System

Abstract

A system is provided.SOLUTION: A system comprising: means for obtaining voice input data; means for transmitting the voice input data to a server; means for translating the voice input data into text data at the server; means for translating the translated text data into another designated language; means for converting the translated text data into voice data; means for transmitting the voice data and the translated text data to a terminal; and means for playing back the received voice data and displaying the translated text data.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] With the increase in global communication in modern times, there is a growing need for systems that allow people who speak different languages ​​to communicate easily. There is also a demand for real-time language translation at meetings, events, and face-to-face customer service, and the widespread use of such systems would help break down language barriers. However, existing language translation systems still lack many areas, such as real-time performance, multilingual support, and translation accuracy, and there is a demand for more convenient systems. The present invention aims to solve these problems and provide a system that enables accurate real-time multilingual translation and speech-to-text output. [Means for solving the problem]

[0005] The system of the present invention first includes a means for acquiring voice input data. It also includes a means for transmitting the voice data to a server and a means for converting the voice input data into text data on the server. It also includes a means for translating the converted text data into another specified language and a means for converting the translated text data into voice data. It also includes a means for transmitting the generated voice data and translated text data to a terminal, and a means for playing back the received voice data and displaying the translated text data, thereby achieving real-time, multilingual, and highly accurate translation. This system can break down language barriers and provide smooth communication in meetings, events, and face-to-face customer service.

[0006] "Voice input data" is data that expresses user speech information captured by a microphone of a terminal as a digital signal.

[0007] A "means" is a hardware or software component for realizing a specific function or role.

[0008] A "server" is a central computer system for processing data over a network, and is a device that analyzes and translates voice data.

[0009] "Text data" is voice data converted into character information, and is data that is in a readable and writable state.

[0010] A "language translation engine" is software or a system for converting text from one language to another.

[0011] A "speech synthesis engine" is software or a system that converts text data into speech signals and generates artificial speech.

[0012] "Audio data" refers to acoustic information stored in a reproducible format, typically represented as a digital file.

[0013] A "terminal" is an input / output device used by a user to capture audio and display data (e.g., a smartphone, tablet, or PC).

[0014] "Lossless compression" is a compression technique that reduces data size without compromising data quality.

[0015] "Cloud services" are services that utilize remote computer resources and software provided over the Internet. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0024] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0037] The system according to the present invention performs real-time translation and speech-to-text output in multiple languages. This system is implemented as follows.

[0038] First, the device captures the user's voice using a microphone. This voice data is converted into a digital signal and processed in real time. The captured data is then transmitted from the device to a server. This transmission process uses lossless compression technology to optimize data volume.

[0039] Next, the server analyzes the received voice data. It converts the voice data into text data using a voice recognition engine. The voice recognition engine divides the voice signal into frames and analyzes them based on acoustic and language models. For example, the voice data "This is a great project" is converted into text data "This is a great project."

[0040] Next, the generated text data is translated into another language specified by a language translation engine. The server uses the language translation engine via a cloud service to convert the text data into the target language. For example, "This is a wonderful project" is translated into English as "This is a wonderful project."

[0041] The translated text data is then converted into speech data using a speech synthesis engine. The server converts the text into speech, taking into account phonology, morphology, and syntax. The generated speech data is saved as an English audio file. This generates the text "This is a wonderful project" as English speech data.

[0042] The generated voice data and translated text data are then sent to the device, where the voice data is played back through the speaker and the text data is displayed on the screen, allowing the user to understand and respond to speech in another language in real time.

[0043] For example, if a user says "This is a wonderful project" in a meeting, the device's microphone captures this speech and immediately sends it to the server. The server converts this into text data, "This is a wonderful project," and then translates it into English as "This is a wonderful project." This is then converted into audio data and sent to the device. The device receives it, plays it as English speech from the speaker, and displays "This is a wonderful project" on the screen.

[0044] Through these processes, the system provides highly accurate multilingual translation in real time, facilitating meetings and face-to-face communication.

[0045] The processing flow will be explained below.

[0046] Step 1:

[0047] The device captures the user's voice using a microphone, and the voice signal is converted into digital data and processed in real time.

[0048] Step 2:

[0049] The device sends the captured audio data to the server, where lossless compression is used to optimize data size.

[0050] Step 3:

[0051] The server analyzes the received voice data using a voice recognition engine and converts the voice data into text data. The voice signal is divided into frames and analyzed based on acoustic and language models.

[0052] Step 4:

[0053] The server translates the generated text data into another language specified by the language translation engine. The translation engine is used via a cloud service to convert the text data into the target language.

[0054] Step 5:

[0055] The server converts the translated text data into speech data using a speech synthesis engine, taking into account phonology, morphology, and syntax, and stores the generated speech data.

[0056] Step 6:

[0057] The server transmits the generated voice data and the translated text data to the terminal.

[0058] Step 7:

[0059] The device receives the voice data, plays it back through the speaker, and displays the translated text data on the screen, allowing users to understand and respond to what is being said in another language in real time.

[0060] Example 1

[0061] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0062] In today's globalized society, there is a demand for real-time translation systems between multiple languages. However, existing translation systems have problems with speech recognition accuracy and translation speed and accuracy, making it difficult to communicate smoothly in real time. This poses a challenge in the smooth conduct of multilingual meetings and face-to-face communication.

[0063] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0064] In this invention, the server includes means for optimizing voice input data using lossless compression technology and transmitting the optimized data to the server, means for converting the voice input data into text data using a physical model in the server, and means for translating the converted text data into another specified language via a cloud service, thereby enabling highly accurate multilingual translation in real time.

[0065] "Voice input data" refers to data obtained by converting the user's speech into digital form.

[0066] "Lossless compression technology" is a compression method that reduces the data volume but allows the original data to be completely restored.

[0067] A "server" is a computer system that processes, stores, and provides data over a network.

[0068] A "voice recognition engine" is software or hardware for converting voice data into text data.

[0069] A "physical model" is an algorithm for analyzing a speech signal, including an acoustic model and a language model.

[0070] "Cloud services" is a general term for various computing resources and services provided via the Internet.

[0071] A "language translation engine" is software or hardware for translating text data in one language into another language.

[0072] A "speech synthesis engine" is software or hardware that converts text data into human speech.

[0073] A "terminal" is an electronic device that can be directly operated by a user and has an interface for inputting and outputting voice.

[0074] A "digital signal" is a signal that is produced by sampling continuous analog data at regular intervals and converting it into digital form.

[0075] The system according to the present invention performs real-time translation and speech-to-text output in multiple languages. This system is implemented as follows.

[0076] First, the device captures the user's voice using a microphone. This capture process uses the device's built-in microphone. For example, say someone says "This is a great project" in a meeting. This voice is picked up by the microphone as an analog signal and then immediately converted into a digital signal.

[0077] The device then converts the captured audio data into a digital signal and optimizes the data size using lossless compression techniques, such as FLAC (Free Lossless Audio Codec), and transmits the compressed audio data to the server via a secure protocol (e.g., HTTPS).

[0078] The server launches a speech recognition engine to analyze the received voice data. Specifically, a common cloud-based voice recognition service is used (e.g., Google® Cloud Speech-to-Text or Microsoft® Azure® Cognitive Services). The server divides the voice signal into frames and analyzes them using an acoustic model and a language model. For example, the voice data "This is a great project" is converted into text data "This is a great project."

[0079] The server sends the generated text data to a language translation engine. Specifically, a common cloud-based translation service is used (e.g., Google Translate or DeepL API). The server converts the text data into the specified target language. For example, "This is a wonderful project" is translated into English as "This is a wonderful project."

