge computing server.

[0260] User terminal operation

[0261] First, the user terminal is equipped with a microphone device that captures the user's voice. For example, when the user says "Hello, how are you?", this voice is captured by the terminal as digital audio data. Next, the user terminal converts the captured audio data into binary format and sends it to a specified endpoint on the edge computing server using an HTTP request. An example of such an endpoint is " / transcribe".

[0262] Edge computing server-side operation

[0263] The edge computing server receives audio data sent from the user's terminal. This received audio data is converted into text data using a speech recognition system (the API used as an example provides a speech recognition algorithm). Examples of speech recognition systems that can be used include the Google Speech-to-Text API. The converted text data is sent back to the user's terminal via an HTTP response.

[0264] Next, the user terminal sends the acquired text data back to the edge computing server, requesting a translation into the specified language. This translation process uses a high-performance translation engine, such as the Google Translate API. Once the translation is complete, the server analyzes the translated text data with a sentiment engine (for example, IBM Watson® Tone Analyzer) and adds sentiment information. This translated data with added sentiment information is then sent back to the user terminal as an HTTP response.

[0265] Display of data by the user terminal

[0266] Finally, the user terminal displays the received translation data and sentiment information to the user. This allows the user to see their spoken content translated into different languages ​​in real time, with added sentiment information.

[0267] Specific example

[0268] For example, a Japanese-speaking user says the phrase "Hello, how are you?" to the user terminal. This audio data is immediately captured and sent to the edge computing server. The edge computing server uses a speech recognition engine to convert the audio data into text "Hello, how are you?" and sends it back to the user terminal. The user terminal resends this text data to the edge computing server, specifying English as the target language. The edge server translates the text data into English "Hello, how are you?" and uses an emotion engine to recognize the user's emotion (e.g., positive emotion) from the original Japanese audio data. This emotion information is added to the translation result and finally sent back to the user terminal. As a result, the user terminal displays the translated text with emotion information, such as "Hello, how are you? [Positive]".

[0269] Through the process described above, the system of the present invention can translate speech into different languages ​​in real time and with ultra-low latency, and can also add emotional information. This system has high practical applications in various fields such as business, tourism, and education, and helps to effectively overcome language barriers.

[0270] Example of a prompt

[0271] For example, by inputting the following prompt into the generating AI model, you can obtain detailed instructions for performing the aforementioned process:

[0272] Prompt message:

[0273] "Capture Japanese voice data, translate it into English, and add emotional information to the translation result for display."

[0274] Based on this prompt sentence, the generative AI model can propose specific technical means such as how to use appropriate APIs and how to specify endpoints.

[0275] The flow of the specific process in Example 2 will be described using FIG. 13.

[0276] Step 1:

[0277] The user starts voice capture.

[0278] Input: The user's voice (e.g., "Hello, how are you?")

[0279] Specific operation: The user starts speaking towards the microphone device of the terminal. The microphone device captures this voice as a digital signal.

[0280] Output: Voice data in digital format

[0281] Step 2:

[0282] The terminal acquires the voice data and sends it to the edge server.

[0283] Input: The captured digital voice data

[0284] Specific operation: The terminal converts the captured voice data into binary format and saves it in a temporary file. Then, it uses an HTTP request to send it to the specified endpoint (e.g., " / transcribe") of the edge server.

[0285] Output: The voice data sent to the edge server

[0286] Step 3:

[0287] The edge server converts voice data into text data.

[0288] Input: Binary voice data sent from the terminal

[0289] Specific operation: The edge server receives the voice data, analyzes the voice data using a voice recognition system (e.g., an API that provides a voice recognition algorithm), and converts it into text data.

[0290] Output: Converted text data (e.g., "Hello, how are you?")

[0291] Step 4:

[0292] The edge server returns the text data to the terminal.

[0293] Input: Text data generated by voice recognition

[0294] Specific operation: The edge server returns the generated text data to the user terminal as an HTTP response. The terminal saves this text data for use in the next process.

[0295] Output: Text data sent to the user terminal

[0296] Step 5:

[0297] The terminal sends the text data to the edge server and requests a translation.

[0298] Input: Text data received from the edge server

[0299] Specific operation: The user terminal sends a request to the edge server again to translate the saved text data into the target language. English is specified as the target language.

[0300] Output: Text data with a translation request sent to the edge server

[0301] Step 6:

[0302] The edge server translates the text data.

[0303] Input: Text data with a translation request sent from the user terminal

[0304] Specific operation: The edge server analyzes the received text data using a translation engine (e.g., an API that provides a translation algorithm) and translates it into the specified target language (here, English).

[0305] Output: Translated text data (e.g., "Hello, how are you?")

[0306] Step 7:

[0307] The edge server performs sentiment analysis on the translated data and adds sentiment information.

[0308] Input: Translated text data

[0309] Specific operation: The edge server analyzes the translated text data using a sentiment engine (e.g., an API that provides a sentiment analysis algorithm) to recognize the sentiment (e.g., positive, negative, neutral) contained in the text. This sentiment information is added as a tag to the translated data.

[0310] Output: Translated text data with sentiment information added (e.g., "Hello, how are you? [positive]")

[0311] Step 8:

[0312] The edge server returns the translated data and sentiment information to the terminal.

[0313] Input: Translated text data with sentiment information added

[0314] Specific operation: The edge server sends translated data with sentiment information back to the user terminal as an HTTP response. The user terminal receives this data.

[0315] Output: Translated data with sentiment information sent to the user's terminal.

[0316] Step 9:

[0317] The device displays translation data and sentiment information to the user.

[0318] Input: Translation data with emotional information

[0319] Specific operation: The user terminal displays the translated data with sentiment information received on the user interface. The user can then view the translated text and its sentiment information on the screen.

[0320] Output: Translated text with sentiment information displayed to the user (e.g., "Hello, how are you? [Positive]")

[0321] This process allows users to translate speech into different languages ​​in real time and then use it in a format that includes emotional information.

[0322] (Application Example 2)

[0323] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0324] In modern society, there is an increasing demand for multilingual support and real-time emotion recognition. However, conventional technologies are unable to adequately translate voice data in real time or add emotional information, leading to decreased communication efficiency. Therefore, there is a need for a system that integrates multilingual support and emotion recognition. In particular, in security services, smooth communication and situational awareness at the site are crucial, and technologies that can solve this problem are in demand.

[0325] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0326] In this invention, the server includes means for capturing voice data, means for transmitting the voice data to an edge computing server, means for the edge computing server to convert the voice data into text data, means for translating the text data into a specified language, means for adding sentiment information to the translated text data, and means for transmitting the translated text data and sentiment information to a user terminal. This enables multilingual real-time translation and sentiment recognition.

[0327] "Audio data" refers to data that represents audio in a digital format.

[0328] "Means of capturing" refers to devices or methods that record a user's voice and acquire it as digital data.

[0329] An "edge computing server" is a distributed server that processes data sent from a user's terminal in real time at a location close to the local area.

[0330] "Text data" refers to data that represents audio data in written form.

[0331] "Translation methods" refer to the techniques and methods used to convert text data expressed in the original language into another specified language.

[0332] "Means of adding emotional information" refers to technologies and methods that analyze the emotions contained in text data and add the results to the data as annotations.

[0333] A "user terminal" is a device (e.g., smart glasses, smartphone, etc.) used to capture audio data and communicate with a server.

[0334] This invention relates to a system that translates voice data into multiple languages ​​in real time and adds emotional information to that data. This system aims to improve effective communication and situational awareness, particularly in security services. The following describes embodiments for carrying out this invention using specific examples.

[0335] System Configuration

[0336] This system consists of user terminals (such as smart glasses or smartphones) that capture voice data and edge computing servers that process the data.

[0337] User terminal operation

[0338] 1. Audio Capture

[0339] When a user begins a conversation, the microphone device on the user's device captures the audio. The captured audio data is collected at short intervals, and the digital data is stored on the device.

[0340] 2. Sending to the edge computing server

[0341] The captured audio data is converted to binary format and sent to the edge computing server using an HTTP request.

[0342] Edge computing server operation

[0343] 1. Receiving and transcribing audio data

[0344] The edge computing server analyzes the audio data received from the user's terminal and converts it into text data using speech recognition software (e.g., Google Speech-to-Text API).

[0345] 2. Translation of text data

[0346] The converted text data is translated into the specified language. A high-performance translation engine (e.g., Google Translate API) is used for the translation.

[0347] 3. Adding emotional information

[0348] The translated text data is subjected to an emotion recognition engine, and emotional information (e.g., positive, negative, neutral) is added. An AI model (e.g., a BERT-based emotion analysis model) is used for emotion recognition.

[0349] Return to user terminal

[0350] The translated text data and sentiment information processed on the edge computing server are sent back to the user terminal via an HTTP response, which the user terminal then displays.

[0351] Specific example

[0352] A security guard is seen conversing with a foreign tourist.

[0353] 1. A tourist says in English, "Excuse me, can you help me find the museum?"

[0354] 2. The security guard's smart glasses capture this audio and send the data to an edge computing server.

[0355] 3. The server converts the speech to text and obtains "Excuse me, can you help me find the museum?".

[0356] 4. The text data is translated into Japanese, and the text "Excuse me, could you help me find the museum?" is generated.

[0357] 5. The translated text is tagged with sentiment and sent back to the user's device as "Excuse me, could you help me find the museum? [Positive]".

[0358] 6. The information is ultimately displayed on the security guard's smart glasses, allowing the guard to take appropriate action.

[0359] Example of a prompt

[0360] User's voice: "Excuse me, can you help me find the museum?"

[0361] Audio data capture:

[0362] Sending voice data to the edge server

[0363] Speech recognition and text conversion: "Excuse me, can you help me find the museum?"

[0364] Text translation: "Excuse me, could you help me find the museum?"

[0365] Emotional perception: "Positive"

[0366] Final message: "Excuse me, could you help me find the museum? [Positive]"

[0367] This enables the system to integrate multilingual support and emotion recognition, providing real-time feedback to users. Furthermore, it can support smooth communication and rapid decision-making in security services.

[0368] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0369] Step 1:

[0370] Audio Capture

[0371] The user's device (e.g., smart glasses, smartphone) captures the user's voice. When the user starts speaking, the device's microphone collects the voice and stores it as digital data. The input is voice data, and the output is digital voice data.

[0372] Step 2:

[0373] Sending digital audio data to an edge computing server

[0374] The terminal converts the captured digital audio data into binary format and sends it to the edge computing server using an HTTP request. The input is digital audio data, and the output is binary audio data sent as an HTTP request.

[0375] Step 3:

[0376] Receiving and transcribing audio data by the server.

[0377] The server receives binary audio data sent from the terminal. The received audio data is converted into text data using speech recognition software (e.g., Google Speech-to-Text API). The input is binary audio data, and the output is text data.

[0378] Step 4:

[0379] Text data translation

[0380] The server translates the converted text data into the specified language. A high-performance translation engine (e.g., Google Translate API) is used for the translation. The input is text data, and the output is translated text data in the target language.

[0381] Step 5:

[0382] Adding emotional information

[0383] The server analyzes the translated text data using an emotion recognition engine (e.g., a BERT-based emotion analysis model) and adds emotion information. The input is the translated text data, and the output is the text data with emotion information added.

[0384] Step 6:

[0385] Sending translated text data and sentiment information back to the device.

[0386] The server sends translated text data with added sentiment information back to the user's terminal using an HTTP response. The input is the text data with added sentiment information, and the output is the text data sent as an HTTP response.

[0387] Step 7:

[0388] Display of results on the user terminal

[0389] The user terminal receives translated text data with sentiment information sent back from the server and displays it to the user. The input is text data with sentiment information added, and the output is the translated result displayed to the user.

[0390] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0391] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0392] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0393] [Second Embodiment]

[0394] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0395] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0396] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0397] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0398] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0399] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0400] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0401] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0402] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0403] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0404] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0405] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0406] This invention is a system that captures audio data on a user terminal, transmits it to an edge computing server, converts the audio data into text data there, translates it into a specified language, and finally returns the translated text data to the user terminal. Embodiments of the present invention will be described in detail below with specific examples.

[0407] User terminal operation

[0408] 1. Audio Capture

[0409] When a user begins a conversation, the microphone device on the user's device captures the audio. This audio is stored on the device as digital data. Because this audio data is captured at short intervals, real-time functionality can be maintained.

[0410] 2. Sending to the edge server

[0411] The captured audio data is converted to binary format and sent to the edge server using an HTTP request. This allows the user terminal to send the audio data for immediate processing.

[0412] 3. Retrieving text data from edge servers

[0413] The edge server processes the audio data and generates text data, which the user terminal receives as an HTTP response. The user terminal then uses this text data in the next processing stage.

[0414] 4. Submitting a translation request

[0415] The system then sends a request to the edge server to translate the received text data into the target language specified by the user.

[0416] 5. Retrieving and displaying translated data

[0417] The system receives translated text data from the edge server and displays this data to the user. This allows the user to see what they said in a translated format in a different language.

[0418] Edge server operation

[0419] 1. Receiving audio data

[0420] The edge server receives audio data transmitted from the user's terminal. The edge server then prepares this data for analysis.

[0421] 2. Speech Recognition Processing

[0422] The edge server analyzes the received audio data using a speech recognition system and converts it into text data. This process employs advanced speech recognition algorithms.

[0423] 3. Return of text data

[0424] The generated text data is sent back to the user's terminal. This allows the user's terminal to begin the next translation process.

[0425] 4. Receiving a translation request

[0426] It receives translation requests for text data sent from the user's terminal. This is for converting text generated by speech recognition into a specified target language.

[0427] 5. Text translation processing

[0428] The received text data is translated into the target language. The edge server uses a high-performance translation engine to perform this transfer.

[0429] 6. Return of translated data

[0430] The translated text data is sent back to the user's terminal. This allows the user to actually check the translation results.

[0431] Specific examples

[0432] For example, a Japanese-speaking user says the phrase "Hello, how are you?" to the user terminal. This audio data is immediately captured and sent to the edge server. The edge server uses a speech recognition engine to convert this audio data into text "Hello, how are you?" and sends it back to the user terminal. Next, the user terminal resends this text data to the edge server, specifying English as the target language. The edge server translates the text data into English "Hello, how are you?" and finally sends it back to the user terminal. The user terminal displays this result, and the user can confirm the translation.

[0433] Through the process described above, the system of the present invention can translate speech into different languages ​​in real time and with ultra-low latency, and provide it to the user. This system has high practical applications in various fields such as business, tourism, and education, and can effectively overcome language barriers.

[0434] The following describes the processing flow.

[0435] Step 1:

[0436] When a user begins voice input, the user's device captures the audio data using its microphone. Specifically, it converts the user's spoken voice into a digital format in real time and temporarily stores it on the device.

[0437] Step 2:

[0438] The device converts the captured audio data into binary format and sends it to a specified endpoint on the edge computing server (e.g., / transcribe) using an HTTP request. The audio data is attached to this request.

[0439] Step 3:

[0440] The edge server receives the audio data and the speech recognition engine performs the analysis. Specifically, it initiates the process of converting the audio data into text data.

[0441] Step 4:

[0442] The edge server sends the text data generated as a result of speech recognition back to the user terminal as an HTTP response. The user terminal receives this response and extracts the text data.

[0443] Step 5:

[0444] The user's terminal uses the received text data to resend the translation request to the edge computing server. This request includes information about the original text and the target language.

[0445] Step 6:

[0446] The edge server processes the received translation request and translates the text data into the specified target language. Specifically, it uses a translation engine on the server to translate the text.

[0447] Step 7:

[0448] The edge server sends the translated text data back to the user's terminal. The user terminal receives this response and extracts the translated text data.