[0080] Next, the server uses a speech synthesis engine to convert the translated text data into speech data. Specifically, a common cloud-based speech synthesis service (e.g., Amazon Polly or Google Text-to-Speech) is used. The server converts the text data into English speech, taking into account phonology, morphology, and syntax. As a result, the text "This is a wonderful project" is generated as English speech data.

[0081] The server then sends the generated voice data and translated text data to the device. The device receives this, and the voice data is played from the speaker and the text data is displayed on the screen. Specifically, for example, an English voice saying "This is a wonderful project" is played from the speaker, and at the same time, the text "This is a wonderful project" is displayed on the screen.

[0082] Specific examples

[0083] During a meeting, if a user says "This is a wonderful project," the device's microphone captures this audio, compresses it in FLAC format, and sends it to the server. The server then uses a common cloud service (e.g., Google Cloud Speech-to-Text) to convert the audio data into text data saying "This is a wonderful project." The server then translates this text data into English using the Google Translate API, obtaining "This is a wonderful project." Next, Amazon Polly converts this text into audio data and sends it back to the device. Finally, the device plays the English audio "This is a wonderful project" from the speaker and displays the translated text on the screen.

[0084] Prompt Sentence Examples

[0085] Below are some example prompts to input to a generative AI model:

[0086] "Please explain the detailed processing flow of a real-time system that takes Japanese speech data as input, translates that speech into English, and plays it back."

[0087] "Please explain in detail, with concrete examples, the step-by-step operation of a system that performs multilingual translation and speech-to-text output."

[0088] Based on the above, this system provides highly accurate multilingual translation in real time, enabling smooth meetings and face-to-face communication.

[0089] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0090] Step 1:

[0091] The user provides speech input. The device captures the speech data through the microphone. The input is what the user says (e.g., "This is a great project"). In operation, the device converts the analog signal to a digital signal. The output is the digital version of the speech data.

[0092] Step 2:

[0093] The device compresses the generated digital audio data using a lossless compression technique (e.g., FLAC). The input is the digital audio data obtained in step 1. The operation is to apply a compression algorithm such as FLAC. The output is the compressed audio data.

[0094] Step 3:

[0095] The device sends the compressed audio data to the server using a secure protocol (e.g., HTTPS). The input is the compressed audio data obtained in step 2. The operation is to send the data to the server using network communication. The output is the compressed audio data received by the server.

[0096] Step 4:

[0097] The server decompresses the received compressed audio data and converts it into text data using a speech recognition engine. The input is the compressed audio data. The operation is to apply a speech recognition algorithm (e.g., a cloud service-based speech recognition engine) to analyze the audio signal. The output is to generate text data (e.g., "This is a great project").

[0098] Step 5:

[0099] The server sends the generated text data to a language translation engine to translate it into the specified target language. The input is the text data obtained in step 4. The action is to apply a language translation algorithm from a cloud service (e.g., Google Translate). The output is the translated text data (e.g., "This is a wonderful project").

[0100] Step 6:

[0101] The server sends the translated text data to a speech synthesis engine and converts it into audio data. The input is the translated text data obtained in step 5. The action is to apply a speech synthesis algorithm (e.g., Amazon Polly). The output is audio data (e.g., an English audio file).

[0102] Step 7:

[0103] The server sends the generated voice data and translated text data to the terminal. The input is the voice data and translated text data obtained in step 6. The operation is to send the data to the terminal using network communication. The output is the voice data and translated text data received by the terminal.

[0104] Step 8:

[0105] The device plays the received audio data from the speaker and displays the translated text data on the screen. The input is the audio data and translated text data obtained in step 7. The operation is to play the audio data from the speaker and display the text data on the display. The output is the audio and text that the user can hear and hear.

[0106] (Application example 1)

[0107] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0108] In manufacturing, service, and other industries where multilingual support is required, workers often communicate in different languages. In these cases, accurate translation in real time is necessary, and misunderstandings and delays in instructions due to language barriers can adversely affect productivity. This problem is particularly pronounced in situations where machine control commands must be translated into multiple languages ​​in real time. Therefore, an objective of this invention is to provide a system that enables real-time translation in multiple languages ​​and instantaneous machine control on-site.

[0109] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0110] In this invention, the server includes means for converting voice input data into text data, means for translating the converted text data into another specified language, means for converting the translated text data into voice data, and means for generating machine control commands based on the text data and controlling the corresponding machine, thereby enabling smooth communication between workers who speak different languages ​​and enabling the machine control commands to be executed in real time.

[0111] "Voice input data" refers to data obtained in digital form from the voice spoken by the user.

[0112] A "server" is a computer system that processes voice input data and provides functions such as conversion to text data, translation, and voice synthesis.

[0113] "Text data" refers to data obtained by converting voice input data into character information.

[0114] "Another language" is a language different from the original language of the speech input data.

[0115] "Translation" is the process of converting text data from one language into another.

[0116] "Audio data" refers to an audio file generated by voice synthesis of text data.

[0117] A "terminal" is a device that receives, plays back, and displays voice data and translated text data.

[0118] A "machine control command" is a command generated based on text data for operating a corresponding mechanical device.

[0119] "Machinery" refers to automated machinery and equipment used in factories, work sites, etc.

[0120] A system embodying this invention is configured as follows: First, the user's speech is captured by a microphone installed in the terminal. This speech input data is converted into a digital format and transmitted from the terminal to a server. At this time, the speech input data is losslessly compressed.

[0121] The server converts the received voice input data into text data using a speech recognition engine, which uses cloud services such as the Google Cloud Speech-to-Text API, and then translates the text data into another specified language using the Google Cloud Translate API.

[0122] The translated text data is converted to audio data using the Google Cloud Text-to-Speech API. The generated audio data is saved on the server as an audio file in the target language. This audio data and the translated text data are then sent back to the device.

[0123] The terminal plays the received voice data through a speaker and displays the translated text data on a display. It also generates machine control commands based on the text data and controls the corresponding machinery. This enables smooth communication between workers who speak different languages ​​and enables machine control commands to be executed in real time.

[0124] As a specific example, if factory worker A says "Please start line 1" in Japanese, this voice is captured by the microphone on the terminal and sent digitally to the server. The server converts the voice into text data and translates "Please start line 1" into English as "Please start line 1." This is converted back into voice data and sent back to the terminal. The terminal's speaker plays "Please start line 1" and the display shows the English text. At the same time, the corresponding machine device receives a command to start line 1.

[0125] Examples of prompts powered by generative AI models include:

[0126] "You are a smart solutions engineer developing a robot assistant for use in factories. The robot assistant needs to respond to multilingual instructions in real time. You will use Google Cloud APIs for implementation. Please generate a specific Python script based on the following requirements:

[0127] 1. Record the voice instructions of the worker.

[0128] 2. Convert the recorded voice into text.

[0129] 3. Translate the text into English.

[0130] 4. Save the translated text as an audio file.

[0131] 5. Play the generated audio file and give instructions to the robot.

[0132] This enables the system to realize real-time multilingual translation and machine control in different language environments, significantly improving work efficiency within factories.

[0133] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0134] Step 1:

[0135] The user speaks a voice command. This voice command is captured by the microphone on the device and obtained as digital voice input data. The input is the user's voice, and the output is digital voice input data.

[0136] Step 2:

[0137] The device losslessly compresses the captured audio input data, then transmits the compressed data to the server. The input is the audio input data in digital form, and the output is the losslessly compressed audio input data.

[0138] Step 3:

[0139] The server converts the received voice input data into text data using the Google Cloud Speech-to-Text API. During this process, the voice signal is divided into frames and analyzed based on acoustic and language models. The input is losslessly compressed voice input data, and the output is converted text data.