[0449] Step 8:

[0450] The user's device displays the final translation result to the user. The user can check the translated text displayed on the screen.

[0451] (Example 1)

[0452] Next, we will describe Example 1. 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."

[0453] Conventional translation systems require real-time performance and accuracy in the process of capturing audio data, converting it to text, and then translating it. Especially in today's society where multilingual communication is essential, systems that perform these processes quickly and efficiently are indispensable. However, delays often occur between capturing audio data and obtaining translated text data, which is stressful for users. Furthermore, challenges exist regarding data transmission formats and security, making it difficult to balance system reliability and speed. Therefore, there is a need for the development of a faster and more accurate real-time speech translation system.

[0454] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0455] In this invention, the server includes means for capturing audio data, means for transmitting the audio data to an edge computing server, means for the edge computing server to convert the audio data into text data, means for translating the text data into a specified language, means for transmitting the translated text data to a user terminal, means for capturing and transmitting audio data at short intervals to maintain real-time performance, means for sending and receiving data between the edge computing server and the user terminal using HTTP requests and responses, and means for displaying the text data on the screen of the user terminal. This enables users to translate audio between multiple languages ​​in real time and accurately, allowing for rapid communication.

[0456] "Audio data" refers to data that records the voice spoken by a user in digital format.

[0457] "Means of capturing" refers to a device or method for acquiring a user's voice as digital data through a microphone device.

[0458] An "edge computing server" is a distributed computing resource located close to the user's terminal for processing voice data.

[0459] "Means of transmission" refers to methods or devices for transferring data from a user terminal to an edge computing server via communication.

[0460] "Means of converting to text data" refers to processes or systems that convert audio data into string data.

[0461] "Specified language" refers to the target language that the user wishes to have translated into.

[0462] "Translation means" refers to the process or system of converting generated text data into a specified target language.

[0463] "Means of sending and receiving" refers to communication methods for transmitting and receiving data.

[0464] "Maintaining real-time performance" means minimizing the delay between capturing audio data and displaying translated text data.

[0465] "Short interval" means processing or transmitting data in the shortest possible time units, such as every 100 milliseconds.

[0466] An "HTTP request" is a request from a client to a server to send data, and it is a protocol primarily used for web communication.

[0467] An "HTTP response" is a data response that a server provides to a client, and it refers to the result of an HTTP request.

[0468] "Means of display" refers to methods or devices for displaying text data on the screen of a user's terminal.

[0469] This invention is a system that captures audio data on a user terminal, sends it to an edge computing server, converts the audio data into text data there, translates it into a specified language, and finally sends the translated text data back to the user terminal.

[0470] User terminal operation

[0471] 1. Audio Capture

[0472] When a user begins a conversation, the microphone device on the user's device captures the audio. This audio is stored on the device as digital data. Because this audio data is captured at short intervals, real-time functionality can be maintained. For example, a smartphone's microphone picks up sound and converts it into digital data through an analog-to-digital converter.

[0473] 2. Sending to the edge server

[0474] The captured audio data is converted to binary format and sent to the edge server using an HTTP request. This allows the user terminal to send the audio data for immediate processing. Alternatively, the terminal's application converts the audio data to binary format and sends it to the edge server using an HTTP library (e.g., OkHttp).

[0475] 3. Retrieving text data from edge servers

[0476] The edge server processes the audio data and generates text data, which the user terminal receives as an HTTP response. The user terminal uses this text data in the next processing stage. To receive the HTTP response, the terminal uses asynchronous communication to wait for the response from the server and saves the received text data to its internal storage or memory.

[0477] 4. Submitting a translation request

[0478] The application sends a request back to the edge server to translate the received text data into the target language specified by the user. The application creates an HTTP POST request, including the text and target language as parameters.

[0479] 5. Retrieving and displaying translated data

[0480] The system receives translated text data from the edge server and displays this data to the user. This allows the user to see what they said in a translated form in a different language. It receives an HTTP response and displays the translated text in a UI component (e.g., TextView) on the user's device.

[0481] Edge server operation

[0482] 1. Receiving audio data

[0483] The edge server receives audio data transmitted from the user's terminal. The edge server prepares this data for analysis. It receives an HTTP request, decodes the audio data, and temporarily stores it in storage.

[0484] 2. Speech Recognition Processing

[0485] The edge server analyzes the received audio data using a speech recognition system and converts it into text data. This process employs advanced speech recognition algorithms. For example, the Google Cloud Speech-to-Text API can be used to convert audio data into text data. The audio data is input into the speech recognition system, and text data is generated. The generated text is temporarily stored in memory.

[0486] 3. Return of text data

[0487] The generated text data is sent back to the user's terminal. This allows the user's terminal to begin the next translation process. The text data is serialized into JSON format and sent as an HTTP response.

[0488] 4. Receiving a translation request

[0489] It receives translation requests for text data sent from the user's terminal. This is for converting text generated by speech recognition into a specified target language. It receives HTTP requests and parses the request parameters.

[0490] 5. Text translation processing

[0491] The received text data is translated into the target language. The edge server uses a high-performance translation engine to perform this transfer. For example, the Google Cloud Translation API is used to translate the text data. A request is sent to the translation API and the result is received.

[0492] 6. Return of translated data

[0493] The translated text data is sent back to the user's terminal. This allows the user to actually check the translation result. The translation result is sent as an HTTP response, and whether the communication was successful is logged.

[0494] Specific example

[0495] For example, a Japanese-speaking user says the phrase "Hello, how are you?" to the user terminal. This audio data is immediately captured and sent to the edge server. The edge server uses a speech recognition engine to convert this audio data into text "Hello, how are you?" and sends it back to the user terminal. Next, the user terminal resends this text data to the edge server, specifying English as the target language. The edge server translates the text data into English "Hello, how are you?" and finally sends it back to the user terminal. The user terminal displays this result, and the user can confirm the translation.

[0496] The following are specific examples of prompt statements to input to a generative AI model.

[0497] Example: "Convert the following Japanese audio to text, and then translate it into English. The audio says 'Hello, how are you?'"

[0498] This system can translate users' voices in real time and accurately, converting them into different languages. This makes it highly practical in diverse fields such as business, tourism, and education, effectively overcoming language barriers.

[0499] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0500] Step 1:

[0501] When a user begins a conversation, the microphone device on the user's device captures the audio. The input is an audio signal, and the output is digital data. Specifically, the smartphone's microphone picks up the audio and converts it into digital data through an ADC (analog-to-digital converter). This data is temporarily stored on the device.

[0502] Step 2:

[0503] The user terminal converts the captured audio data into binary format and sends it to the edge server using an HTTP POST request. The input is digital audio data, and the output is binary data. Specifically, the terminal's application converts the audio data into binary format and sends it to the edge server using a network library (e.g., OkHttp).

[0504] Step 3:

[0505] The edge server receives audio data sent from the user's terminal. The input is binary data sent in an HTTP request, and the output is audio data ready for analysis. Specifically, the server receives the HTTP request, decodes the binary data, and temporarily stores it in storage.

[0506] Step 4:

[0507] The edge server inputs the received audio data into a speech recognition engine and converts it into text data. The input is digital audio data, and the output is text data. Specifically, the edge server uses a speech recognition API (e.g., Google Cloud Speech-to-Text API) to analyze the audio and generate text data.

[0508] Step 5:

[0509] The edge server sends the generated text data to the user's terminal as an HTTP response. The input is text data, and the output is an HTTP response. Specifically, the server serializes the text data into JSON format and sends it to the user's terminal as an HTTP response.

[0510] Step 6:

[0511] The user terminal receives text data from the edge server and uses this data for the next processing step. The input is the text data sent in the HTTP response, and the output is the text data used in the request. Specifically, it receives the HTTP response and saves the text data to memory or internal storage.

[0512] Step 7:

[0513] The user terminal sends a request to the edge server to translate the received text data into the target language. The input is the text data and the specified target language, and the output is the translation request. Specifically, the application creates an HTTP POST request, including the text and target language as parameters.

[0514] Step 8:

[0515] The edge server receives translation requests from user terminals and inputs them into the translation engine. The input consists of text and the target language, and the output is translated text data. Specifically, the server calls a translation API (e.g., Google Cloud Translation API) to translate the text into the specified language.

[0516] Step 9:

[0517] The translated text data is sent back from the edge server to the user terminal. The input is the translated text data, and the output is the translated data as an HTTP response. Specifically, the server serializes the translated text into JSON format and sends it to the user terminal as an HTTP response.

[0518] Step 10:

[0519] The user terminal receives translated text data from the edge server and displays it to the user. The input is the translated data sent via HTTP response, and the output is the text data displayed on the user screen. Specifically, it receives the HTTP response and displays the translated text in a UI component (e.g., TextView). This allows the user to see the translated content in real time.

[0520] (Application Example 1)

[0521] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0522] In multilingual food delivery services, a challenge exists: users often find it difficult to place orders smoothly in their native language. Therefore, it is necessary to provide a system that allows users who speak different languages ​​to comfortably use the service without experiencing language barriers. Furthermore, utilizing voice input is required to provide a more intuitive and convenient ordering experience.

[0523] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0524] In this invention, the server includes means for capturing voice data, means for transmitting the voice data to an edge computing server, means for the edge computing server to convert the voice data into text data, means for translating the text data into a specified language, means for transmitting the translated text data to a user terminal and for the user terminal to display the translated text data, means for the user terminal to place an order based on voice input, and means for the edge computing server to process the order details and notify the delivery service. This enables users to order food delivery using voice input in different languages, providing a comfortable ordering experience that transcends language barriers.

[0525] "Audio data" refers to data obtained by converting audio signals input by a user through a microphone device into a digital format.

[0526] "Capture method" refers to a function that captures and records audio data using the microphone device on the user's terminal.

[0527] An "edge computing server" is a server that receives voice and text data sent from user terminals and processes it in real time.

[0528] "Text data" refers to data in string format generated from audio data through speech recognition.

[0529] A "means of translation" is a function for converting text data written in one language into another specified language.

[0530] A "user terminal" refers to a device used by a user, such as a smartphone or tablet, and is a device for capturing audio and sending / receiving data.

[0531] "Means of placing an order" refers to a function that allows users to confirm their food delivery order via voice input or other means and send the details to the system.

[0532] "Means of notifying delivery services" refers to a function that transmits order details processed on edge computing servers to food delivery personnel and operations systems.

[0533] "Means of display" refers to functions that visually present translated text data and order information to the user on the user's terminal.

[0534] This invention is a system that captures voice data and translates it into different languages ​​in real time via an edge computing server. It is particularly applicable to food delivery services, enabling a multilingual ordering process.

[0535] The user terminal first has the function to capture audio data. This includes devices such as smartphones and tablets, which use microphone devices to capture audio as digital data. The captured audio data is converted to binary format and sent to the edge computing server via an HTTP request.

[0536] The edge computing server converts received audio data into text data through speech recognition processing. This speech recognition uses speech recognition libraries such as Google Cloud Speech-to-Text or Amazon Transcribe. The converted text data is then translated into the language specified by the user. High-performance translation engines such as the Google Cloud Translation API or Microsoft Translator Text API are used for the translation process.

[0537] The translated text data is sent back to the user's terminal and displayed on the user's screen. This allows the user to see the results of their voice input being translated into different languages. Furthermore, users can use voice input to place food delivery orders, which are processed on an edge computing server and notified to the delivery service.

[0538] For example, when a user says "I want to order curry" into their smartphone, this voice data is captured and sent to an edge server. The edge server converts the voice into text "I want to order curry," and then translates it into English as "I want to order curry." This translation is then displayed on the user's smartphone. The order details are also processed by the edge server and notified to the food delivery service.

[0539] Examples of prompt statements are as follows:

[0540] User: "I want to order curry."

[0541] Method:

[0542] 1. Audio signal capture.

[0543] 2. Sent audio signal to edge server.

[0544] 3. Received text: "I want to order curry."

[0545] 4. Sent text for translation.

[0546] 5. Received translated text: "I want to order curry"

[0547] 6. Displaying translated text: "I want to order curry"

[0548] This allows users to order food delivery through voice input in different languages, without being hindered by language barriers. This system can also be used for a variety of purposes, including business, tourism, and education.

[0549] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0550] Step 1:

[0551] The user captures audio data using the microphone device on their smartphone or tablet. As input, the user's voice signal is converted into a digital format. As output, digital audio data is generated. This data is captured at short time intervals and immediately prepared to proceed to the next processing step.

[0552] Step 2:

[0553] The terminal converts the captured audio data into binary format and sends it to the edge computing server as an HTTP request. Digital audio data exists as input. The output is the audio data converted to binary format and configured as the payload for the HTTP request. A POST request is sent to the URL of the edge computing server.

[0554] Step 3:

[0555] The server receives an HTTP request and retrieves audio data. The input is audio data in binary format, delivered via the HTTP request. The output is decoded audio data generated on the server side. This is then passed to the speech recognition process.

[0556] Step 4:

[0557] The server performs speech recognition processing and converts the audio data into text data. The input is the decoded audio data. The output is the text data generated by speech recognition. This process utilizes speech recognition libraries such as Google Cloud Speech-to-Text and Amazon Transcribe.

[0558] Step 5:

[0559] The server sends the generated text data back to the user's terminal. Text data exists as input. Text data is returned as an HTTP response as output. The user's terminal receives this data and proceeds to the next translation processing step.

[0560] Step 6:

[0561] The terminal sends a request back to the edge computing server to translate the received text data into the specified language. The input includes the text data and information about the target language. The output is an HTTP request sent back to the edge computing server.

[0562] Step 7:

[0563] The server receives a translation request and translates the text data into the target language. The input consists of the text data to be translated and information about the target language. The output is translated text data generated using a high-performance translation engine (e.g., Google Cloud Translation API or Microsoft Translator Text API).

[0564] Step 8:

[0565] The server returns the translated text data to the user's terminal. The input is the translated text data. The output is the translated text data returned as an HTTP response. The user's terminal receives this data and displays it visually to the user.

[0566] Step 9:

[0567] The device displays the translated text data on the screen for the user to review. The input is the translated text data. The output is the translation result displayed on the device screen. This allows the user to immediately see how their voice input has been translated.

[0568] Step 10:

[0569] When a user places an order using voice input, the terminal sends the order details to an edge computing server. The input is text data containing the user's order. The output is an HTTP request sent to the edge computing server. This allows the server to process the order information and notify the appropriate food delivery service.

[0570] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0571] The present invention is a system that captures voice data on a user terminal, transmits it to an edge computing server, converts the voice data into text data there, translates it into a specified language, performs emotion recognition using an emotion engine, and finally adds emotion information to the translated text data before returning it to the user terminal. Embodiments of the present invention will be described in detail below with specific examples.

[0572] User terminal operation

[0573] 1. Audio Capture

[0574] When a user begins a conversation, the microphone device on the user's device captures the audio. This audio is stored on the device as digital data. Because this audio data is captured at short intervals, real-time functionality can be maintained.

[0575] 2. Sending to the edge server

[0576] The captured audio data is converted to binary format and sent to a specified endpoint on the edge server (e.g., / transcribe) using an HTTP request. This allows the user terminal to send the audio data for immediate processing.

[0577] 3. Retrieving text data from edge servers

[0578] The edge server processes the audio data and generates text data, which the user terminal receives as an HTTP response. The user terminal then uses this text data in the next processing stage.

[0579] 4. Submitting a translation request

[0580] The system then sends a request to the edge server to translate the received text data into the target language specified by the user.

[0581] 5. Retrieving and displaying translated data

[0582] The system receives translated text data from the edge server and displays this data to the user. This allows the user to see what they said in a translated format in a different language.

[0583] Edge server operation

[0584] 1. Receiving audio data

[0585] The edge server receives audio data transmitted from the user's terminal. The edge server then prepares this data for analysis.

[0586] 2. Speech Recognition Processing

[0587] The edge server analyzes the received audio data using a speech recognition system and converts it into text data. This process employs advanced speech recognition algorithms.

[0588] 3. Return of text data

[0589] The generated text data is sent back to the user's terminal. This allows the user's terminal to begin the next translation process.

[0590] 4. Receiving a text translation request

[0591] It receives translation requests for text data sent from the user's terminal. This is for converting text generated by speech recognition into a specified target language.

[0592] 5. Text translation processing

[0593] The received text data is translated into the target language. The edge server uses a high-performance translation engine to perform this transfer.

[0594] 6. Performing emotion recognition

[0595] After translation processing, an emotion engine is used to analyze the emotions contained in the text data. Based on the emotion information obtained from the audio data, emotion tags are added to the text data.

[0596] 7. Return of translated data and sentiment information

[0597] The translated text data and added sentiment information are sent back to the user's terminal. This allows the user's terminal to display the translation result with the sentiment information included.

[0598] Specific examples

[0599] For example, a Japanese-speaking user says the phrase "Hello, how are you?" to the user terminal. This audio data is immediately captured and sent to the edge server. The edge server uses a speech recognition engine to convert this audio data into text "Hello, how are you?" and sends it back to the user terminal. Next, the user terminal resends this text data to the edge server, specifying English as the target language. The edge server translates the text data into English "Hello, how are you?" and uses an emotion engine to recognize the user's emotion (e.g., positive emotion) from the original Japanese audio data. This emotion information is added to the translation result and finally sent back to the user terminal. As a result, the user terminal displays the translated text with emotion information, such as "Hello, how are you? [Positive]".

[0600] Through the process described above, the system of the present invention can translate speech into different languages ​​in real time and with ultra-low latency, and can also add emotional information. This system has high practical applications in various fields such as business, tourism, and education, and helps to effectively overcome language barriers.

[0601] The following describes the processing flow.

[0602] Step 1:

[0603] When a user starts a conversation, their device captures the audio. Specifically, it uses a microphone to pick up the user's voice and converts it to a digital format in real time.

[0604] Step 2:

[0605] The device captures audio data, converts it to binary format, and sends it to the edge computing server using an HTTP request. This request includes the audio data.

[0606] Step 3:

[0607] The edge server receives the audio data and the speech recognition engine begins analysis. Specifically, the process of converting the audio data into text data is executed.

[0608] Step 4:

[0609] The edge server sends the text data generated as a result of speech recognition back to the user terminal as an HTTP response. The user terminal receives this response and extracts the text data.

[0610] Step 5:

[0611] The user's terminal uses the received text data to resend the translation request to the edge computing server. This request includes information about the original text and the target language.

[0612] Step 6:

[0613] The edge server receives the text translation request and uses a high-performance translation engine to translate the text data into the specified target language.

[0614] Step 7:

[0615] The edge server performs sentiment recognition on the translated text data. Specifically, the sentiment engine analyzes the audio data and text to evaluate relevant sentiment information.

[0616] Step 8:

[0617] The edge server adds sentiment information to the translation result and sends it to the user terminal as an HTTP response. The user terminal receives this response and extracts the translated text data and sentiment information.

[0618] Step 9:

[0619] The user's device displays the final translation result and sentiment information to the user. Specifically, it visually displays the translated text and its sentiment on the screen, making it easy for the user to understand.

[0620] (Example 2)

[0621] Next, we will describe Example 2. 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".

[0622] Conventional speech recognition systems could convert speech data to text and translate languages, but they could not add emotional information, making it difficult to achieve communication that reflected the user's emotions. Furthermore, they often lacked processing speed and real-time capabilities, limiting their practicality. As a result, they were unable to effectively overcome language barriers in fields such as business, tourism, and education.

[0623] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0624] In this invention, the server includes means for capturing audio data, means for transmitting the audio data to an edge computing server, means for the edge computing server to convert the audio data into text data, means for translating the text data into a specified language, means for adding sentiment information to the translated text data, and means for transmitting the translated text data and sentiment information to a user terminal. This enables the addition of sentiment information in addition to the conversion and translation of audio data into text. As a result, users can use text with sentiment information that has been translated in real time, enabling richer communication. Furthermore, high-speed edge computing improves processing speed and real-time capabilities, increasing its practicality in various fields.

[0625] "Audio data" refers to a digital audio signal acquired using a microphone device.

[0626] An "edge computing server" refers to a server located close to the user's terminal that performs data processing with an emphasis on real-time performance.

[0627] "Text data" refers to string data generated by analyzing audio data.

[0628] "Translation" refers to the process of converting text data expressed in one language into another language.

[0629] "Emotional information" refers to information that indicates a user's emotions (e.g., positive, negative, neutral) as recognized from text data or audio data.

[0630] An "HTTP request" refers to a request made by a client to a server using a protocol for sending data.

[0631] This invention relates to a system that captures audio data, converts it into text data in real time, translates it into a specified language, and adds emotional information. This system operates based on a user terminal and an edge computing server.

[0632] User terminal operation

[0633] First, the user terminal is equipped with a microphone device that captures the user's voice. For example, when the user says "Hello, how are you?", this voice is captured by the terminal as digital audio data. Next, the user terminal converts the captured audio data into binary format and sends it to a specified endpoint on the edge computing server using an HTTP request. An example of such an endpoint is " / transcribe".

[0634] Edge computing server-side operation

[0635] The edge computing server receives audio data sent from the user's terminal. This received audio data is converted into text data using a speech recognition system (the API used as an example provides a speech recognition algorithm). Examples of speech recognition systems that can be used include the Google Speech-to-Text API. The converted text data is sent back to the user's terminal via an HTTP response.

[0636] Next, the user terminal sends the acquired text data back to the edge computing server, requesting a translation into the specified language. This translation process uses a high-performance translation engine, such as the Google Translate API. Once the translation is complete, the server analyzes the translated text data with a sentiment engine (for example, IBM Watson Tone Analyzer) and adds sentiment information. This translated data with added sentiment information is then sent back to the user terminal as an HTTP response.

[0637] Display of data by the user terminal

[0638] Finally, the user terminal displays the received translation data and sentiment information to the user. This allows the user to see their spoken content translated into different languages ​​in real time, with added sentiment information.

[0639] Specific example

[0640] For example, a Japanese-speaking user says the phrase "Hello, how are you?" to the user terminal. This audio data is immediately captured and sent to the edge computing server. The edge computing server uses a speech recognition engine to convert the audio data into text "Hello, how are you?" and sends it back to the user terminal. The user terminal resends this text data to the edge computing server, specifying English as the target language. The edge server translates the text data into English "Hello, how are you?" and uses an emotion engine to recognize the user's emotion (e.g., positive emotion) from the original Japanese audio data. This emotion information is added to the translation result and finally sent back to the user terminal. As a result, the user terminal displays the translated text with emotion information, such as "Hello, how are you? [Positive]".

[0641] Through the process described above, the system of the present invention can translate speech into different languages ​​in real time and with ultra-low latency, and can also add emotional information. This system has high practical applications in various fields such as business, tourism, and education, and helps to effectively overcome language barriers.

[0642] Example of a prompt

[0643] For example, by inputting the following prompt into the generating AI model, you can obtain detailed instructions for performing the aforementioned process:

[0644] Prompt message:

[0645] "Capture Japanese audio data, translate it into English, and display the translation result with added sentiment information."

[0646] This prompt allows the generative AI model to suggest specific technical measures, such as how to use the appropriate API and how to specify the endpoint.

[0647] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0648] Step 1:

[0649] The user starts voice capture.

[0650] Input: User's voice (e.g., "Hello, how are you?")

[0651] Specific action: The user begins speaking into the device's microphone. The microphone captures this audio as a digital signal.

[0652] Output: Digital audio data

[0653] Step 2:

[0654] The device acquires voice data and sends it to the edge server.

[0655] Input: Captured digital audio data

[0656] Specific operation: The terminal converts the captured audio data into binary format and saves it to a temporary file. Then, it sends it to the specified endpoint on the edge server (e.g., " / transcribe") using an HTTP request.

[0657] Output: Audio data sent to the edge server

[0658] Step 3:

[0659] The edge server converts the audio data into text data.

[0660] Input: Binary format audio data sent from the terminal.

[0661] Specific operation: The edge server receives the audio data, analyzes it using a speech recognition system (e.g., an API that provides a speech recognition algorithm), and converts it into text data.

[0662] Output: Converted text data (e.g., "Hello, how are you?")

[0663] Step 4:

[0664] The edge server sends text data back to the terminal.

[0665] Input: Text data generated by speech recognition

[0666] Specific operation: The edge server sends the generated text data back to the user terminal as an HTTP response. The terminal saves this text data for use in the next process.

[0667] Output: Text data sent to the user's terminal

[0668] Step 5:

[0669] The device sends text data to the edge server and requests a translation.

[0670] Input: Text data received from the edge server

[0671] Specific action: The user terminal sends a request to the edge server again to translate the saved text data into the target language. English is specified as the target language.

[0672] Output: Text data with translation request sent to the edge server

[0673] Step 6:

[0674] The edge server translates the text data.

[0675] Input: Text data with translation request sent from the user terminal.

[0676] Specific operation: The edge server analyzes the received text data using a translation engine (e.g., an API that provides translation algorithms) and translates it into the specified target language (in this case, English).

[0677] Output: Translated text data (e.g., "Hello, how are you?")

[0678] Step 7:

[0679] The edge server analyzes the sentiment of the translated data and adds emotional information.

[0680] Input: Translated text data

[0681] Specific operation: The edge server analyzes the translated text data using an emotion engine (e.g., an API that provides an emotion analysis algorithm) and recognizes the emotions contained in the text (e.g., positive, negative, neutral). This emotion information is then added to the translated data as a tag.

[0682] Output: Translated text data with added sentiment information (e.g., "Hello, how are you? [Positive]")

[0683] Step 8:

[0684] The edge server sends the translation data and sentiment information back to the terminal.

[0685] Input: Translated text data with added emotional information

[0686] Specific operation: The edge server sends translated data with sentiment information back to the user terminal as an HTTP response. The user terminal receives this data.

[0687] Output: Translated data with sentiment information sent to the user's terminal.

[0688] Step 9:

[0689] The device displays translation data and sentiment information to the user.

[0690] Input: Translation data with emotional information

[0691] Specific operation: The user terminal displays the translated data with sentiment information received on the user interface. The user can then view the translated text and its sentiment information on the screen.

[0692] Output: Translated text with sentiment information displayed to the user (e.g., "Hello, how are you? [Positive]")

[0693] This process allows users to translate speech into different languages ​​in real time and then use it in a format that includes emotional information.

[0694] (Application Example 2)

[0695] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0696] In modern society, there is an increasing demand for multilingual support and real-time emotion recognition. However, conventional technologies are unable to adequately translate voice data in real time or add emotional information, leading to decreased communication efficiency. Therefore, there is a need for a system that integrates multilingual support and emotion recognition. In particular, in security services, smooth communication and situational awareness at the site are crucial, and technologies that can solve this problem are in demand.

[0697] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0698] In this invention, the server includes means for capturing voice data, means for transmitting the voice data to an edge computing server, means for the edge computing server to convert the voice data into text data, means for translating the text data into a specified language, means for adding sentiment information to the translated text data, and means for transmitting the translated text data and sentiment information to a user terminal. This enables multilingual real-time translation and sentiment recognition.

[0699] "Audio data" refers to data that represents audio in a digital format.

[0700] "Means of capturing" refers to devices or methods that record a user's voice and acquire it as digital data.

[0701] An "edge computing server" is a distributed server that processes data sent from a user's terminal in real time at a location close to the local area.

[0702] "Text data" refers to data that represents audio data in written form.

[0703] "Translation methods" refer to the techniques and methods used to convert text data expressed in the original language into another specified language.

[0704] "Means of adding emotional information" refers to technologies and methods that analyze the emotions contained in text data and add the results to the data as annotations.

[0705] A "user terminal" is a device (e.g., smart glasses, smartphone, etc.) used to capture audio data and communicate with a server.

[0706] This invention relates to a system that translates voice data into multiple languages ​​in real time and adds emotional information to that data. This system aims to improve effective communication and situational awareness, particularly in security services. The following describes embodiments for carrying out this invention using specific examples.

[0707] System Configuration

[0708] This system consists of user terminals (such as smart glasses or smartphones) that capture voice data and edge computing servers that process the data.

[0709] User terminal operation

[0710] 1. Audio Capture

[0711] When a user begins a conversation, the microphone device on the user's device captures the audio. The captured audio data is collected at short intervals, and the digital data is stored on the device.

[0712] 2. Sending to the edge computing server

[0713] The captured audio data is converted to binary format and sent to the edge computing server using an HTTP request.

[0714] Edge computing server operation

[0715] 1. Receiving and transcribing audio data

[0716] The edge computing server analyzes the audio data received from the user's terminal and converts it into text data using speech recognition software (e.g., Google Speech-to-Text API).

[0717] 2. Translation of text data

[0718] The converted text data is translated into the specified language. A high-performance translation engine (e.g., Google Translate API) is used for the translation.

[0719] 3. Adding emotional information

[0720] The translated text data is subjected to an emotion recognition engine, and emotional information (e.g., positive, negative, neutral) is added. An AI model (e.g., a BERT-based emotion analysis model) is used for emotion recognition.

[0721] Return to user terminal

[0722] The translated text data and sentiment information processed on the edge computing server are sent back to the user terminal via an HTTP response, and the user terminal displays it.

[0723] Specific example

[0724] A security guard is seen conversing with a foreign tourist.

[0725] 1. A tourist says in English, "Excuse me, can you help me find the museum?"

[0726] 2. The security guard's smart glasses capture this audio and send the data to an edge computing server.

[0727] 3. The server converts the speech to text and obtains "Excuse me, can you help me find the museum?".

[0728] 4. The text data is translated into Japanese, and the text "Excuse me, could you help me find the museum?" is generated.

[0729] 5. The translated text is tagged with sentiment and sent back to the user's device as "Excuse me, could you help me find the museum? [Positive]".

[0730] 6. The information is ultimately displayed on the security guard's smart glasses, allowing the guard to take appropriate action.

[0731] Example of a prompt

[0732] User's voice: "Excuse me, can you help me find the museum?"

[0733] Audio data capture:

[0734] Sending voice data to the edge server

[0735] Speech recognition and text conversion: "Excuse me, can you help me find the museum?"

[0736] Text translation: "Excuse me, could you help me find the museum?"

[0737] Emotional perception: "Positive"

[0738] Final message: "Excuse me, could you help me find the museum? [Positive]"

[0739] This enables the system to integrate multilingual support and emotion recognition, providing real-time feedback to users. Furthermore, it can support smooth communication and rapid decision-making in security services.

[0740] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0741] Step 1:

[0742] Audio Capture

[0743] The user's device (e.g., smart glasses, smartphone) captures the user's voice. When the user starts speaking, the device's microphone collects the voice and stores it as digital data. The input is voice data, and the output is digital voice data.

[0744] Step 2:

[0745] Sending digital audio data to an edge computing server

[0746] The terminal converts the captured digital audio data into binary format and sends it to the edge computing server using an HTTP request. The input is digital audio data, and the output is binary audio data sent as an HTTP request.

[0747] Step 3:

[0748] Receiving and transcribing audio data by the server.