[0140] Step 4:

[0141] The server translates the converted text data into another specified language using the Google Cloud Translate API. This process involves language identification and translation of the text. The input is the text data, and the output is the translated text data.

[0142] Step 5:

[0143] The server converts the translated text data into audio data using the Google Cloud Text-to-Speech API. In this step, the text is synthesized into audio taking into account phonology, morphology, and syntax. The input is the translated text data, and the output is audio data.

[0144] Step 6:

[0145] The server sends the generated voice data and translated text data to the terminal. The input is the voice data and translated text data, and the output is the voice data and translated text data sent to the terminal.

[0146] Step 7:

[0147] The terminal plays the received voice data from a speaker, displays the translated text data on a display, and generates machine control commands based on the text data to control the corresponding machine. The inputs are the received voice data and the translated text data, and the outputs are voice playback, text display, and machine control execution.

[0148] This process flow allows smooth communication between workers who speak different languages ​​and enables machine control commands to be executed in real time.

[0149] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0150] The system of this invention performs real-time translation and speech / text output in multiple languages, and by combining it with an emotion engine that recognizes the user's emotions, it realizes more natural and rich communication that includes emotions. This system is implemented as follows.

[0151] First, the device captures the user's voice using a microphone. The voice signal is converted into digital data and processed in real time. The captured data is then sent from the device to a server. Lossless compression technology is used to optimize data volume during the transmission process.

[0152] The server then analyzes the received voice data using a speech recognition engine and converts it into text data. The voice signal is divided into frames and analyzed based on acoustic and language models to generate accurate text data.

[0153] The server translates the generated text data into another language specified by the language translation engine. The server uses the translation engine via a cloud service to convert the text data into the target language. For example, "This is a wonderful project" is translated into English as "This is a wonderful project."

[0154] The server also includes an emotion engine, which analyzes the user's emotional state from the voice input data. The emotion engine analyzes the tone, pitch, speed, rhythm, etc. of the voice to recognize the user's emotions, such as happiness, anger, or sadness. The recognized emotional information is reflected in the generated translated text and voice data.

[0155] The server then converts the translated text data into speech data using a speech synthesis engine. The text data is converted into speech while taking into account phonology, morphology, and syntax, and the generated speech data is saved. During this process, the emotion engine also applies the emotions recognized by the server, ensuring that the translated speech data reflects the emotions.

[0156] The server then sends the generated voice data and translated text data to the device. The voice data received by the device is played back through the speaker, and the text data is displayed on the screen. Emotional information is also reflected on the device's display. This allows the user to understand and respond to content spoken in another language, along with emotional information, in real time.

[0157] As a concrete example, if a user says with joy, "This is a wonderful project," in a meeting, the device's microphone captures this voice and emotion and sends it to the server. The server converts this into text data, "This is a wonderful project," and then translates it into English as "This is a wonderful project." Emotional information is also added to this translated text. This is then converted into audio data and sent to the device as audio reflecting the emotion of joy. The device receives this and plays the joyful English voice saying "This is a wonderful project" from its speaker, while the same content along with the emotional information is displayed on the screen.

[0158] Through these processes, the system provides highly accurate multilingual translation in real time, facilitating meetings and face-to-face communication while incorporating emotions.

[0159] The processing flow will be explained below.

[0160] Step 1:

[0161] The device captures the user's voice using a microphone. The voice signal is converted into digital data and processed in real time. The voice data includes the user's voice tone and pitch.

[0162] Step 2:

[0163] The device sends the captured audio data to the server, where lossless compression is used to optimize data size.

[0164] Step 3:

[0165] The server analyzes the received voice data using a voice recognition engine and converts the voice data into text data. The voice signal is divided into frames and analyzed based on acoustic and language models to generate accurate text data.

[0166] Step 4:

[0167] The server analyzes the voice input data using an emotion engine to extract the user's emotional state. This analysis uses the tone, pitch, speed, and rhythm of the voice. For example, if the voice tone is high and fast, it detects the emotion of joy.

[0168] Step 5:

[0169] The server translates the generated text data into another language specified by the language translation engine. The translation engine is used via a cloud service to convert the text data into the target language. For example, "This is a wonderful project" is translated into English as "This is a wonderful project."

[0170] Step 6:

[0171] The server adds the extracted emotion information to the translated text data, so that the emotion is reflected in the translated text.

[0172] Step 7:

[0173] The server converts the translated text data into speech data using a speech synthesis engine. The text data is converted into speech taking into account phonology, morphology, and syntax, and emotional information from the emotion engine is also reflected in the speech data. For example, the text "This is a wonderful project" is converted into speech with a joyful tone.

[0174] Step 8:

[0175] The server transmits the generated voice data and the translated text data to the terminal.

[0176] Step 9:

[0177] The device plays the received voice data through the speaker and displays the translated text data on the screen. Furthermore, emotional information is reflected in the text data, so the user can check the emotional content of the voice and text in real time.

[0178] For example, if a user says with joy, "This is a wonderful project," in a meeting, the device's microphone captures this voice and emotion and sends it to the server. The server converts this into text data, "This is a wonderful project," and then translates it into English as "This is a wonderful project." Emotional information is also added to this translated text. This is then converted into audio data and sent to the device as audio reflecting the emotion of joy. The device receives this and plays the joyful English voice saying "This is a wonderful project" from its speaker, while the same content along with the emotional information is displayed on the screen.

[0179] Through these steps, the system provides highly accurate multilingual translation in real time, facilitating meetings and face-to-face communication with full emotional impact.

[0180] Example 2

[0181] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0182] Translation systems that facilitate smooth communication between multiple languages ​​require real-time performance and the ability to accurately reflect emotions. Conventional translation systems face the challenge of simultaneously achieving both real-time performance and the ability to accurately reflect emotions. Furthermore, there are limitations to methods for efficiently transmitting speech data while maintaining its quality, and degradation can occur, particularly during data compression and decompression.

[0183] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0184] In this invention, the server includes means for transmitting voice input data to the processing device, means for converting the voice input data into text data in the processing device, means for translating the converted text data into another specified language, means for converting the translated text data into voice data, means for transmitting the voice data and the translated text data to the terminal, and means for analyzing emotional information and reflecting it in the corresponding data. This not only enables accurate multilingual translation in real time while preventing degradation of the voice data, but also enables natural communication that reflects the user's emotions.

[0185] "Voice input data" is data that represents in digital form the voice uttered by the user.

[0186] A "processing device" is a device that has the functionality to analyze, convert, and communicate voice data. It is synonymous with a server.

[0187] "Text data" is character information data generated by voice recognition.

[0188] "Translation" is the process of converting text data in one language into text data in another language.

[0189] "Audio data" is digital data in which the contents of text data are synthesized as audio.

[0190] "Lossless compression" is a technology that reduces data size without compromising data quality.

[0191] "Remote services" are services provided via the Internet using cloud computing technology.

[0192] A "voice recognition engine" is software or hardware that analyzes voice data and converts it into corresponding text data.

[0193] A "language translation engine" is software or hardware for translating text data into another language.

[0194] "Emotion information" is information that indicates the emotional state of the user analyzed from the voice data.

[0195] The system of this invention performs real-time translation and voice / text output in multiple languages, and by combining it with an emotion engine that recognizes the user's emotions, it achieves more natural and rich communication that includes emotions.

[0196] First, the device captures the user's voice using a microphone. The captured voice signal is converted from analog to digital and then transmitted to a processing unit using a lossless compression technique to optimize data volume. This compression technique uses a common lossless compression technique.

[0197] The server then analyzes the received voice data using a speech recognition engine. The voice signal is divided into frames and converted into text data based on acoustic and language models. This process can be performed using a cloud-based speech recognition service, for example.

[0198] The server translates the generated text data into another specified language using a language translation engine provided via a cloud service, for example, translating the text data "This is a wonderful project" into "This is a wonderful project."

[0199] The server is also equipped with an emotion engine that analyzes the user's emotional state from the voice input data. Based on the tone, pitch, speed, and rhythm of the voice, the engine recognizes the user's emotions, such as happiness, anger, and sadness. The analyzed emotional information is reflected in the generated text data and voice data.

[0200] Next, the server converts the translated text data into speech data using a speech synthesis engine. The speech synthesis engine converts the text data into speech while taking into account phonology, morphology, and syntax. During this process, the emotion engine also applies the emotions recognized and reflects them in the translated speech data. For example, a general-purpose speech synthesis service can be used as the speech synthesis engine.

[0201] Finally, the generated voice data and translated text data are sent from the server to the device. The device then plays the received voice data over a speaker and displays the text data on the screen. At this time, emotional information is also reflected in the display. This allows the user to understand the content of speech in another language, along with emotional information, in real time.

[0202] As a concrete example, if a user says with joy, "This is a wonderful project," in a meeting, the device's microphone captures this voice and emotion and sends it to the server. The server converts this into text data, "This is a wonderful project," and then translates it into English as "This is a wonderful project." Emotional information is also added to this translated text. This is then converted into audio data and sent to the device as audio reflecting the emotion of joy. The device receives this and plays the joyful English voice saying "This is a wonderful project" from its speaker, while the same content along with the emotional information is displayed on the screen.

[0203] An example of a prompt might be:

[0204] Please explain the process of capturing a user's speech with a device microphone, sending it to a server for translation, and generating speech and text data with added emotion. Please provide a concrete example, including the translation of "This is a great project" into English.

[0205] This system enables highly accurate real-time multilingual translation and communication that includes emotions, facilitating smoother meetings and face-to-face interactions.

[0206] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0207] The processing flow of this system's program is explained below, divided into processing steps.

[0208] Step 1:

[0209] The user inputs voice. The device's built-in microphone captures what the user is saying. The input voice signal enters the microphone as an analog signal and is then converted into a digital signal. The converted digital voice data proceeds to the next step.

[0210] Step 2:

[0211] The device sends audio data to the server. The digital audio data is compressed in a lossless compression format such as FLAC and then transmitted to the server over the network. The input is digital audio data, and the output is compressed audio data.

[0212] Step 3:

[0213] The server decompresses the received audio data and analyzes it using a speech recognition engine. The FLAC format compressed data is decompressed, and then the speech recognition engine is used to convert the audio data into text data. This engine divides the audio signal into frames, each of which is analyzed using a phonological model and a language model. The input is compressed audio data, and the output is text data.

[0214] Step 4:

[0215] The server translates the converted text data. The text data is translated into the target language using a language translation engine via a cloud service. For example, the Japanese text "This is a wonderful project" is translated into the English text "This is a wonderful project." The input is Japanese text data, and the output is English text data.

[0216] Step 5:

[0217] The server's emotion engine analyzes the user's emotions from the voice input data. It identifies the user's emotional state based on the tone, pitch, speed, rhythm, etc. of the voice. This emotional information is reflected in the translated text data and subsequent voice data. The input is voice data, and the output is emotional information.

[0218] Step 6:

[0219] The server converts the translated text data into speech data using a speech synthesis engine. The translated text is converted into speech while taking into account phonology, morphology, and syntax. At this time, emotional information obtained from the emotion engine is reflected in the speech data. For example, a speech with an emotional expression of joy is generated. The input is the translated text data and emotional information, and the output is speech data that reflects the emotion.

[0220] Step 7:

[0221] The server sends the generated voice data and translated text data to the terminal. The voice data received by the terminal is played back using a speaker, and the translated text data is displayed on the screen. At this time, emotional information is also included in the display. The input is voice data and text data, and the output is voice and text display on the terminal.

[0222] In this way, each process in the system is generated from the user's input speech, and translated speech and text containing emotions are output in real time.

[0223] (Application example 2)

[0224] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0225] In autonomous vehicles, the language barrier makes it difficult for passengers to communicate in multiple languages. Another problem is that natural and rich communication cannot be achieved because passenger emotions are not reflected. Furthermore, it is necessary to appropriately reflect passenger emotions in real-time translation and voice playback.

[0226] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for transmitting voice input data to a data storage device, means for converting the voice input data into text data in the data storage device, and means for recognizing emotional states and reflecting emotional information in the voice data and text data. This makes it possible to translate the multilingual voices of passengers in an autonomous vehicle in real time and output natural voices and text that reflect emotions.

[0227] "Voice input data" refers to data obtained by converting the user's speech into digital form.

[0228] A "data storage device" is a device that temporarily stores voice input data and text data and performs the necessary processing.

[0229] "Text data" is data that is generated by analyzing voice input data and expressing it as a character string.

[0230] "Translation" is the process of converting text data expressed in one language into text data expressed in another language.

[0231] "Audio data" refers to data obtained by converting text data back into audio format.

[0232] The "emotional state" is a state in which the user is feeling some emotion, which is analyzed from the user's voice.

[0233] "Emotional information" is data that reflects the emotional state of the user and is information that is applied to translated text data or voice data.

[0234] A "receiving terminal" is a device that receives voice data and translated text data and plays or displays them.

[0235] A "voice recognition engine" is software or hardware for converting voice input data into a string of characters.

[0236] A "language translation engine" is software or hardware for translating text data expressed in one language into another language.

[0237] "Information and communication services" refers to infrastructure for utilizing various information processing functions provided on the cloud.

[0238] MODE FOR CARRYING OUT THE INVENTION

[0239] This invention is a multilingual real-time speech translation and emotion recognition system for passengers in autonomous vehicles. To realize this system, the following specific configuration is required.

[0240] First, the user's voice data is captured by a microphone installed in the vehicle. The captured voice data is converted into digital data and then compressed using lossless compression technology (e.g., Zlib). The compressed voice data is then temporarily stored in a data storage device in the vehicle.

[0241] The server then takes this voice data and converts it into text using a speech recognition engine (e.g., Google Speech-to-Text). The converted text data is then sent to a language translation engine (e.g., Google Cloud Translate) for translation into the specified target language. This converts the text data expressed in the original language into another language.

[0242] Furthermore, the server uses an emotion recognition engine (e.g., IBM Watson® Tone Analyzer) to analyze the user's emotional state from the voice data, and the analyzed emotional information is reflected in the translated text data and the voice data that is later synthesized.

[0243] Next, based on the translated text data and emotion information, a speech synthesis engine (e.g., Amazon Polly) is used to generate voice data that reflects the user's emotions.

[0244] The generated voice data and translated text data are then transmitted to a receiving terminal inside the vehicle, which then displays the text data and emotion information on a display and simultaneously plays the voice data reflecting the emotion from a speaker.

[0245] Examples:

[0246] For example, if a user says in Japanese, "Good morning, please head to the airport," this voice is captured and sent to a data storage device. The server converts this voice into text data and translates it into English as "Good morning, please head to the airport." In addition, the emotion of the user's voice is recognized as "polite and calm," and a voice that reflects this emotion is generated. Finally, the in-car speaker plays "Good morning, please head to the airport" in a polite tone, and the same message is displayed on the in-car display.

[0247] Example prompt sentence:

[0248] If the user says "This is a wonderful project!" in a joyful voice, translate that into English and generate a joyful voice saying "This is a wonderful project!"

[0249] In this way, the system of the present invention realizes natural and rich multilingual communication within an autonomous vehicle.