[0749] The server receives binary audio data sent from the terminal. The received audio data is converted into text data using speech recognition software (e.g., Google Speech-to-Text API). The input is binary audio data, and the output is text data.

[0750] Step 4:

[0751] Text data translation

[0752] The server translates the converted text data into the specified language. A high-performance translation engine (e.g., Google Translate API) is used for the translation. The input is text data, and the output is translated text data in the target language.

[0753] Step 5:

[0754] Adding emotional information

[0755] The server analyzes the translated text data using an emotion recognition engine (e.g., a BERT-based emotion analysis model) and adds emotion information. The input is the translated text data, and the output is the text data with emotion information added.

[0756] Step 6:

[0757] Sending translated text data and sentiment information back to the device

[0758] The server sends translated text data with added sentiment information back to the user's terminal using an HTTP response. The input is the text data with added sentiment information, and the output is the text data sent as an HTTP response.

[0759] Step 7:

[0760] Display of results on the user terminal

[0761] The user terminal receives translated text data with sentiment information sent back from the server and displays it to the user. The input is text data with sentiment information added, and the output is the translated result displayed to the user.

[0762] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0763] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0764] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0765] [Third Embodiment]

[0766] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0767] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0768] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0769] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0770] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0771] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0772] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0773] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0774] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0775] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0776] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0777] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0778] This invention is a system that captures audio data on a user terminal, transmits it to an edge computing server, converts the audio data into text data there, translates it into a specified language, and finally returns the translated text data to the user terminal. Embodiments of the present invention will be described in detail below with specific examples.

[0779] User terminal operation

[0780] 1. Audio Capture

[0781] When a user begins a conversation, the microphone device on the user's device captures the audio. This audio is stored on the device as digital data. Because this audio data is captured at short intervals, real-time functionality can be maintained.

[0782] 2. Sending to the edge server

[0783] The captured audio data is converted to binary format and sent to the edge server using an HTTP request. This allows the user terminal to send the audio data for immediate processing.

[0784] 3. Retrieving text data from edge servers

[0785] The edge server processes the audio data and generates text data, which the user terminal receives as an HTTP response. The user terminal then uses this text data in the next processing stage.

[0786] 4. Submitting a translation request

[0787] The system then sends a request to the edge server to translate the received text data into the target language specified by the user.

[0788] 5. Retrieving and displaying translated data

[0789] The system receives translated text data from the edge server and displays this data to the user. This allows the user to see what they said in a translated format in a different language.

[0790] Edge server operation

[0791] 1. Receiving audio data

[0792] The edge server receives audio data transmitted from the user's terminal. The edge server then prepares this data for analysis.

[0793] 2. Speech Recognition Processing

[0794] The edge server analyzes the received audio data using a speech recognition system and converts it into text data. This process employs advanced speech recognition algorithms.

[0795] 3. Return of text data

[0796] The generated text data is sent back to the user's terminal. This allows the user's terminal to begin the next translation process.

[0797] 4. Receiving a translation request

[0798] It receives translation requests for text data sent from the user's terminal. This is for converting text generated by speech recognition into a specified target language.

[0799] 5. Text translation processing

[0800] The received text data is translated into the target language. The edge server uses a high-performance translation engine to perform this transfer.

[0801] 6. Return of translated data

[0802] The translated text data is sent back to the user's terminal. This allows the user to actually check the translation results.

[0803] Specific examples

[0804] For example, a Japanese-speaking user says the phrase "Hello, how are you?" to the user terminal. This audio data is immediately captured and sent to the edge server. The edge server uses a speech recognition engine to convert this audio data into text "Hello, how are you?" and sends it back to the user terminal. Next, the user terminal resends this text data to the edge server, specifying English as the target language. The edge server translates the text data into English "Hello, how are you?" and finally sends it back to the user terminal. The user terminal displays this result, and the user can confirm the translation.

[0805] Through the process described above, the system of the present invention can translate speech into different languages ​​in real time and with ultra-low latency, and provide it to the user. This system has high practical applications in various fields such as business, tourism, and education, and can effectively overcome language barriers.

[0806] The following describes the processing flow.

[0807] Step 1:

[0808] When a user begins voice input, the user's device captures the audio data using its microphone. Specifically, it converts the user's spoken voice into a digital format in real time and temporarily stores it on the device.

[0809] Step 2:

[0810] The device converts the captured audio data into binary format and sends it to a specified endpoint on the edge computing server (e.g., / transcribe) using an HTTP request. The audio data is attached to this request.

[0811] Step 3:

[0812] The edge server receives the audio data and the speech recognition engine performs the analysis. Specifically, it initiates the process of converting the audio data into text data.

[0813] Step 4:

[0814] The edge server sends the text data generated as a result of speech recognition back to the user terminal as an HTTP response. The user terminal receives this response and extracts the text data.

[0815] Step 5:

[0816] The user's terminal uses the received text data to resend the translation request to the edge computing server. This request includes information about the original text and the target language.

[0817] Step 6:

[0818] The edge server processes the received translation request and translates the text data into the specified target language. Specifically, it uses a translation engine on the server to translate the text.

[0819] Step 7:

[0820] The edge server sends the translated text data back to the user's terminal. The user terminal receives this response and extracts the translated text data.

[0821] Step 8:

[0822] The user's device displays the final translation result to the user. The user can check the translated text displayed on the screen.

[0823] (Example 1)

[0824] Next, we will describe Example 1. 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."

[0825] Conventional translation systems require real-time performance and accuracy in the process of capturing audio data, converting it to text, and then translating it. Especially in today's society where multilingual communication is essential, systems that perform these processes quickly and efficiently are indispensable. However, delays often occur between capturing audio data and obtaining translated text data, which is stressful for users. Furthermore, challenges exist regarding data transmission formats and security, making it difficult to balance system reliability and speed. Therefore, there is a need for the development of a faster and more accurate real-time speech translation system.

[0826] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0827] In this invention, the server includes means for capturing audio data, means for transmitting the audio data to an edge computing server, means for the edge computing server to convert the audio data into text data, means for translating the text data into a specified language, means for transmitting the translated text data to a user terminal, means for capturing and transmitting audio data at short intervals to maintain real-time performance, means for sending and receiving data between the edge computing server and the user terminal using HTTP requests and responses, and means for displaying the text data on the screen of the user terminal. This enables users to translate audio between multiple languages ​​in real time and accurately, allowing for rapid communication.

[0828] "Audio data" refers to data that records the voice spoken by a user in digital format.

[0829] "Means of capturing" refers to a device or method for acquiring a user's voice as digital data through a microphone device.

[0830] An "edge computing server" is a distributed computing resource located close to the user's terminal for processing voice data.

[0831] "Means of transmission" refers to methods or devices for transferring data from a user terminal to an edge computing server via communication.

[0832] "Means of converting to text data" refers to processes or systems that convert audio data into string data.

[0833] "Specified language" refers to the target language that the user wishes to have translated into.

[0834] "Translation means" refers to the process or system of converting generated text data into a specified target language.

[0835] "Means of sending and receiving" refers to communication methods for transmitting and receiving data.

[0836] "Maintaining real-time performance" means minimizing the delay between capturing audio data and displaying translated text data.

[0837] "Short interval" means processing or transmitting data in the shortest possible time units, such as every 100 milliseconds.

[0838] An "HTTP request" is a request from a client to a server to send data, and it is a protocol primarily used for web communication.

[0839] An "HTTP response" is a data response that a server provides to a client, and it refers to the result of an HTTP request.

[0840] "Means of display" refers to methods or devices for displaying text data on the screen of a user's terminal.

[0841] This invention is a system that captures audio data on a user terminal, sends it to an edge computing server, converts the audio data into text data there, translates it into a specified language, and finally sends the translated text data back to the user terminal.

[0842] User terminal operation

[0843] 1. Audio Capture

[0844] When a user begins a conversation, the microphone device on the user's device captures the audio. This audio is stored on the device as digital data. Because this audio data is captured at short intervals, real-time functionality can be maintained. For example, a smartphone's microphone picks up sound and converts it into digital data through an analog-to-digital converter.

[0845] 2. Sending to the edge server

[0846] The captured audio data is converted to binary format and sent to the edge server using an HTTP request. This allows the user terminal to send the audio data for immediate processing. Alternatively, the terminal's application converts the audio data to binary format and sends it to the edge server using an HTTP library (e.g., OkHttp).

[0847] 3. Retrieving text data from edge servers

[0848] The edge server processes the audio data and generates text data, which the user terminal receives as an HTTP response. The user terminal uses this text data in the next processing stage. To receive the HTTP response, the terminal uses asynchronous communication to wait for the response from the server and saves the received text data to its internal storage or memory.

[0849] 4. Submitting a translation request

[0850] The application sends a request back to the edge server to translate the received text data into the target language specified by the user. The application creates an HTTP POST request, including the text and target language as parameters.

[0851] 5. Retrieving and displaying translated data

[0852] The system receives translated text data from the edge server and displays this data to the user. This allows the user to see what they said in a translated form in a different language. It receives an HTTP response and displays the translated text in a UI component (e.g., TextView) on the user's device.

[0853] Edge server operation

[0854] 1. Receiving audio data

[0855] The edge server receives audio data transmitted from the user's terminal. The edge server prepares this data for analysis. It receives an HTTP request, decodes the audio data, and temporarily stores it in storage.

[0856] 2. Speech Recognition Processing

[0857] The edge server analyzes the received audio data using a speech recognition system and converts it into text data. This process employs advanced speech recognition algorithms. For example, the Google Cloud Speech-to-Text API can be used to convert audio data into text data. The audio data is input into the speech recognition system, and text data is generated. The generated text is temporarily stored in memory.

[0858] 3. Return of text data

[0859] The generated text data is sent back to the user's terminal. This allows the user's terminal to begin the next translation process. The text data is serialized into JSON format and sent as an HTTP response.

[0860] 4. Receiving a translation request

[0861] It receives translation requests for text data sent from the user's terminal. This is for converting text generated by speech recognition into a specified target language. It receives HTTP requests and parses the request parameters.

[0862] 5. Text translation processing

[0863] The received text data is translated into the target language. The edge server uses a high-performance translation engine to perform this transfer. For example, the Google Cloud Translation API is used to translate the text data. A request is sent to the translation API and the result is received.

[0864] 6. Return of translated data

[0865] The translated text data is sent back to the user's terminal. This allows the user to actually check the translation result. The translation result is sent as an HTTP response, and whether the communication was successful is logged.

[0866] Specific example

[0867] For example, a Japanese-speaking user says the phrase "Hello, how are you?" to the user terminal. This audio data is immediately captured and sent to the edge server. The edge server uses a speech recognition engine to convert this audio data into text "Hello, how are you?" and sends it back to the user terminal. Next, the user terminal resends this text data to the edge server, specifying English as the target language. The edge server translates the text data into English "Hello, how are you?" and finally sends it back to the user terminal. The user terminal displays this result, and the user can confirm the translation.

[0868] The following are specific examples of prompt statements to input to a generative AI model.

[0869] Example: "Convert the following Japanese audio to text, and then translate it into English. The audio says 'Hello, how are you?'"

[0870] This system can translate users' voices in real time and accurately, converting them into different languages. This makes it highly practical in diverse fields such as business, tourism, and education, effectively overcoming language barriers.

[0871] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0872] Step 1:

[0873] When a user begins a conversation, the microphone device on the user's device captures the audio. The input is an audio signal, and the output is digital data. Specifically, the smartphone's microphone picks up the audio and converts it into digital data through an ADC (analog-to-digital converter). This data is temporarily stored on the device.

[0874] Step 2:

[0875] The user terminal converts the captured audio data into binary format and sends it to the edge server using an HTTP POST request. The input is digital audio data, and the output is binary data. Specifically, the terminal's application converts the audio data into binary format and sends it to the edge server using a network library (e.g., OkHttp).

[0876] Step 3:

[0877] The edge server receives audio data sent from the user's terminal. The input is binary data sent in an HTTP request, and the output is audio data ready for analysis. Specifically, the server receives the HTTP request, decodes the binary data, and temporarily stores it in storage.

[0878] Step 4:

[0879] The edge server inputs the received audio data into a speech recognition engine and converts it into text data. The input is digital audio data, and the output is text data. Specifically, the edge server uses a speech recognition API (e.g., Google Cloud Speech-to-Text API) to analyze the audio and generate text data.

[0880] Step 5:

[0881] The edge server sends the generated text data to the user's terminal as an HTTP response. The input is text data, and the output is an HTTP response. Specifically, the server serializes the text data into JSON format and sends it to the user's terminal as an HTTP response.

[0882] Step 6:

[0883] The user terminal receives text data from the edge server and uses this data for the next processing step. The input is the text data sent in the HTTP response, and the output is the text data used in the request. Specifically, it receives the HTTP response and saves the text data to memory or internal storage.

[0884] Step 7:

[0885] The user terminal sends a request to the edge server to translate the received text data into the target language. The input is the text data and the specified target language, and the output is the translation request. Specifically, the application creates an HTTP POST request, including the text and target language as parameters.

[0886] Step 8:

[0887] The edge server receives translation requests from user terminals and inputs them into the translation engine. The input consists of text and the target language, and the output is translated text data. Specifically, the server calls a translation API (e.g., Google Cloud Translation API) to translate the text into the specified language.

[0888] Step 9:

[0889] The translated text data is sent back from the edge server to the user terminal. The input is the translated text data, and the output is the translated data as an HTTP response. Specifically, the server serializes the translated text into JSON format and sends it to the user terminal as an HTTP response.

[0890] Step 10:

[0891] The user terminal receives translated text data from the edge server and displays it to the user. The input is the translated data sent via HTTP response, and the output is the text data displayed on the user screen. Specifically, it receives the HTTP response and displays the translated text in a UI component (e.g., TextView). This allows the user to see the translated content in real time.

[0892] (Application Example 1)

[0893] Next, we will explain Application Example 1. In the following explanation, 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."

[0894] In multilingual food delivery services, a challenge exists: users often find it difficult to place orders smoothly in their native language. Therefore, it is necessary to provide a system that allows users who speak different languages ​​to comfortably use the service without experiencing language barriers. Furthermore, utilizing voice input is required to provide a more intuitive and convenient ordering experience.

[0895] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0896] In this invention, the server includes means for capturing voice data, means for transmitting the voice data to an edge computing server, means for the edge computing server to convert the voice data into text data, means for translating the text data into a specified language, means for transmitting the translated text data to a user terminal and for the user terminal to display the translated text data, means for the user terminal to place an order based on voice input, and means for the edge computing server to process the order details and notify the delivery service. This enables users to order food delivery using voice input in different languages, providing a comfortable ordering experience that transcends language barriers.

[0897] "Audio data" refers to data obtained by converting audio signals input by a user through a microphone device into a digital format.

[0898] "Capture method" refers to a function that captures and records audio data using the microphone device on the user's terminal.

[0899] An "edge computing server" is a server that receives voice and text data sent from user terminals and processes it in real time.

[0900] "Text data" refers to data in string format generated from audio data through speech recognition.

[0901] A "means of translation" is a function for converting text data written in one language into another specified language.

[0902] A "user terminal" refers to a device used by a user, such as a smartphone or tablet, and is a device for capturing audio and sending / receiving data.

[0903] "Means of placing an order" refers to a function that allows users to confirm their food delivery order via voice input or other means and send the details to the system.

[0904] "Means of notifying delivery services" refers to a function that transmits order details processed on edge computing servers to food delivery personnel and operations systems.

[0905] "Means of display" refers to functions that visually present translated text data and order information to the user on the user's terminal.

[0906] This invention is a system that captures voice data and translates it into different languages ​​in real time via an edge computing server. It is particularly applicable to food delivery services, enabling a multilingual ordering process.

[0907] The user terminal first has the function to capture audio data. This includes devices such as smartphones and tablets, which use microphone devices to capture audio as digital data. The captured audio data is converted to binary format and sent to the edge computing server via an HTTP request.

[0908] The edge computing server converts received audio data into text data through speech recognition processing. This speech recognition uses speech recognition libraries such as Google Cloud Speech-to-Text or Amazon Transcribe. The converted text data is then translated into the language specified by the user. High-performance translation engines such as the Google Cloud Translation API or Microsoft Translator Text API are used for the translation process.

[0909] The translated text data is sent back to the user's terminal and displayed on the user's screen. This allows the user to see the results of their voice input being translated into different languages. Furthermore, users can use voice input to place food delivery orders, which are processed on an edge computing server and notified to the delivery service.

[0910] For example, when a user says "I want to order curry" into their smartphone, this voice data is captured and sent to an edge server. The edge server converts the voice into text "I want to order curry," and then translates it into English as "I want to order curry." This translation is then displayed on the user's smartphone. The order details are also processed by the edge server and notified to the food delivery service.

[0911] Examples of prompt statements are as follows:

[0912] User: "I want to order curry."

[0913] Method:

[0914] 1. Audio signal capture.

[0915] 2. Sent audio signal to edge server.

[0916] 3. Received text: "I want to order curry."

[0917] 4. Sent text for translation.

[0918] 5. Received translated text: "I want to order curry"

[0919] 6. Displaying translated text: "I want to order curry"

[0920] This allows users to order food delivery through voice input in different languages, without being hindered by language barriers. This system can also be used for a variety of purposes, including business, tourism, and education.

[0921] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0922] Step 1:

[0923] The user captures audio data using the microphone device on their smartphone or tablet. As input, the user's voice signal is converted into a digital format. As output, digital audio data is generated. This data is captured at short time intervals and immediately prepared to proceed to the next processing step.

[0924] Step 2:

[0925] The terminal converts the captured audio data into binary format and sends it to the edge computing server as an HTTP request. Digital audio data exists as input. The output is the audio data converted to binary format and configured as the payload for the HTTP request. A POST request is sent to the URL of the edge computing server.

[0926] Step 3:

[0927] The server receives an HTTP request and retrieves audio data. The input is audio data in binary format, delivered via the HTTP request. The output is decoded audio data generated on the server side. This is then passed to the speech recognition process.

[0928] Step 4:

[0929] The server performs speech recognition processing and converts the audio data into text data. The input is the decoded audio data. The output is the text data generated by speech recognition. This process utilizes speech recognition libraries such as Google Cloud Speech-to-Text and Amazon Transcribe.

[0930] Step 5:

[0931] The server sends the generated text data back to the user's terminal. Text data exists as input. Text data is returned as an HTTP response as output. The user's terminal receives this data and proceeds to the next translation processing step.

[0932] Step 6:

[0933] The terminal sends a request back to the edge computing server to translate the received text data into the specified language. The input includes the text data and information about the target language. The output is an HTTP request sent back to the edge computing server.

[0934] Step 7:

[0935] The server receives a translation request and translates the text data into the target language. The input consists of the text data to be translated and information about the target language. The output is translated text data generated using a high-performance translation engine (e.g., Google Cloud Translation API or Microsoft Translator Text API).

[0936] Step 8:

[0937] The server returns the translated text data to the user's terminal. The input is the translated text data. The output is the translated text data returned as an HTTP response. The user's terminal receives this data and displays it visually to the user.

[0938] Step 9:

[0939] The device displays the translated text data on the screen for the user to review. The input is the translated text data. The output is the translation result displayed on the device screen. This allows the user to immediately see how their voice input has been translated.

[0940] Step 10:

[0941] When a user places an order using voice input, the terminal sends the order details to an edge computing server. The input is text data containing the user's order. The output is an HTTP request sent to the edge computing server. This allows the server to process the order information and notify the appropriate food delivery service.

[0942] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0943] The present invention is a system that captures voice data on a user terminal, transmits it to an edge computing server, converts the voice data into text data there, translates it into a specified language, performs emotion recognition using an emotion engine, and finally adds emotion information to the translated text data before returning it to the user terminal. Embodiments of the present invention will be described in detail below with specific examples.

[0944] User terminal operation

[0945] 1. Audio Capture

[0946] When a user begins a conversation, the microphone device on the user's device captures the audio. This audio is stored on the device as digital data. Because this audio data is captured at short intervals, real-time functionality can be maintained.

[0947] 2. Sending to the edge server

[0948] The captured audio data is converted to binary format and sent to a specified endpoint on the edge server (e.g., / transcribe) using an HTTP request. This allows the user terminal to send the audio data for immediate processing.

[0949] 3. Retrieving text data from edge servers

[0950] The edge server processes the audio data and generates text data, which the user terminal receives as an HTTP response. The user terminal then uses this text data in the next processing stage.

[0951] 4. Submitting a translation request

[0952] The system then sends a request to the edge server to translate the received text data into the target language specified by the user.

[0953] 5. Retrieving and displaying translated data

[0954] The system receives translated text data from the edge server and displays this data to the user. This allows the user to see what they said in a translated format in a different language.

[0955] Edge server operation

[0956] 1. Receiving audio data

[0957] The edge server receives audio data transmitted from the user's terminal. The edge server then prepares this data for analysis.

[0958] 2. Speech Recognition Processing

[0959] The edge server analyzes the received audio data using a speech recognition system and converts it into text data. This process employs advanced speech recognition algorithms.

[0960] 3. Return of text data

[0961] The generated text data is sent back to the user's terminal. This allows the user's terminal to begin the next translation process.

[0962] 4. Receiving a text translation request

[0963] It receives translation requests for text data sent from the user's terminal. This is for converting text generated by speech recognition into a specified target language.

[0964] 5. Text translation processing

[0965] The received text data is translated into the target language. The edge server uses a high-performance translation engine to perform this transfer.

[0966] 6. Performing emotion recognition

[0967] After translation processing, an emotion engine is used to analyze the emotions contained in the text data. Based on the emotion information obtained from the audio data, emotion tags are added to the text data.

[0968] 7. Return of translated data and sentiment information

[0969] The translated text data and added sentiment information are sent back to the user's terminal. This allows the user's terminal to display the translation result with the sentiment information included.

[0970] Specific examples

[0971] For example, a Japanese-speaking user says the phrase "Hello, how are you?" to the user terminal. This audio data is immediately captured and sent to the edge server. The edge server uses a speech recognition engine to convert this audio data into text "Hello, how are you?" and sends it back to the user terminal. Next, the user terminal resends this text data to the edge server, specifying English as the target language. The edge server translates the text data into English "Hello, how are you?" and uses an emotion engine to recognize the user's emotion (e.g., positive emotion) from the original Japanese audio data. This emotion information is added to the translation result and finally sent back to the user terminal. As a result, the user terminal displays the translated text with emotion information, such as "Hello, how are you? [Positive]".

[0972] Through the process described above, the system of the present invention can translate speech into different languages ​​in real time and with ultra-low latency, and can also add emotional information. This system has high practical applications in various fields such as business, tourism, and education, and helps to effectively overcome language barriers.

[0973] The following describes the processing flow.

[0974] Step 1:

[0975] When a user starts a conversation, their device captures the audio. Specifically, it uses a microphone to pick up the user's voice and converts it to a digital format in real time.

[0976] Step 2:

[0977] The device captures audio data, converts it to binary format, and sends it to the edge computing server using an HTTP request. This request includes the audio data.

[0978] Step 3:

[0979] The edge server receives the audio data and the speech recognition engine begins analysis. Specifically, the process of converting the audio data into text data is executed.

[0980] Step 4:

[0981] The edge server sends the text data generated as a result of speech recognition back to the user terminal as an HTTP response. The user terminal receives this response and extracts the text data.

[0982] Step 5:

[0983] The user's terminal uses the received text data to resend the translation request to the edge computing server. This request includes information about the original text and the target language.

[0984] Step 6:

[0985] The edge server receives the text translation request and uses a high-performance translation engine to translate the text data into the specified target language.

[0986] Step 7:

[0987] The edge server performs sentiment recognition on the translated text data. Specifically, the sentiment engine analyzes the audio data and text to evaluate relevant sentiment information.

[0988] Step 8:

[0989] The edge server adds sentiment information to the translation result and sends it to the user terminal as an HTTP response. The user terminal receives this response and extracts the translated text data and sentiment information.

[0990] Step 9:

[0991] The user's device displays the final translation result and sentiment information to the user. Specifically, it visually displays the translated text and its sentiment on the screen, making it easy for the user to understand.

[0992] (Example 2)

[0993] Next, we will describe Example 2. 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."

[0994] Conventional speech recognition systems could convert speech data to text and translate languages, but they could not add emotional information, making it difficult to achieve communication that reflected the user's emotions. Furthermore, they often lacked processing speed and real-time capabilities, limiting their practicality. As a result, they were unable to effectively overcome language barriers in fields such as business, tourism, and education.

[0995] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0996] In this invention, the server includes means for capturing audio data, means for transmitting the audio data to an edge computing server, means for the edge computing server to convert the audio data into text data, means for translating the text data into a specified language, means for adding sentiment information to the translated text data, and means for transmitting the translated text data and sentiment information to a user terminal. This enables the addition of sentiment information in addition to the conversion and translation of audio data into text. As a result, users can use text with sentiment information that has been translated in real time, enabling richer communication. Furthermore, high-speed edge computing improves processing speed and real-time capabilities, increasing its practicality in various fields.

[0997] "Audio data" refers to a digital audio signal acquired using a microphone device.

[0998] An "edge computing server" refers to a server located close to the user's terminal that performs data processing with an emphasis on real-time performance.

[0999] "Text data" refers to string data generated by analyzing audio data.

[1000] "Translation" refers to the process of converting text data expressed in one language into another language.

[1001] "Emotional information" refers to information that indicates a user's emotions (e.g., positive, negative, neutral) as recognized from text data or audio data.

[1002] An "HTTP request" refers to a request made by a client to a server using a protocol for sending data.

[1003] This invention relates to a system that captures audio data, converts it into text data in real time, translates it into a specified language, and adds emotional information. This system operates based on a user terminal and an edge computing server.

[1004] User terminal operation

[1005] First, the user terminal is equipped with a microphone device that captures the user's voice. For example, when the user says "Hello, how are you?", this voice is captured by the terminal as digital audio data. Next, the user terminal converts the captured audio data into binary format and sends it to a specified endpoint on the edge computing server using an HTTP request. An example of such an endpoint is " / transcribe".

[1006] Edge computing server-side operation

[1007] The edge computing server receives audio data sent from the user's terminal. This received audio data is converted into text data using a speech recognition system (the API used as an example provides a speech recognition algorithm). Examples of speech recognition systems that can be used include the Google Speech-to-Text API. The converted text data is sent back to the user's terminal via an HTTP response.

[1008] Next, the user terminal sends the acquired text data back to the edge computing server, requesting a translation into the specified language. This translation process uses a high-performance translation engine, such as the Google Translate API. Once the translation is complete, the server analyzes the translated text data with a sentiment engine (for example, IBM Watson Tone Analyzer) and adds sentiment information. This translated data with added sentiment information is then sent back to the user terminal as an HTTP response.

[1009] Display of data by the user terminal

[1010] Finally, the user terminal displays the received translation data and sentiment information to the user. This allows the user to see their spoken content translated into different languages ​​in real time, with added sentiment information.

[1011] Specific example

[1012] For example, a Japanese-speaking user says the phrase "Hello, how are you?" to the user terminal. This audio data is immediately captured and sent to the edge computing server. The edge computing server uses a speech recognition engine to convert the audio data into text "Hello, how are you?" and sends it back to the user terminal. The user terminal resends this text data to the edge computing server, specifying English as the target language. The edge server translates the text data into English "Hello, how are you?" and uses an emotion engine to recognize the user's emotion (e.g., positive emotion) from the original Japanese audio data. This emotion information is added to the translation result and finally sent back to the user terminal. As a result, the user terminal displays the translated text with emotion information, such as "Hello, how are you? [Positive]".

[1013] Through the process described above, the system of the present invention can translate speech into different languages ​​in real time and with ultra-low latency, and can also add emotional information. This system has high practical applications in various fields such as business, tourism, and education, and helps to effectively overcome language barriers.

[1014] Example of a prompt

[1015] For example, by inputting the following prompt into the generating AI model, you can obtain detailed instructions for performing the aforementioned process:

[1016] Prompt message:

[1017] "Capture Japanese audio data, translate it into English, and display the translation result with added sentiment information."

[1018] This prompt allows the generative AI model to suggest specific technical measures, such as how to use the appropriate API and how to specify the endpoint.

[1019] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1020] Step 1:

[1021] The user starts voice capture.

[1022] Input: User's voice (e.g., "Hello, how are you?")

[1023] Specific action: The user begins speaking into the device's microphone. The microphone captures this audio as a digital signal.

[1024] Output: Digital audio data

[1025] Step 2:

[1026] The device acquires voice data and sends it to the edge server.

[1027] Input: Captured digital audio data

[1028] Specific operation: The terminal converts the captured audio data into binary format and saves it to a temporary file. Then, it sends it to the specified endpoint on the edge server (e.g., " / transcribe") using an HTTP request.

[1029] Output: Audio data sent to the edge server

[1030] Step 3:

[1031] The edge server converts the audio data into text data.

[1032] Input: Binary format audio data sent from the terminal.

[1033] Specific operation: The edge server receives the audio data, analyzes it using a speech recognition system (e.g., an API that provides a speech recognition algorithm), and converts it into text data.

[1034] Output: Converted text data (e.g., "Hello, how are you?")

[1035] Step 4:

[1036] The edge server sends text data back to the terminal.

[1037] Input: Text data generated by speech recognition

[1038] Specific operation: The edge server sends the generated text data back to the user terminal as an HTTP response. The terminal saves this text data for use in the next process.

[1039] Output: Text data sent to the user's terminal

[1040] Step 5:

[1041] The device sends text data to the edge server and requests a translation.

[1042] Input: Text data received from the edge server

[1043] Specific action: The user terminal sends a request to the edge server again to translate the saved text data into the target language. English is specified as the target language.

[1044] Output: Text data with translation request sent to the edge server

[1045] Step 6:

[1046] The edge server translates the text data.

[1047] Input: Text data with translation request sent from the user terminal.

[1048] Specific operation: The edge server analyzes the received text data using a translation engine (e.g., an API that provides translation algorithms) and translates it into the specified target language (in this case, English).

[1049] Output: Translated text data (e.g., "Hello, how are you?")

[1050] Step 7:

[1051] The edge server analyzes the sentiment of the translated data and adds emotional information.

[1052] Input: Translated text data

[1053] Specific operation: The edge server analyzes the translated text data using an emotion engine (e.g., an API that provides an emotion analysis algorithm) and recognizes the emotions contained in the text (e.g., positive, negative, neutral). This emotion information is then added to the translated data as a tag.

[1054] Output: Translated text data with added sentiment information (e.g., "Hello, how are you? [Positive]")

[1055] Step 8:

[1056] The edge server sends the translation data and sentiment information back to the terminal.

[1057] Input: Translated text data with added emotional information

[1058] Specific operation: The edge server sends translated data with sentiment information back to the user terminal as an HTTP response. The user terminal receives this data.

[1059] Output: Translated data with sentiment information sent to the user's terminal.

[1060] Step 9:

[1061] The device displays translation data and sentiment information to the user.

[1062] Input: Translation data with emotional information

[1063] Specific operation: The user terminal displays the translated data with sentiment information received on the user interface. The user can then view the translated text and its sentiment information on the screen.