[0250] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0251] Step 1:

[0252] The user speaks voice commands into a microphone installed inside the vehicle. The user's voice data is captured as input. This voice data is converted from an analog signal to a digital signal, making it easier to process (e.g., PCM format).

[0253] Step 2:

[0254] The audio data captured by the terminal is sent to the data storage device using lossless compression technology (e.g., Zlib). The input is digital audio data, and the output is compressed audio data. This improves data transmission efficiency.

[0255] Step 3:

[0256] The server decompresses the compressed audio data it receives and converts it into text data using a speech recognition engine (e.g., Google Speech-to-Text). The input is compressed audio data and the output is text data. The speech recognition engine analyzes the audio signal frame by frame based on an acoustic model and a language model to generate a string of characters.

[0257] Step 4:

[0258] The server sends the generated text data to a language translation engine (e.g., Google Cloud Translate) to translate it into the specified target language. The input is text data, and the output is text data in the target language. The translation engine uses a neural network to perform highly accurate translation.

[0259] Step 5:

[0260] The server uses an emotion recognition engine (e.g., IBM Watson Tone Analyzer) to analyze the emotional state from the voice data. The input is the original voice data, and the output is emotional information (e.g., joy, anger, surprise, etc.). The emotion engine analyzes the tone, pitch, speed, etc. of the voice to determine the emotional state.

[0261] Step 6:

[0262] The server generates speech data using a speech synthesis engine (e.g., Amazon Polly) based on the translated text data and emotional information. The input is the translated text data and emotional information, and the output is speech data that reflects the emotion. The speech synthesis engine generates natural-sounding speech while taking into account phonology, morphology, and syntax.

[0263] Step 7:

[0264] The server sends the generated voice data and translated text data to the terminal. The input is the voice data and translated text data, and the output is the data received by the terminal. An appropriate protocol (e.g. HTTP / 2) is used for data transmission.

[0265] Step 8:

[0266] The terminal receives voice data and plays it back through the vehicle's speakers, and displays the translated text data and emotional information on the display. The input is the received data, and the output is audiovisual information presentation. This enables users to communicate in multiple languages ​​in real time.

[0267] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0268] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0269] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0270] [Second embodiment]

[0271] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0272] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0273] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0274] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0275] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0276] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0277] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0278] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0279] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0280] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0281] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0282] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0283] The system according to the present invention performs real-time translation and speech-to-text output in multiple languages. This system is implemented as follows.

[0284] First, the device captures the user's voice using a microphone. This voice data is converted into a digital signal and processed in real time. The captured data is then transmitted from the device to a server. This transmission process uses lossless compression technology to optimize data volume.

[0285] Next, the server analyzes the received voice data. It converts the voice data into text data using a voice recognition engine. The voice recognition engine divides the voice signal into frames and analyzes them based on acoustic and language models. For example, the voice data "This is a great project" is converted into text data "This is a great project."

[0286] Next, the generated text data is translated into another language specified by a language translation engine. The server uses the language translation engine via a cloud service to convert the text data into the target language. For example, "This is a wonderful project" is translated into English as "This is a wonderful project."

[0287] The translated text data is then converted into speech data using a speech synthesis engine. The server converts the text into speech, taking into account phonology, morphology, and syntax. The generated speech data is saved as an English audio file. This generates the text "This is a wonderful project" as English speech data.

[0288] The generated voice data and translated text data are then sent to the device, where the voice data is played back through the speaker and the text data is displayed on the screen, allowing the user to understand and respond to speech in another language in real time.

[0289] For example, if a user says "This is a wonderful project" in a meeting, the device's microphone captures this speech and immediately sends it to the server. The server converts this into text data, "This is a wonderful project," and then translates it into English as "This is a wonderful project." This is then converted into audio data and sent to the device. The device receives it, plays it as English speech from the speaker, and displays "This is a wonderful project" on the screen.

[0290] Through these processes, the system provides highly accurate multilingual translation in real time, facilitating meetings and face-to-face communication.

[0291] The processing flow will be explained below.

[0292] Step 1:

[0293] The device captures the user's voice using a microphone, and the voice signal is converted into digital data and processed in real time.

[0294] Step 2:

[0295] The device sends the captured audio data to the server, where lossless compression is used to optimize data size.

[0296] Step 3:

[0297] The server analyzes the received voice data using a voice recognition engine and converts the voice data into text data. The voice signal is divided into frames and analyzed based on acoustic and language models.

[0298] Step 4:

[0299] The server translates the generated text data into another language specified by the language translation engine. The translation engine is used via a cloud service to convert the text data into the target language.

[0300] Step 5:

[0301] The server converts the translated text data into speech data using a speech synthesis engine, taking into account phonology, morphology, and syntax, and stores the generated speech data.

[0302] Step 6:

[0303] The server transmits the generated voice data and the translated text data to the terminal.

[0304] Step 7:

[0305] The device receives the voice data, plays it back through the speaker, and displays the translated text data on the screen, allowing users to understand and respond to what is being said in another language in real time.

[0306] Example 1

[0307] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0308] In today's globalized society, there is a demand for real-time translation systems between multiple languages. However, existing translation systems have problems with speech recognition accuracy and translation speed and accuracy, making it difficult to communicate smoothly in real time. This poses a challenge in the smooth conduct of multilingual meetings and face-to-face communication.

[0309] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0310] In this invention, the server includes means for optimizing voice input data using lossless compression technology and transmitting the optimized data to the server, means for converting the voice input data into text data using a physical model in the server, and means for translating the converted text data into another specified language via a cloud service, thereby enabling highly accurate multilingual translation in real time.

[0311] "Voice input data" refers to data obtained by converting the user's speech into digital form.

[0312] "Lossless compression technology" is a compression method that reduces the data volume but allows the original data to be completely restored.

[0313] A "server" is a computer system that processes, stores, and provides data over a network.

[0314] A "voice recognition engine" is software or hardware for converting voice data into text data.

[0315] A "physical model" is an algorithm for analyzing a speech signal, including an acoustic model and a language model.

[0316] "Cloud services" is a general term for various computing resources and services provided via the Internet.

[0317] A "language translation engine" is software or hardware for translating text data in one language into another language.

[0318] A "speech synthesis engine" is software or hardware that converts text data into human speech.

[0319] A "terminal" is an electronic device that can be directly operated by a user and has an interface for inputting and outputting voice.

[0320] A "digital signal" is a signal that is produced by sampling continuous analog data at regular intervals and converting it into digital form.

[0321] The system according to the present invention performs real-time translation and speech-to-text output in multiple languages. This system is implemented as follows.

[0322] First, the device captures the user's voice using a microphone. This capture process uses the device's built-in microphone. For example, say someone says "This is a great project" in a meeting. This voice is picked up by the microphone as an analog signal and then immediately converted into a digital signal.

[0323] The device then converts the captured audio data into a digital signal and optimizes the data size using lossless compression techniques, such as FLAC (Free Lossless Audio Codec), and transmits the compressed audio data to the server via a secure protocol (e.g., HTTPS).

[0324] The server launches a speech recognition engine to analyze the received voice data. A typical cloud-based speech recognition service (e.g., Google Cloud Speech-to-Text or Microsoft Azure Cognitive Services) is used. The server divides the voice signal into frames and analyzes them using an acoustic model and a language model. For example, the voice data "This is a great project" is converted into text data "This is a great project."

[0325] The server sends the generated text data to a language translation engine. Specifically, a common cloud-based translation service is used (e.g., Google Translate or DeepL API). The server converts the text data into the specified target language. For example, "This is a wonderful project" is translated into English as "This is a wonderful project."

[0326] Next, the server uses a speech synthesis engine to convert the translated text data into speech data. Specifically, a common cloud-based speech synthesis service (e.g., Amazon Polly or Google Text-to-Speech) is used. The server converts the text data into English speech, taking into account phonology, morphology, and syntax. As a result, the text "This is a wonderful project" is generated as English speech data.