[1064] Output: Translated text with sentiment information displayed to the user (e.g., "Hello, how are you? [Positive]")

[1065] This process allows users to translate speech into different languages ​​in real time and then use it in a format that includes emotional information.

[1066] (Application Example 2)

[1067] Next, we will explain application example 2. In the following explanation, 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."

[1068] In modern society, there is an increasing demand for multilingual support and real-time emotion recognition. However, conventional technologies are unable to adequately translate voice data in real time or add emotional information, leading to decreased communication efficiency. Therefore, there is a need for a system that integrates multilingual support and emotion recognition. In particular, in security services, smooth communication and situational awareness at the site are crucial, and technologies that can solve this problem are in demand.

[1069] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1070] In this invention, the server includes means for capturing voice data, means for transmitting the voice data to an edge computing server, means for the edge computing server to convert the voice data into text data, means for translating the text data into a specified language, means for adding sentiment information to the translated text data, and means for transmitting the translated text data and sentiment information to a user terminal. This enables multilingual real-time translation and sentiment recognition.

[1071] "Audio data" refers to data that represents audio in a digital format.

[1072] "Means of capturing" refers to devices or methods that record a user's voice and acquire it as digital data.

[1073] An "edge computing server" is a distributed server that processes data sent from a user's terminal in real time at a location close to the local area.

[1074] "Text data" refers to data that represents audio data in written form.

[1075] "Translation methods" refer to the techniques and methods used to convert text data expressed in the original language into another specified language.

[1076] "Means of adding emotional information" refers to technologies and methods that analyze the emotions contained in text data and add the results to the data as annotations.

[1077] A "user terminal" is a device (e.g., smart glasses, smartphone, etc.) used to capture audio data and communicate with a server.

[1078] This invention relates to a system that translates voice data into multiple languages ​​in real time and adds emotional information to that data. This system aims to improve effective communication and situational awareness, particularly in security services. The following describes embodiments for carrying out this invention using specific examples.

[1079] System Configuration

[1080] This system consists of user terminals (such as smart glasses or smartphones) that capture voice data and edge computing servers that process the data.

[1081] User terminal operation

[1082] 1. Audio Capture

[1083] When a user begins a conversation, the microphone device on the user's device captures the audio. The captured audio data is collected at short intervals, and the digital data is stored on the device.

[1084] 2. Sending to the edge computing server

[1085] The captured audio data is converted to binary format and sent to the edge computing server using an HTTP request.

[1086] Edge computing server operation

[1087] 1. Receiving and transcribing audio data

[1088] The edge computing server analyzes the audio data received from the user's terminal and converts it into text data using speech recognition software (e.g., Google Speech-to-Text API).

[1089] 2. Translation of text data

[1090] The converted text data is translated into the specified language. A high-performance translation engine (e.g., Google Translate API) is used for the translation.

[1091] 3. Adding emotional information

[1092] The translated text data is subjected to an emotion recognition engine, and emotional information (e.g., positive, negative, neutral) is added. An AI model (e.g., a BERT-based emotion analysis model) is used for emotion recognition.

[1093] Return to user terminal

[1094] The translated text data and sentiment information processed on the edge computing server are sent back to the user terminal via an HTTP response, and the user terminal displays it.

[1095] Specific example

[1096] A security guard is seen conversing with a foreign tourist.

[1097] 1. A tourist says in English, "Excuse me, can you help me find the museum?"

[1098] 2. The security guard's smart glasses capture this audio and send the data to an edge computing server.

[1099] 3. The server converts the speech to text and obtains "Excuse me, can you help me find the museum?".

[1100] 4. The text data is translated into Japanese, and the text "Excuse me, could you help me find the museum?" is generated.

[1101] 5. The translated text is tagged with sentiment and sent back to the user's device as "Excuse me, could you help me find the museum? [Positive]".

[1102] 6. The information is ultimately displayed on the security guard's smart glasses, allowing the guard to take appropriate action.

[1103] Example of a prompt

[1104] User's voice: "Excuse me, can you help me find the museum?"

[1105] Audio data capture:

[1106] Sending voice data to the edge server

[1107] Speech recognition and text conversion: "Excuse me, can you help me find the museum?"

[1108] Text translation: "Excuse me, could you help me find the museum?"

[1109] Emotional perception: "Positive"

[1110] Final message: "Excuse me, could you help me find the museum? [Positive]"

[1111] This enables the system to integrate multilingual support and emotion recognition, providing real-time feedback to users. Furthermore, it can support smooth communication and rapid decision-making in security services.

[1112] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1113] Step 1:

[1114] Audio Capture

[1115] The user's device (e.g., smart glasses, smartphone) captures the user's voice. When the user starts speaking, the device's microphone collects the voice and stores it as digital data. The input is voice data, and the output is digital voice data.

[1116] Step 2:

[1117] Sending digital audio data to an edge computing server

[1118] The terminal converts the captured digital audio data into binary format and sends it to the edge computing server using an HTTP request. The input is digital audio data, and the output is binary audio data sent as an HTTP request.

[1119] Step 3:

[1120] Receiving and transcribing audio data by the server.

[1121] The server receives binary audio data sent from the terminal. The received audio data is converted into text data using speech recognition software (e.g., Google Speech-to-Text API). The input is binary audio data, and the output is text data.

[1122] Step 4:

[1123] Text data translation

[1124] The server translates the converted text data into the specified language. A high-performance translation engine (e.g., Google Translate API) is used for the translation. The input is text data, and the output is translated text data in the target language.

[1125] Step 5:

[1126] Adding emotional information

[1127] The server analyzes the translated text data using an emotion recognition engine (e.g., a BERT-based emotion analysis model) and adds emotion information. The input is the translated text data, and the output is the text data with emotion information added.

[1128] Step 6:

[1129] Sending translated text data and sentiment information back to the device

[1130] The server sends translated text data with added sentiment information back to the user's terminal using an HTTP response. The input is the text data with added sentiment information, and the output is the text data sent as an HTTP response.

[1131] Step 7:

[1132] Display of results on the user terminal

[1133] The user terminal receives translated text data with sentiment information sent back from the server and displays it to the user. The input is text data with sentiment information added, and the output is the translated result displayed to the user.

[1134] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1135] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1136] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1137] [Fourth Embodiment]

[1138] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1139] As shown in Figure 7, the 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.

[1140] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1141] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1142] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1143] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1144] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1145] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1146] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1147] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[1148] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1149] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1150] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1151] This invention is a system that captures audio data on a user terminal, transmits it to an edge computing server, converts the audio data into text data there, translates it into a specified language, and finally returns the translated text data to the user terminal. Embodiments of the present invention will be described in detail below with specific examples.

[1152] User terminal operation

[1153] 1. Audio Capture

[1154] When a user begins a conversation, the microphone device on the user's device captures the audio. This audio is stored on the device as digital data. Because this audio data is captured at short intervals, real-time functionality can be maintained.

[1155] 2. Sending to the edge server

[1156] The captured audio data is converted to binary format and sent to the edge server using an HTTP request. This allows the user terminal to send the audio data for immediate processing.

[1157] 3. Retrieving text data from edge servers

[1158] The edge server processes the audio data and generates text data, which the user terminal receives as an HTTP response. The user terminal then uses this text data in the next processing stage.

[1159] 4. Submitting a translation request

[1160] The system then sends a request to the edge server to translate the received text data into the target language specified by the user.

[1161] 5. Retrieving and displaying translated data

[1162] The system receives translated text data from the edge server and displays this data to the user. This allows the user to see what they said in a translated format in a different language.

[1163] Edge server operation

[1164] 1. Receiving audio data

[1165] The edge server receives audio data transmitted from the user's terminal. The edge server then prepares this data for analysis.

[1166] 2. Speech Recognition Processing

[1167] The edge server analyzes the received audio data using a speech recognition system and converts it into text data. This process employs advanced speech recognition algorithms.

[1168] 3. Return of text data

[1169] The generated text data is sent back to the user's terminal. This allows the user's terminal to begin the next translation process.

[1170] 4. Receiving a translation request

[1171] It receives translation requests for text data sent from the user's terminal. This is for converting text generated by speech recognition into a specified target language.

[1172] 5. Text translation processing

[1173] The received text data is translated into the target language. The edge server uses a high-performance translation engine to perform this transfer.

[1174] 6. Return of translated data

[1175] The translated text data is sent back to the user's terminal. This allows the user to actually check the translation results.

[1176] Specific examples

[1177] For example, a Japanese-speaking user says the phrase "Hello, how are you?" to the user terminal. This audio data is immediately captured and sent to the edge server. The edge server uses a speech recognition engine to convert this audio data into text "Hello, how are you?" and sends it back to the user terminal. Next, the user terminal resends this text data to the edge server, specifying English as the target language. The edge server translates the text data into English "Hello, how are you?" and finally sends it back to the user terminal. The user terminal displays this result, and the user can confirm the translation.

[1178] Through the process described above, the system of the present invention can translate speech into different languages ​​in real time and with ultra-low latency, and provide it to the user. This system has high practical applications in various fields such as business, tourism, and education, and can effectively overcome language barriers.

[1179] The following describes the processing flow.

[1180] Step 1:

[1181] When a user begins voice input, the user's device captures the audio data using its microphone. Specifically, it converts the user's spoken voice into a digital format in real time and temporarily stores it on the device.

[1182] Step 2:

[1183] The device converts the captured audio data into binary format and sends it to a specified endpoint on the edge computing server (e.g., / transcribe) using an HTTP request. The audio data is attached to this request.

[1184] Step 3:

[1185] The edge server receives the audio data and the speech recognition engine performs the analysis. Specifically, it initiates the process of converting the audio data into text data.

[1186] Step 4:

[1187] The edge server sends the text data generated as a result of speech recognition back to the user terminal as an HTTP response. The user terminal receives this response and extracts the text data.

[1188] Step 5:

[1189] The user's terminal uses the received text data to resend the translation request to the edge computing server. This request includes information about the original text and the target language.

[1190] Step 6:

[1191] The edge server processes the received translation request and translates the text data into the specified target language. Specifically, it uses a translation engine on the server to translate the text.

[1192] Step 7:

[1193] The edge server sends the translated text data back to the user's terminal. The user terminal receives this response and extracts the translated text data.

[1194] Step 8:

[1195] The user's device displays the final translation result to the user. The user can check the translated text displayed on the screen.

[1196] (Example 1)

[1197] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1198] Conventional translation systems require real-time performance and accuracy in the process of capturing audio data, converting it to text, and then translating it. Especially in today's society where multilingual communication is essential, systems that perform these processes quickly and efficiently are indispensable. However, delays often occur between capturing audio data and obtaining translated text data, which is stressful for users. Furthermore, challenges exist regarding data transmission formats and security, making it difficult to balance system reliability and speed. Therefore, there is a need for the development of a faster and more accurate real-time speech translation system.

[1199] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1200] In this invention, the server includes means for capturing audio data, means for transmitting the audio data to an edge computing server, means for the edge computing server to convert the audio data into text data, means for translating the text data into a specified language, means for transmitting the translated text data to a user terminal, means for capturing and transmitting audio data at short intervals to maintain real-time performance, means for sending and receiving data between the edge computing server and the user terminal using HTTP requests and responses, and means for displaying the text data on the screen of the user terminal. This enables users to translate audio between multiple languages ​​in real time and accurately, allowing for rapid communication.

[1201] "Audio data" refers to data that records the voice spoken by a user in digital format.

[1202] "Means of capturing" refers to a device or method for acquiring a user's voice as digital data through a microphone device.

[1203] An "edge computing server" is a distributed computing resource located close to the user's terminal for processing voice data.

[1204] "Means of transmission" refers to methods or devices for transferring data from a user terminal to an edge computing server via communication.

[1205] "Means of converting to text data" refers to processes or systems that convert audio data into string data.

[1206] "Specified language" refers to the target language that the user wishes to have translated into.

[1207] "Translation means" refers to the process or system of converting generated text data into a specified target language.

[1208] "Means of sending and receiving" refers to communication methods for transmitting and receiving data.

[1209] "Maintaining real-time performance" means minimizing the delay between capturing audio data and displaying translated text data.

[1210] "Short interval" means processing or transmitting data in the shortest possible time units, such as every 100 milliseconds.

[1211] An "HTTP request" is a request from a client to a server to send data, and it is a protocol primarily used for web communication.

[1212] An "HTTP response" is a data response that a server provides to a client, and it refers to the result of an HTTP request.

[1213] "Means of display" refers to methods or devices for displaying text data on the screen of a user's terminal.

[1214] This invention is a system that captures audio data on a user terminal, sends it to an edge computing server, converts the audio data into text data there, translates it into a specified language, and finally sends the translated text data back to the user terminal.

[1215] User terminal operation

[1216] 1. Audio Capture

[1217] When a user begins a conversation, the microphone device on the user's device captures the audio. This audio is stored on the device as digital data. Because this audio data is captured at short intervals, real-time functionality can be maintained. For example, a smartphone's microphone picks up sound and converts it into digital data through an analog-to-digital converter.

[1218] 2. Sending to the edge server

[1219] The captured audio data is converted to binary format and sent to the edge server using an HTTP request. This allows the user terminal to send the audio data for immediate processing. Alternatively, the terminal's application converts the audio data to binary format and sends it to the edge server using an HTTP library (e.g., OkHttp).

[1220] 3. Retrieving text data from edge servers

[1221] The edge server processes the audio data and generates text data, which the user terminal receives as an HTTP response. The user terminal uses this text data in the next processing stage. To receive the HTTP response, the terminal uses asynchronous communication to wait for the response from the server and saves the received text data to its internal storage or memory.

[1222] 4. Submitting a translation request

[1223] The application sends a request back to the edge server to translate the received text data into the target language specified by the user. The application creates an HTTP POST request, including the text and target language as parameters.

[1224] 5. Retrieving and displaying translated data

[1225] The system receives translated text data from the edge server and displays this data to the user. This allows the user to see what they said in a translated form in a different language. It receives an HTTP response and displays the translated text in a UI component (e.g., TextView) on the user's device.

[1226] Edge server operation

[1227] 1. Receiving audio data

[1228] The edge server receives audio data transmitted from the user's terminal. The edge server prepares this data for analysis. It receives an HTTP request, decodes the audio data, and temporarily stores it in storage.

[1229] 2. Speech Recognition Processing

[1230] The edge server analyzes the received audio data using a speech recognition system and converts it into text data. This process employs advanced speech recognition algorithms. For example, the Google Cloud Speech-to-Text API can be used to convert audio data into text data. The audio data is input into the speech recognition system, and text data is generated. The generated text is temporarily stored in memory.

[1231] 3. Return of text data

[1232] The generated text data is sent back to the user's terminal. This allows the user's terminal to begin the next translation process. The text data is serialized into JSON format and sent as an HTTP response.

[1233] 4. Receiving a translation request

[1234] It receives translation requests for text data sent from the user's terminal. This is for converting text generated by speech recognition into a specified target language. It receives HTTP requests and parses the request parameters.

[1235] 5. Text translation processing

[1236] The received text data is translated into the target language. The edge server uses a high-performance translation engine to perform this transfer. For example, the Google Cloud Translation API is used to translate the text data. A request is sent to the translation API and the result is received.

[1237] 6. Return of translated data

[1238] The translated text data is sent back to the user's terminal. This allows the user to actually check the translation result. The translation result is sent as an HTTP response, and whether the communication was successful is logged.

[1239] Specific example

[1240] For example, a Japanese-speaking user says the phrase "Hello, how are you?" to the user terminal. This audio data is immediately captured and sent to the edge server. The edge server uses a speech recognition engine to convert this audio data into text "Hello, how are you?" and sends it back to the user terminal. Next, the user terminal resends this text data to the edge server, specifying English as the target language. The edge server translates the text data into English "Hello, how are you?" and finally sends it back to the user terminal. The user terminal displays this result, and the user can confirm the translation.

[1241] The following are specific examples of prompt statements to input to a generative AI model.

[1242] Example: "Convert the following Japanese audio to text, and then translate it into English. The audio says 'Hello, how are you?'"

[1243] This system can translate users' voices in real time and accurately, converting them into different languages. This makes it highly practical in diverse fields such as business, tourism, and education, effectively overcoming language barriers.

[1244] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1245] Step 1:

[1246] When a user begins a conversation, the microphone device on the user's device captures the audio. The input is an audio signal, and the output is digital data. Specifically, the smartphone's microphone picks up the audio and converts it into digital data through an ADC (analog-to-digital converter). This data is temporarily stored on the device.

[1247] Step 2:

[1248] The user terminal converts the captured audio data into binary format and sends it to the edge server using an HTTP POST request. The input is digital audio data, and the output is binary data. Specifically, the terminal's application converts the audio data into binary format and sends it to the edge server using a network library (e.g., OkHttp).

[1249] Step 3:

[1250] The edge server receives audio data sent from the user's terminal. The input is binary data sent in an HTTP request, and the output is audio data ready for analysis. Specifically, the server receives the HTTP request, decodes the binary data, and temporarily stores it in storage.

[1251] Step 4:

[1252] The edge server inputs the received audio data into a speech recognition engine and converts it into text data. The input is digital audio data, and the output is text data. Specifically, the edge server uses a speech recognition API (e.g., Google Cloud Speech-to-Text API) to analyze the audio and generate text data.

[1253] Step 5:

[1254] The edge server sends the generated text data to the user's terminal as an HTTP response. The input is text data, and the output is an HTTP response. Specifically, the server serializes the text data into JSON format and sends it to the user's terminal as an HTTP response.

[1255] Step 6:

[1256] The user terminal receives text data from the edge server and uses this data for the next processing step. The input is the text data sent in the HTTP response, and the output is the text data used in the request. Specifically, it receives the HTTP response and saves the text data to memory or internal storage.

[1257] Step 7:

[1258] The user terminal sends a request to the edge server to translate the received text data into the target language. The input is the text data and the specified target language, and the output is the translation request. Specifically, the application creates an HTTP POST request, including the text and target language as parameters.

[1259] Step 8:

[1260] The edge server receives translation requests from user terminals and inputs them into the translation engine. The input consists of text and the target language, and the output is translated text data. Specifically, the server calls a translation API (e.g., Google Cloud Translation API) to translate the text into the specified language.

[1261] Step 9:

[1262] The translated text data is sent back from the edge server to the user terminal. The input is the translated text data, and the output is the translated data as an HTTP response. Specifically, the server serializes the translated text into JSON format and sends it to the user terminal as an HTTP response.

[1263] Step 10:

[1264] The user terminal receives translated text data from the edge server and displays it to the user. The input is the translated data sent via HTTP response, and the output is the text data displayed on the user screen. Specifically, it receives the HTTP response and displays the translated text in a UI component (e.g., TextView). This allows the user to see the translated content in real time.

[1265] (Application Example 1)

[1266] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1267] In multilingual food delivery services, a challenge exists: users often find it difficult to place orders smoothly in their native language. Therefore, it is necessary to provide a system that allows users who speak different languages ​​to comfortably use the service without experiencing language barriers. Furthermore, utilizing voice input is required to provide a more intuitive and convenient ordering experience.

[1268] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1269] In this invention, the server includes means for capturing voice data, means for transmitting the voice data to an edge computing server, means for the edge computing server to convert the voice data into text data, means for translating the text data into a specified language, means for transmitting the translated text data to a user terminal and for the user terminal to display the translated text data, means for the user terminal to place an order based on voice input, and means for the edge computing server to process the order details and notify the delivery service. This enables users to order food delivery using voice input in different languages, providing a comfortable ordering experience that transcends language barriers.

[1270] "Audio data" refers to data obtained by converting audio signals input by a user through a microphone device into a digital format.

[1271] "Capture method" refers to a function that captures and records audio data using the microphone device on the user's terminal.

[1272] An "edge computing server" is a server that receives voice and text data sent from user terminals and processes it in real time.

[1273] "Text data" refers to data in string format generated from audio data through speech recognition.

[1274] A "means of translation" is a function for converting text data written in one language into another specified language.

[1275] A "user terminal" refers to a device used by a user, such as a smartphone or tablet, and is a device for capturing audio and sending / receiving data.

[1276] "Means of placing an order" refers to a function that allows users to confirm their food delivery order via voice input or other means and send the details to the system.

[1277] "Means of notifying delivery services" refers to a function that transmits order details processed on edge computing servers to food delivery personnel and operations systems.

[1278] "Means of display" refers to functions that visually present translated text data and order information to the user on the user's terminal.

[1279] This invention is a system that captures voice data and translates it into different languages ​​in real time via an edge computing server. It is particularly applicable to food delivery services, enabling a multilingual ordering process.

[1280] The user terminal first has the function to capture audio data. This includes devices such as smartphones and tablets, which use microphone devices to capture audio as digital data. The captured audio data is converted to binary format and sent to the edge computing server via an HTTP request.

[1281] The edge computing server converts received audio data into text data through speech recognition processing. This speech recognition uses speech recognition libraries such as Google Cloud Speech-to-Text or Amazon Transcribe. The converted text data is then translated into the language specified by the user. High-performance translation engines such as the Google Cloud Translation API or Microsoft Translator Text API are used for the translation process.

[1282] The translated text data is sent back to the user's terminal and displayed on the user's screen. This allows the user to see the results of their voice input being translated into different languages. Furthermore, users can use voice input to place food delivery orders, which are processed on an edge computing server and notified to the delivery service.

[1283] For example, when a user says "I want to order curry" into their smartphone, this voice data is captured and sent to an edge server. The edge server converts the voice into text "I want to order curry," and then translates it into English as "I want to order curry." This translation is then displayed on the user's smartphone. The order details are also processed by the edge server and notified to the food delivery service.

[1284] Examples of prompt statements are as follows:

[1285] User: "I want to order curry."

[1286] Method:

[1287] 1. Audio signal capture.

[1288] 2. Sent audio signal to edge server.

[1289] 3. Received text: "I want to order curry."

[1290] 4. Sent text for translation.

[1291] 5. Received translated text: "I want to order curry"

[1292] 6. Displaying translated text: "I want to order curry"

[1293] This allows users to order food delivery through voice input in different languages, without being hindered by language barriers. This system can also be used for a variety of purposes, including business, tourism, and education.

[1294] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1295] Step 1:

[1296] The user captures audio data using the microphone device on their smartphone or tablet. As input, the user's voice signal is converted into a digital format. As output, digital audio data is generated. This data is captured at short time intervals and immediately prepared to proceed to the next processing step.

[1297] Step 2:

[1298] The terminal converts the captured audio data into binary format and sends it to the edge computing server as an HTTP request. Digital audio data exists as input. The output is the audio data converted to binary format and configured as the payload for the HTTP request. A POST request is sent to the URL of the edge computing server.

[1299] Step 3:

[1300] The server receives an HTTP request and retrieves audio data. The input is audio data in binary format, delivered via the HTTP request. The output is decoded audio data generated on the server side. This is then passed to the speech recognition process.

[1301] Step 4:

[1302] The server performs speech recognition processing and converts the audio data into text data. The input is the decoded audio data. The output is the text data generated by speech recognition. This process utilizes speech recognition libraries such as Google Cloud Speech-to-Text and Amazon Transcribe.

[1303] Step 5:

[1304] The server sends the generated text data back to the user's terminal. Text data exists as input. Text data is returned as an HTTP response as output. The user's terminal receives this data and proceeds to the next translation processing step.

[1305] Step 6:

[1306] The terminal sends a request back to the edge computing server to translate the received text data into the specified language. The input includes the text data and information about the target language. The output is an HTTP request sent back to the edge computing server.

[1307] Step 7:

[1308] The server receives a translation request and translates the text data into the target language. The input consists of the text data to be translated and information about the target language. The output is translated text data generated using a high-performance translation engine (e.g., Google Cloud Translation API or Microsoft Translator Text API).

[1309] Step 8:

[1310] The server returns the translated text data to the user's terminal. The input is the translated text data. The output is the translated text data returned as an HTTP response. The user's terminal receives this data and displays it visually to the user.

[1311] Step 9:

[1312] The device displays the translated text data on the screen for the user to review. The input is the translated text data. The output is the translation result displayed on the device screen. This allows the user to immediately see how their voice input has been translated.

[1313] Step 10:

[1314] When a user places an order using voice input, the terminal sends the order details to an edge computing server. The input is text data containing the user's order. The output is an HTTP request sent to the edge computing server. This allows the server to process the order information and notify the appropriate food delivery service.

[1315] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1316] The present invention is a system that captures voice data on a user terminal, transmits it to an edge computing server, converts the voice data into text data there, translates it into a specified language, performs emotion recognition using an emotion engine, and finally adds emotion information to the translated text data before returning it to the user terminal. Embodiments of the present invention will be described in detail below with specific examples.

[1317] User terminal operation

[1318] 1. Audio Capture

[1319] When a user begins a conversation, the microphone device on the user's device captures the audio. This audio is stored on the device as digital data. Because this audio data is captured at short intervals, real-time functionality can be maintained.

[1320] 2. Sending to the edge server

[1321] The captured audio data is converted to binary format and sent to a specified endpoint on the edge server (e.g., / transcribe) using an HTTP request. This allows the user terminal to send the audio data for immediate processing.

[1322] 3. Retrieving text data from edge servers

[1323] The edge server processes the audio data and generates text data, which the user terminal receives as an HTTP response. The user terminal then uses this text data in the next processing stage.

[1324] 4. Submitting a translation request

[1325] The system then sends a request to the edge server to translate the received text data into the target language specified by the user.

[1326] 5. Retrieving and displaying translated data

[1327] The system receives translated text data from the edge server and displays this data to the user. This allows the user to see what they said in a translated format in a different language.

[1328] Edge server operation

[1329] 1. Receiving audio data

[1330] The edge server receives audio data transmitted from the user's terminal. The edge server then prepares this data for analysis.

[1331] 2. Speech Recognition Processing

[1332] The edge server analyzes the received audio data using a speech recognition system and converts it into text data. This process employs advanced speech recognition algorithms.

[1333] 3. Return of text data

[1334] The generated text data is sent back to the user's terminal. This allows the user's terminal to begin the next translation process.

[1335] 4. Receiving a text translation request

[1336] It receives translation requests for text data sent from the user's terminal. This is for converting text generated by speech recognition into a specified target language.

[1337] 5. Text translation processing

[1338] The received text data is translated into the target language. The edge server uses a high-performance translation engine to perform this transfer.

[1339] 6. Performing emotion recognition

[1340] After translation processing, an emotion engine is used to analyze the emotions contained in the text data. Based on the emotion information obtained from the audio data, emotion tags are added to the text data.

[1341] 7. Return of translated data and sentiment information

[1342] The translated text data and added sentiment information are sent back to the user's terminal. This allows the user's terminal to display the translation result with the sentiment information included.

[1343] Specific examples

[1344] For example, a Japanese-speaking user says the phrase "Hello, how are you?" to the user terminal. This audio data is immediately captured and sent to the edge server. The edge server uses a speech recognition engine to convert this audio data into text "Hello, how are you?" and sends it back to the user terminal. Next, the user terminal resends this text data to the edge server, specifying English as the target language. The edge server translates the text data into English "Hello, how are you?" and uses an emotion engine to recognize the user's emotion (e.g., positive emotion) from the original Japanese audio data. This emotion information is added to the translation result and finally sent back to the user terminal. As a result, the user terminal displays the translated text with emotion information, such as "Hello, how are you? [Positive]".

[1345] Through the process described above, the system of the present invention can translate speech into different languages ​​in real time and with ultra-low latency, and can also add emotional information. This system has high practical applications in various fields such as business, tourism, and education, and helps to effectively overcome language barriers.

[1346] The following describes the processing flow.

[1347] Step 1:

[1348] When a user starts a conversation, their device captures the audio. Specifically, it uses a microphone to pick up the user's voice and converts it to a digital format in real time.

[1349] Step 2:

[1350] The device captures audio data, converts it to binary format, and sends it to the edge computing server using an HTTP request. This request includes the audio data.

[1351] Step 3:

[1352] The edge server receives the audio data and the speech recognition engine begins analysis. Specifically, the process of converting the audio data into text data is executed.

[1353] Step 4:

[1354] The edge server sends the text data generated as a result of speech recognition back to the user terminal as an HTTP response. The user terminal receives this response and extracts the text data.

[1355] Step 5:

[1356] The user's terminal uses the received text data to resend the translation request to the edge computing server. This request includes information about the original text and the target language.

[1357] Step 6:

[1358] The edge server receives the text translation request and uses a high-performance translation engine to translate the text data into the specified target language.

[1359] Step 7:

[1360] The edge server performs sentiment recognition on the translated text data. Specifically, the sentiment engine analyzes the audio data and text to evaluate relevant sentiment information.

[1361] Step 8:

[1362] The edge server adds sentiment information to the translation result and sends it to the user terminal as an HTTP response. The user terminal receives this response and extracts the translated text data and sentiment information.

[1363] Step 9:

[1364] The user's device displays the final translation result and sentiment information to the user. Specifically, it visually displays the translated text and its sentiment on the screen, making it easy for the user to understand.

[1365] (Example 2)

[1366] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1367] Conventional speech recognition systems could convert speech data to text and translate languages, but they could not add emotional information, making it difficult to achieve communication that reflected the user's emotions. Furthermore, they often lacked processing speed and real-time capabilities, limiting their practicality. As a result, they were unable to effectively overcome language barriers in fields such as business, tourism, and education.

[1368] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1369] In this invention, the server includes means for capturing audio data, means for transmitting the audio data to an edge computing server, means for the edge computing server to convert the audio data into text data, means for translating the text data into a specified language, means for adding sentiment information to the translated text data, and means for transmitting the translated text data and sentiment information to a user terminal. This enables the addition of sentiment information in addition to the conversion and translation of audio data into text. As a result, users can use text with sentiment information that has been translated in real time, enabling richer communication. Furthermore, high-speed edge computing improves processing speed and real-time capabilities, increasing its practicality in various fields.

[1370] "Audio data" refers to a digital audio signal acquired using a microphone device.

[1371] An "edge computing server" refers to a server located close to the user's terminal that performs data processing with an emphasis on real-time performance.

[1372] "Text data" refers to string data generated by analyzing audio data.

[1373] "Translation" refers to the process of converting text data expressed in one language into another language.

[1374] "Emotional information" refers to information that indicates a user's emotions (e.g., positive, negative, neutral) as recognized from text data or audio data.

[1375] An "HTTP request" refers to a request made by a client to a server using a protocol for sending data.

[1376] This invention relates to a system that captures audio data, converts it into text data in real time, translates it into a specified language, and adds emotional information. This system operates based on a user terminal and an edge computing server.

[1377] User terminal operation

[1378] First, the user terminal is equipped with a microphone device that captures the user's voice. For example, when the user says "Hello, how are you?", this voice is captured by the terminal as digital audio data. Next, the user terminal converts the captured audio data into binary format and sends it to a specified endpoint on the edge computing server using an HTTP request. An example of such an endpoint is " / transcribe".