[0327] The server then sends the generated voice data and translated text data to the device. The device receives this, and the voice data is played from the speaker and the text data is displayed on the screen. Specifically, for example, an English voice saying "This is a wonderful project" is played from the speaker, and at the same time, the text "This is a wonderful project" is displayed on the screen.

[0328] Specific examples

[0329] During a meeting, if a user says "This is a wonderful project," the device's microphone captures this audio, compresses it in FLAC format, and sends it to the server. The server then uses a common cloud service (e.g., Google Cloud Speech-to-Text) to convert the audio data into text data saying "This is a wonderful project." The server then translates this text data into English using the Google Translate API, obtaining "This is a wonderful project." Next, Amazon Polly converts this text into audio data and sends it back to the device. Finally, the device plays the English audio "This is a wonderful project" from the speaker and displays the translated text on the screen.

[0330] Prompt Sentence Examples

[0331] Below are some example prompts to input to a generative AI model:

[0332] "Please explain the detailed processing flow of a real-time system that takes Japanese speech data as input, translates that speech into English, and plays it back."

[0333] "Please explain in detail, with concrete examples, the step-by-step operation of a system that performs multilingual translation and speech-to-text output."

[0334] Based on the above, this system provides highly accurate multilingual translation in real time, enabling smooth meetings and face-to-face communication.

[0335] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0336] Step 1:

[0337] The user provides speech input. The device captures the speech data through the microphone. The input is what the user says (e.g., "This is a great project"). In operation, the device converts the analog signal to a digital signal. The output is the digital version of the speech data.

[0338] Step 2:

[0339] The device compresses the generated digital audio data using a lossless compression technique (e.g., FLAC). The input is the digital audio data obtained in step 1. The operation is to apply a compression algorithm such as FLAC. The output is the compressed audio data.

[0340] Step 3:

[0341] The device sends the compressed audio data to the server using a secure protocol (e.g., HTTPS). The input is the compressed audio data obtained in step 2. The operation is to send the data to the server using network communication. The output is the compressed audio data received by the server.

[0342] Step 4:

[0343] The server decompresses the received compressed audio data and converts it into text data using a speech recognition engine. The input is the compressed audio data. The operation is to apply a speech recognition algorithm (e.g., a cloud service-based speech recognition engine) to analyze the audio signal. The output is to generate text data (e.g., "This is a great project").

[0344] Step 5:

[0345] The server sends the generated text data to a language translation engine to translate it into the specified target language. The input is the text data obtained in step 4. The action is to apply a language translation algorithm from a cloud service (e.g., Google Translate). The output is the translated text data (e.g., "This is a wonderful project").

[0346] Step 6:

[0347] The server sends the translated text data to a speech synthesis engine and converts it into audio data. The input is the translated text data obtained in step 5. The action is to apply a speech synthesis algorithm (e.g., Amazon Polly). The output is audio data (e.g., an English audio file).

[0348] Step 7:

[0349] The server sends the generated voice data and translated text data to the terminal. The input is the voice data and translated text data obtained in step 6. The operation is to send the data to the terminal using network communication. The output is the voice data and translated text data received by the terminal.

[0350] Step 8:

[0351] The device plays the received audio data from the speaker and displays the translated text data on the screen. The input is the audio data and translated text data obtained in step 7. The operation is to play the audio data from the speaker and display the text data on the display. The output is the audio and text that the user can hear and hear.

[0352] (Application example 1)

[0353] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0354] In manufacturing, service, and other industries where multilingual support is required, workers often communicate in different languages. In these cases, accurate translation in real time is necessary, and misunderstandings and delays in instructions due to language barriers can adversely affect productivity. This problem is particularly pronounced in situations where machine control commands must be translated into multiple languages ​​in real time. Therefore, an objective of this invention is to provide a system that enables real-time translation in multiple languages ​​and instantaneous machine control on-site.

[0355] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0356] In this invention, the server includes means for converting voice input data into text data, means for translating the converted text data into another specified language, means for converting the translated text data into voice data, and means for generating machine control commands based on the text data and controlling the corresponding machine, thereby enabling smooth communication between workers who speak different languages ​​and enabling the machine control commands to be executed in real time.

[0357] "Voice input data" refers to data obtained in digital form from the voice spoken by the user.

[0358] A "server" is a computer system that processes voice input data and provides functions such as conversion to text data, translation, and voice synthesis.

[0359] "Text data" refers to data obtained by converting voice input data into character information.

[0360] "Another language" is a language different from the original language of the speech input data.

[0361] "Translation" is the process of converting text data from one language into another.

[0362] "Audio data" refers to an audio file generated by voice synthesis of text data.

[0363] A "terminal" is a device that receives, plays back, and displays voice data and translated text data.

[0364] A "machine control command" is a command generated based on text data for operating a corresponding mechanical device.

[0365] "Machinery" refers to automated machinery and equipment used in factories, work sites, etc.

[0366] A system embodying this invention is configured as follows: First, the user's speech is captured by a microphone installed in the terminal. This speech input data is converted into a digital format and transmitted from the terminal to a server. At this time, the speech input data is losslessly compressed.

[0367] The server converts the received voice input data into text data using a speech recognition engine, which uses cloud services such as the Google Cloud Speech-to-Text API, and then translates the text data into another specified language using the Google Cloud Translate API.

[0368] The translated text data is converted to audio data using the Google Cloud Text-to-Speech API. The generated audio data is saved on the server as an audio file in the target language. This audio data and the translated text data are then sent back to the device.

[0369] The terminal plays the received voice data through a speaker and displays the translated text data on a display. It also generates machine control commands based on the text data and controls the corresponding machinery. This enables smooth communication between workers who speak different languages ​​and enables machine control commands to be executed in real time.

[0370] As a specific example, if factory worker A says "Please start line 1" in Japanese, this voice is captured by the microphone on the terminal and sent digitally to the server. The server converts the voice into text data and translates "Please start line 1" into English as "Please start line 1." This is converted back into voice data and sent back to the terminal. The terminal's speaker plays "Please start line 1" and the display shows the English text. At the same time, the corresponding machine device receives a command to start line 1.

[0371] Examples of prompts powered by generative AI models include:

[0372] "You are a smart solutions engineer developing a robot assistant for use in factories. The robot assistant needs to respond to multilingual instructions in real time. You will use Google Cloud APIs for implementation. Please generate a specific Python script based on the following requirements:

[0373] 1. Record the voice instructions of the worker.

[0374] 2. Convert the recorded voice into text.

[0375] 3. Translate the text into English.

[0376] 4. Save the translated text as an audio file.

[0377] 5. Play the generated audio file and give instructions to the robot.

[0378] This enables the system to realize real-time multilingual translation and machine control in different language environments, significantly improving work efficiency within factories.

[0379] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0380] Step 1:

[0381] The user speaks a voice command. This voice command is captured by the microphone on the device and obtained as digital voice input data. The input is the user's voice, and the output is digital voice input data.

[0382] Step 2:

[0383] The device losslessly compresses the captured audio input data, then transmits the compressed data to the server. The input is the audio input data in digital form, and the output is the losslessly compressed audio input data.

[0384] Step 3:

[0385] The server converts the received voice input data into text data using the Google Cloud Speech-to-Text API. During this process, the voice signal is divided into frames and analyzed based on acoustic and language models. The input is losslessly compressed voice input data, and the output is converted text data.

[0386] Step 4:

[0387] The server translates the converted text data into another specified language using the Google Cloud Translate API. This process involves language identification and translation of the text. The input is the text data, and the output is the translated text data.