[1379] Edge computing server-side operation

[1380] The edge computing server receives audio data sent from the user's terminal. This received audio data is converted into text data using a speech recognition system (the API used as an example provides a speech recognition algorithm). Examples of speech recognition systems that can be used include the Google Speech-to-Text API. The converted text data is sent back to the user's terminal via an HTTP response.

[1381] Next, the user terminal sends the acquired text data back to the edge computing server, requesting a translation into the specified language. This translation process uses a high-performance translation engine, such as the Google Translate API. Once the translation is complete, the server analyzes the translated text data with a sentiment engine (for example, IBM Watson Tone Analyzer) and adds sentiment information. This translated data with added sentiment information is then sent back to the user terminal as an HTTP response.

[1382] Display of data by the user terminal

[1383] Finally, the user terminal displays the received translation data and sentiment information to the user. This allows the user to see their spoken content translated into different languages ​​in real time, with added sentiment information.

[1384] Specific example

[1385] For example, a Japanese-speaking user says the phrase "Hello, how are you?" to the user terminal. This audio data is immediately captured and sent to the edge computing server. The edge computing server uses a speech recognition engine to convert the audio data into text "Hello, how are you?" and sends it back to the user terminal. The user terminal resends this text data to the edge computing server, specifying English as the target language. The edge server translates the text data into English "Hello, how are you?" and uses an emotion engine to recognize the user's emotion (e.g., positive emotion) from the original Japanese audio data. This emotion information is added to the translation result and finally sent back to the user terminal. As a result, the user terminal displays the translated text with emotion information, such as "Hello, how are you? [Positive]".

[1386] Through the process described above, the system of the present invention can translate speech into different languages ​​in real time and with ultra-low latency, and can also add emotional information. This system has high practical applications in various fields such as business, tourism, and education, and helps to effectively overcome language barriers.

[1387] Example of a prompt

[1388] For example, by inputting the following prompt into the generating AI model, you can obtain detailed instructions for performing the aforementioned process:

[1389] Prompt message:

[1390] "Capture Japanese audio data, translate it into English, and display the translation result with added sentiment information."

[1391] This prompt allows the generative AI model to suggest specific technical measures, such as how to use the appropriate API and how to specify the endpoint.

[1392] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1393] Step 1:

[1394] The user starts voice capture.

[1395] Input: User's voice (e.g., "Hello, how are you?")

[1396] Specific action: The user begins speaking into the device's microphone. The microphone captures this audio as a digital signal.

[1397] Output: Digital audio data

[1398] Step 2:

[1399] The device acquires voice data and sends it to the edge server.

[1400] Input: Captured digital audio data

[1401] Specific operation: The terminal converts the captured audio data into binary format and saves it to a temporary file. Then, it sends it to the specified endpoint on the edge server (e.g., " / transcribe") using an HTTP request.

[1402] Output: Audio data sent to the edge server

[1403] Step 3:

[1404] The edge server converts the audio data into text data.

[1405] Input: Binary format audio data sent from the terminal.

[1406] Specific operation: The edge server receives the audio data, analyzes it using a speech recognition system (e.g., an API that provides a speech recognition algorithm), and converts it into text data.

[1407] Output: Converted text data (e.g., "Hello, how are you?")

[1408] Step 4:

[1409] The edge server sends text data back to the terminal.

[1410] Input: Text data generated by speech recognition

[1411] Specific operation: The edge server sends the generated text data back to the user terminal as an HTTP response. The terminal saves this text data for use in the next process.

[1412] Output: Text data sent to the user's terminal

[1413] Step 5:

[1414] The device sends text data to the edge server and requests a translation.

[1415] Input: Text data received from the edge server

[1416] Specific action: The user terminal sends a request to the edge server again to translate the saved text data into the target language. English is specified as the target language.

[1417] Output: Text data with translation request sent to the edge server

[1418] Step 6:

[1419] The edge server translates the text data.

[1420] Input: Text data with translation request sent from the user terminal.

[1421] Specific operation: The edge server analyzes the received text data using a translation engine (e.g., an API that provides translation algorithms) and translates it into the specified target language (in this case, English).

[1422] Output: Translated text data (e.g., "Hello, how are you?")

[1423] Step 7:

[1424] The edge server analyzes the sentiment of the translated data and adds emotional information.

[1425] Input: Translated text data

[1426] Specific operation: The edge server analyzes the translated text data using an emotion engine (e.g., an API that provides an emotion analysis algorithm) and recognizes the emotions contained in the text (e.g., positive, negative, neutral). This emotion information is then added to the translated data as a tag.

[1427] Output: Translated text data with added sentiment information (e.g., "Hello, how are you? [Positive]")

[1428] Step 8:

[1429] The edge server sends the translation data and sentiment information back to the terminal.

[1430] Input: Translated text data with added emotional information

[1431] Specific operation: The edge server sends translated data with sentiment information back to the user terminal as an HTTP response. The user terminal receives this data.

[1432] Output: Translated data with sentiment information sent to the user's terminal.

[1433] Step 9:

[1434] The device displays translation data and sentiment information to the user.

[1435] Input: Translation data with emotional information

[1436] Specific operation: The user terminal displays the translated data with sentiment information received on the user interface. The user can then view the translated text and its sentiment information on the screen.

[1437] Output: Translated text with sentiment information displayed to the user (e.g., "Hello, how are you? [Positive]")

[1438] This process allows users to translate speech into different languages ​​in real time and then use it in a format that includes emotional information.

[1439] (Application Example 2)

[1440] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1441] In modern society, there is an increasing demand for multilingual support and real-time emotion recognition. However, conventional technologies are unable to adequately translate voice data in real time or add emotional information, leading to decreased communication efficiency. Therefore, there is a need for a system that integrates multilingual support and emotion recognition. In particular, in security services, smooth communication and situational awareness at the site are crucial, and technologies that can solve this problem are in demand.

[1442] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1443] In this invention, the server includes means for capturing voice data, means for transmitting the voice data to an edge computing server, means for the edge computing server to convert the voice data into text data, means for translating the text data into a specified language, means for adding sentiment information to the translated text data, and means for transmitting the translated text data and sentiment information to a user terminal. This enables multilingual real-time translation and sentiment recognition.

[1444] "Audio data" refers to data that represents audio in a digital format.

[1445] "Means of capturing" refers to devices or methods that record a user's voice and acquire it as digital data.

[1446] An "edge computing server" is a distributed server that processes data sent from a user's terminal in real time at a location close to the local area.

[1447] "Text data" refers to data that represents audio data in written form.

[1448] "Translation methods" refer to the techniques and methods used to convert text data expressed in the original language into another specified language.

[1449] "Means of adding emotional information" refers to technologies and methods that analyze the emotions contained in text data and add the results to the data as annotations.

[1450] A "user terminal" is a device (e.g., smart glasses, smartphone, etc.) used to capture audio data and communicate with a server.

[1451] This invention relates to a system that translates voice data into multiple languages ​​in real time and adds emotional information to that data. This system aims to improve effective communication and situational awareness, particularly in security services. The following describes embodiments for carrying out this invention using specific examples.

[1452] System Configuration

[1453] This system consists of user terminals (such as smart glasses or smartphones) that capture voice data and edge computing servers that process the data.

[1454] User terminal operation

[1455] 1. Audio Capture

[1456] When a user begins a conversation, the microphone device on the user's device captures the audio. The captured audio data is collected at short intervals, and the digital data is stored on the device.

[1457] 2. Sending to the edge computing server

[1458] The captured audio data is converted to binary format and sent to the edge computing server using an HTTP request.

[1459] Edge computing server operation

[1460] 1. Receiving and transcribing audio data

[1461] The edge computing server analyzes the audio data received from the user's terminal and converts it into text data using speech recognition software (e.g., Google Speech-to-Text API).

[1462] 2. Translation of text data

[1463] The converted text data is translated into the specified language. A high-performance translation engine (e.g., Google Translate API) is used for the translation.

[1464] 3. Adding emotional information

[1465] The translated text data is subjected to an emotion recognition engine, and emotional information (e.g., positive, negative, neutral) is added. An AI model (e.g., a BERT-based emotion analysis model) is used for emotion recognition.

[1466] Return to user terminal

[1467] The translated text data and sentiment information processed on the edge computing server are sent back to the user terminal via an HTTP response, and the user terminal displays it.

[1468] Specific example

[1469] A security guard is seen conversing with a foreign tourist.

[1470] 1. A tourist says in English, "Excuse me, can you help me find the museum?"

[1471] 2. The security guard's smart glasses capture this audio and send the data to an edge computing server.

[1472] 3. The server converts the speech to text and obtains "Excuse me, can you help me find the museum?".

[1473] 4. The text data is translated into Japanese, and the text "Excuse me, could you help me find the museum?" is generated.

[1474] 5. The translated text is tagged with sentiment and sent back to the user's device as "Excuse me, could you help me find the museum? [Positive]".

[1475] 6. The information is ultimately displayed on the security guard's smart glasses, allowing the guard to take appropriate action.

[1476] Example of a prompt

[1477] User's voice: "Excuse me, can you help me find the museum?"

[1478] Audio data capture:

[1479] Sending voice data to the edge server

[1480] Speech recognition and text conversion: "Excuse me, can you help me find the museum?"

[1481] Text translation: "Excuse me, could you help me find the museum?"

[1482] Emotional perception: "Positive"

[1483] Final message: "Excuse me, could you help me find the museum? [Positive]"

[1484] This enables the system to integrate multilingual support and emotion recognition, providing real-time feedback to users. Furthermore, it can support smooth communication and rapid decision-making in security services.

[1485] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1486] Step 1:

[1487] Audio Capture

[1488] The user's device (e.g., smart glasses, smartphone) captures the user's voice. When the user starts speaking, the device's microphone collects the voice and stores it as digital data. The input is voice data, and the output is digital voice data.

[1489] Step 2:

[1490] Sending digital audio data to an edge computing server

[1491] The terminal converts the captured digital audio data into binary format and sends it to the edge computing server using an HTTP request. The input is digital audio data, and the output is binary audio data sent as an HTTP request.

[1492] Step 3:

[1493] Receiving and transcribing audio data by the server.

[1494] The server receives binary audio data sent from the terminal. The received audio data is converted into text data using speech recognition software (e.g., Google Speech-to-Text API). The input is binary audio data, and the output is text data.

[1495] Step 4:

[1496] Text data translation

[1497] The server translates the converted text data into the specified language. A high-performance translation engine (e.g., Google Translate API) is used for the translation. The input is text data, and the output is translated text data in the target language.

[1498] Step 5:

[1499] Adding emotional information

[1500] The server analyzes the translated text data using an emotion recognition engine (e.g., a BERT-based emotion analysis model) and adds emotion information. The input is the translated text data, and the output is the text data with emotion information added.

[1501] Step 6:

[1502] Sending translated text data and sentiment information back to the device

[1503] The server sends translated text data with added sentiment information back to the user's terminal using an HTTP response. The input is the text data with added sentiment information, and the output is the text data sent as an HTTP response.

[1504] Step 7:

[1505] Display of results on the user terminal

[1506] The user terminal receives translated text data with sentiment information sent back from the server and displays it to the user. The input is text data with sentiment information added, and the output is the translated result displayed to the user.

[1507] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1508] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1509] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1510] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1511] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1512] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1513] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1514] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1515] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1516] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1517] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1518] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1519] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1520] 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.

[1521] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1522] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1523] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1524] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1525] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1526] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1527] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[1528] The following is further disclosed regarding the embodiments described above.

[1529] (Claim 1)

[1530] A means of capturing audio data,

[1531] Means for transmitting the aforementioned audio data to an edge computing server,

[1532] The edge computing server includes means for converting the audio data into text data,

[1533] A means for translating the aforementioned text data into a specified language,

[1534] Means for transmitting the translated text data to the user terminal,

[1535] A system that includes this.

[1536] (Claim 2)

[1537] The system according to claim 1, wherein the edge computing server performs speech recognition processing.

[1538] (Claim 3)

[1539] The system according to claim 1, wherein the user terminal converts the audio data into binary format.

[1540] "Example 1"

[1541] (Claim 1)

[1542] A means of capturing audio data,

[1543] Means for transmitting the aforementioned audio data to an edge computing server,

[1544] The edge computing server includes means for converting the audio data into text data,

[1545] A means for translating the aforementioned text data into a specified language,

[1546] Means for transmitting the translated text data to the user terminal,

[1547] Means for capturing and transmitting audio data at short intervals in order to maintain real-time performance,

[1548] Means for sending and receiving data between the edge computing server and the user terminal using HTTP requests and responses,

[1549] Means for displaying the aforementioned text data on the screen of the user terminal,

[1550] A system that includes this.

[1551] (Claim 2)

[1552] The system according to claim 1, wherein the edge computing server performs speech recognition processing and translation processing.

[1553] (Claim 3)

[1554] The system according to claim 1, which includes a process in which the user terminal converts voice data into binary format and sends it to an edge computing server using an HTTP request.

[1555] "Application Example 1"

[1556] (Claim 1)

[1557] A means of capturing audio data,

[1558] Means for transmitting the aforementioned audio data to an edge computing server,

[1559] The edge computing server includes means for converting the audio data into text data,

[1560] A means for translating the aforementioned text data into a specified language,

[1561] A means for transmitting the translated text data to a user terminal and for the user terminal to display the translated text data,

[1562] The user terminal provides a means for placing an order based on voice input,

[1563] A means for processing the details of the aforementioned order on an edge computing server and notifying the delivery service,

[1564] A system that includes this.

[1565] (Claim 2)

[1566] The system according to claim 1, wherein the edge computing server performs speech recognition processing and converts the order details into text data.

[1567] (Claim 3)

[1568] The system according to claim 1, wherein the user terminal converts the voice data into binary format and sends it to the edge computing server as an HTTP request.

[1569] "Example 2 of combining an emotion engine"

[1570] (Claim 1)

[1571] A means of capturing audio data,

[1572] Means for transmitting the aforementioned audio data to an edge computing server,

[1573] The edge computing server includes means for converting the audio data into text data,

[1574] A means for translating the aforementioned text data into a specified language,

[1575] A means for adding emotional information to the translated text data,

[1576] Means for transmitting the translated text data and sentiment information to the user terminal,

[1577] A system that includes this.

[1578] (Claim 2)

[1579] The system according to claim 1, wherein the edge computing server performs speech recognition processing and emotion analysis processing.

[1580] (Claim 3)

[1581] The system according to claim 1, wherein the user terminal converts the audio data into binary format and transmits it using an HTTP request.

[1582] "Application example 2 when combining with an emotional engine"

[1583] (Claim 1)

[1584] A means of capturing audio data,

[1585] Means for transmitting the aforementioned audio data to an edge computing server,

[1586] The edge computing server includes means for converting the audio data into text data,

[1587] A means for translating the aforementioned text data into a specified language,

[1588] A means of adding emotional information to translated text data,

[1589] A means for transmitting the translated text data and sentiment information to the user terminal,

[1590] A system that includes this.

[1591] (Claim 2)

[1592] The system according to claim 1, wherein the edge computing server performs speech recognition processing.

[1593] (Claim 3)

[1594] The system according to claim 1, wherein the user terminal converts the audio data into binary format. [Explanation of Symbols]

[1595] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of capturing audio data, Means for transmitting the aforementioned audio data to an edge computing server, The edge computing server includes means for converting the audio data into text data, A means for translating the aforementioned text data into a specified language, Means for transmitting the translated text data to the user terminal, A system that includes this.

2. The system according to claim 1, wherein the edge computing server performs speech recognition processing.

3. The system according to claim 1, wherein the user terminal converts the audio data into binary format.

Images