[0388] Step 5:

[0389] The server converts the translated text data into audio data using the Google Cloud Text-to-Speech API. In this step, the text is synthesized into audio taking into account phonology, morphology, and syntax. The input is the translated text data, and the output is audio data.

[0390] Step 6:

[0391] The server sends the generated voice data and translated text data to the terminal. The input is the voice data and translated text data, and the output is the voice data and translated text data sent to the terminal.

[0392] Step 7:

[0393] The terminal plays the received voice data from a speaker, displays the translated text data on a display, and generates machine control commands based on the text data to control the corresponding machine. The inputs are the received voice data and the translated text data, and the outputs are voice playback, text display, and machine control execution.

[0394] This process flow allows smooth communication between workers who speak different languages ​​and enables machine control commands to be executed in real time.

[0395] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0396] The system of this invention performs real-time translation and speech / text output in multiple languages, and by combining it with an emotion engine that recognizes the user's emotions, it realizes more natural and rich communication that includes emotions. This system is implemented as follows.

[0397] First, the device captures the user's voice using a microphone. The voice signal is converted into digital data and processed in real time. The captured data is then sent from the device to a server. Lossless compression technology is used to optimize data volume during the transmission process.

[0398] The server then analyzes the received voice data using a speech recognition engine and converts it into text data. The voice signal is divided into frames and analyzed based on acoustic and language models to generate accurate text data.

[0399] The server translates the generated text data into another language specified by the language translation engine. The server uses the translation engine via a cloud service to convert the text data into the target language. For example, "This is a wonderful project" is translated into English as "This is a wonderful project."

[0400] The server also includes an emotion engine, which analyzes the user's emotional state from the voice input data. The emotion engine analyzes the tone, pitch, speed, rhythm, etc. of the voice to recognize the user's emotions, such as happiness, anger, or sadness. The recognized emotional information is reflected in the generated translated text and voice data.

[0401] The server then converts the translated text data into speech data using a speech synthesis engine. The text data is converted into speech while taking into account phonology, morphology, and syntax, and the generated speech data is saved. During this process, the emotion engine also applies the emotions recognized by the server, ensuring that the translated speech data reflects the emotions.

[0402] The server then sends the generated voice data and translated text data to the device. The voice data received by the device is played back through the speaker, and the text data is displayed on the screen. Emotional information is also reflected on the device's display. This allows the user to understand and respond to content spoken in another language, along with emotional information, in real time.

[0403] As a concrete example, if a user says with joy, "This is a wonderful project," in a meeting, the device's microphone captures this voice and emotion and sends it to the server. The server converts this into text data, "This is a wonderful project," and then translates it into English as "This is a wonderful project." Emotional information is also added to this translated text. This is then converted into audio data and sent to the device as audio reflecting the emotion of joy. The device receives this and plays the joyful English voice saying "This is a wonderful project" from its speaker, while the same content along with the emotional information is displayed on the screen.

[0404] Through these processes, the system provides highly accurate multilingual translation in real time, facilitating meetings and face-to-face communication while incorporating emotions.

[0405] The processing flow will be explained below.

[0406] Step 1:

[0407] The device captures the user's voice using a microphone. The voice signal is converted into digital data and processed in real time. The voice data includes the user's voice tone and pitch.

[0408] Step 2:

[0409] The device sends the captured audio data to the server, where lossless compression is used to optimize data size.

[0410] Step 3:

[0411] The server analyzes the received voice data using a voice recognition engine and converts the voice data into text data. The voice signal is divided into frames and analyzed based on acoustic and language models to generate accurate text data.

[0412] Step 4:

[0413] The server analyzes the voice input data using an emotion engine to extract the user's emotional state. This analysis uses the tone, pitch, speed, and rhythm of the voice. For example, if the voice tone is high and fast, it detects the emotion of joy.

[0414] Step 5:

[0415] The server translates the generated text data into another language specified by the language translation engine. The translation engine is used via a cloud service to convert the text data into the target language. For example, "This is a wonderful project" is translated into English as "This is a wonderful project."

[0416] Step 6:

[0417] The server adds the extracted emotion information to the translated text data, so that the emotion is reflected in the translated text.

[0418] Step 7:

[0419] The server converts the translated text data into speech data using a speech synthesis engine. The text data is converted into speech taking into account phonology, morphology, and syntax, and emotional information from the emotion engine is also reflected in the speech data. For example, the text "This is a wonderful project" is converted into speech with a joyful tone.

[0420] Step 8:

[0421] The server transmits the generated voice data and the translated text data to the terminal.

[0422] Step 9:

[0423] The device plays the received voice data through the speaker and displays the translated text data on the screen. Furthermore, emotional information is reflected in the text data, so the user can check the emotional content of the voice and text in real time.

[0424] For example, if a user says with joy, "This is a wonderful project," in a meeting, the device's microphone captures this voice and emotion and sends it to the server. The server converts this into text data, "This is a wonderful project," and then translates it into English as "This is a wonderful project." Emotional information is also added to this translated text. This is then converted into audio data and sent to the device as audio reflecting the emotion of joy. The device receives this and plays the joyful English voice saying "This is a wonderful project" from its speaker, while the same content along with the emotional information is displayed on the screen.

[0425] Through these steps, the system provides highly accurate multilingual translation in real time, facilitating meetings and face-to-face communication with full emotional impact.

[0426] Example 2

[0427] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0428] Translation systems that facilitate smooth communication between multiple languages ​​require real-time performance and the ability to accurately reflect emotions. Conventional translation systems face the challenge of simultaneously achieving both real-time performance and the ability to accurately reflect emotions. Furthermore, there are limitations to methods for efficiently transmitting speech data while maintaining its quality, and degradation can occur, particularly during data compression and decompression.

[0429] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0430] In this invention, the server includes means for transmitting voice input data to the processing device, means for converting the voice input data into text data in the processing device, means for translating the converted text data into another specified language, means for converting the translated text data into voice data, means for transmitting the voice data and the translated text data to the terminal, and means for analyzing emotional information and reflecting it in the corresponding data. This not only enables accurate multilingual translation in real time while preventing degradation of the voice data, but also enables natural communication that reflects the user's emotions.

[0431] "Voice input data" is data that represents in digital form the voice uttered by the user.

[0432] A "processing device" is a device that has the functionality to analyze, convert, and communicate voice data. It is synonymous with a server.

[0433] "Text data" is character information data generated by voice recognition.

[0434] "Translation" is the process of converting text data in one language into text data in another language.

[0435] "Audio data" is digital data in which the contents of text data are synthesized as audio.

[0436] "Lossless compression" is a technology that reduces data size without compromising data quality.

[0437] "Remote services" are services provided via the Internet using cloud computing technology.

[0438] A "voice recognition engine" is software or hardware that analyzes voice data and converts it into corresponding text data.

[0439] A "language translation engine" is software or hardware for translating text data into another language.

[0440] "Emotion information" is information that indicates the emotional state of the user analyzed from the voice data.

[0441] The system of this invention performs real-time translation and voice / text output in multiple languages, and by combining it with an emotion engine that recognizes the user's emotions, it achieves more natural and rich communication that includes emotions.

[0442] First, the device captures the user's voice using a microphone. The captured voice signal is converted from analog to digital and then transmitted to a processing unit using a lossless compression technique to optimize data volume. This compression technique uses a common lossless compression technique.

[0443] The server then analyzes the received voice data using a speech recognition engine. The voice signal is divided into frames and converted into text data based on acoustic and language models. This process can be performed using a cloud-based speech recognition service, for example.

[0444] The server translates the generated text data into another specified language using a language translation engine provided via a cloud service, for example, translating the text data "This is a wonderful project" into "This is a wonderful project."

[0445] The server is also equipped with an emotion engine that analyzes the user's emotional state from the voice input data. Based on the tone, pitch, speed, and rhythm of the voice, the engine recognizes the user's emotions, such as happiness, anger, and sadness. The analyzed emotional information is reflected in the generated text data and voice data.

[0446] Next, the server converts the translated text data into speech data using a speech synthesis engine. The speech synthesis engine converts the text data into speech while taking into account phonology, morphology, and syntax. During this process, the emotion engine also applies the emotions recognized and reflects them in the translated speech data. For example, a general-purpose speech synthesis service can be used as the speech synthesis engine.

[0447] Finally, the generated voice data and translated text data are sent from the server to the device. The device then plays the received voice data over a speaker and displays the text data on the screen. At this time, emotional information is also reflected in the display. This allows the user to understand the content of speech in another language, along with emotional information, in real time.

[0448] As a concrete example, if a user says with joy, "This is a wonderful project," in a meeting, the device's microphone captures this voice and emotion and sends it to the server. The server converts this into text data, "This is a wonderful project," and then translates it into English as "This is a wonderful project." Emotional information is also added to this translated text. This is then converted into audio data and sent to the device as audio reflecting the emotion of joy. The device receives this and plays the joyful English voice saying "This is a wonderful project" from its speaker, while the same content along with the emotional information is displayed on the screen.

[0449] An example of a prompt might be:

[0450] Please explain the process of capturing a user's speech with a device microphone, sending it to a server for translation, and generating speech and text data with added emotion. Please provide a concrete example, including the translation of "This is a great project" into English.

[0451] This system enables highly accurate real-time multilingual translation and communication that includes emotions, facilitating smoother meetings and face-to-face interactions.

[0452] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0453] The processing flow of this system's program is explained below, divided into processing steps.

[0454] Step 1:

[0455] The user inputs voice. The device's built-in microphone captures what the user is saying. The input voice signal enters the microphone as an analog signal and is then converted into a digital signal. The converted digital voice data proceeds to the next step.

[0456] Step 2:

[0457] The device sends audio data to the server. The digital audio data is compressed in a lossless compression format such as FLAC and then transmitted to the server over the network. The input is digital audio data, and the output is compressed audio data.

[0458] Step 3:

[0459] The server decompresses the received audio data and analyzes it using a speech recognition engine. The FLAC format compressed data is decompressed, and then the speech recognition engine is used to convert the audio data into text data. This engine divides the audio signal into frames, each of which is analyzed using a phonological model and a language model. The input is compressed audio data, and the output is text data.

[0460] Step 4:

[0461] The server translates the converted text data. The text data is translated into the target language using a language translation engine via a cloud service. For example, the Japanese text "This is a wonderful project" is translated into the English text "This is a wonderful project." The input is Japanese text data, and the output is English text data.

[0462] Step 5:

[0463] The server's emotion engine analyzes the user's emotions from the voice input data. It identifies the user's emotional state based on the tone, pitch, speed, rhythm, etc. of the voice. This emotional information is reflected in the translated text data and subsequent voice data. The input is voice data, and the output is emotional information.

[0464] Step 6:

[0465] The server converts the translated text data into speech data using a speech synthesis engine. The translated text is converted into speech while taking into account phonology, morphology, and syntax. At this time, emotional information obtained from the emotion engine is reflected in the speech data. For example, a speech with an emotional expression of joy is generated. The input is the translated text data and emotional information, and the output is speech data that reflects the emotion.

[0466] Step 7:

[0467] The server sends the generated voice data and translated text data to the terminal. The voice data received by the terminal is played back using a speaker, and the translated text data is displayed on the screen. At this time, emotional information is also included in the display. The input is voice data and text data, and the output is voice and text display on the terminal.

[0468] In this way, each process in the system is generated from the user's input speech, and translated speech and text containing emotions are output in real time.

[0469] (Application example 2)

[0470] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0471] In autonomous vehicles, the language barrier makes it difficult for passengers to communicate in multiple languages. Another problem is that natural and rich communication cannot be achieved because passenger emotions are not reflected. Furthermore, it is necessary to appropriately reflect passenger emotions in real-time translation and voice playback.

[0472] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for transmitting voice input data to a data storage device, means for converting the voice input data into text data in the data storage device, and means for recognizing emotional states and reflecting emotional information in the voice data and text data. This makes it possible to translate the multilingual voices of passengers in an autonomous vehicle in real time and output natural voices and text that reflect emotions.

[0473] "Voice input data" refers to data obtained by converting the user's speech into digital form.

[0474] A "data storage device" is a device that temporarily stores voice input data and text data and performs the necessary processing.

[0475] "Text data" is data that is generated by analyzing voice input data and expressing it as a character string.

[0476] "Translation" is the process of converting text data expressed in one language into text data expressed in another language.

[0477] "Audio data" refers to data obtained by converting text data back into audio format.

[0478] The "emotional state" is a state in which the user is feeling some emotion, which is analyzed from the user's voice.

[0479] "Emotional information" is data that reflects the emotional state of the user and is information that is applied to translated text data or voice data.

[0480] A "receiving terminal" is a device that receives voice data and translated text data and plays or displays them.

[0481] A "voice recognition engine" is software or hardware for converting voice input data into a string of characters.

[0482] A "language translation engine" is software or hardware for translating text data expressed in one language into another language.

[0483] "Information and communication services" refers to infrastructure for utilizing various information processing functions provided on the cloud.

[0484] MODE FOR CARRYING OUT THE INVENTION

[0485] This invention is a multilingual real-time speech translation and emotion recognition system for passengers in autonomous vehicles. To realize this system, the following specific configuration is required.

[0486] First, the user's voice data is captured by a microphone installed in the vehicle. The captured voice data is converted into digital data and then compressed using lossless compression technology (e.g., Zlib). The compressed voice data is then temporarily stored in a data storage device in the vehicle.

[0487] The server then takes this voice data and converts it into text using a speech recognition engine (e.g., Google Speech-to-Text). The converted text data is then sent to a language translation engine (e.g., Google Cloud Translate) for translation into the specified target language. This converts the text data expressed in the original language into another language.

[0488] Furthermore, the server uses an emotion recognition engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotional state from the voice data, and the analyzed emotional information is reflected in the translated text data and the voice data that is later synthesized.

[0489] Next, based on the translated text data and emotion information, a speech synthesis engine (e.g., Amazon Polly) is used to generate voice data that reflects the user's emotions.

[0490] The generated voice data and translated text data are then transmitted to a receiving terminal inside the vehicle, which then displays the text data and emotion information on a display and simultaneously plays the voice data reflecting the emotion from a speaker.

[0491] Examples:

[0492] For example, if a user says in Japanese, "Good morning, please head to the airport," this voice is captured and sent to a data storage device. The server converts this voice into text data and translates it into English as "Good morning, please head to the airport." In addition, the emotion of the user's voice is recognized as "polite and calm," and a voice that reflects this emotion is generated. Finally, the in-car speaker plays "Good morning, please head to the airport" in a polite tone, and the same message is displayed on the in-car display.

[0493] Example prompt sentence:

[0494] If the user says "This is a wonderful project!" in a joyful voice, translate that into English and generate a joyful voice saying "This is a wonderful project!"

[0495] In this way, the system of the present invention realizes natural and rich multilingual communication within an autonomous vehicle.

[0496] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0497] Step 1:

[0498] The user speaks voice commands into a microphone installed inside the vehicle. The user's voice data is captured as input. This voice data is converted from an analog signal to a digital signal, making it easier to process (e.g., PCM format).

[0499] Step 2:

[0500] The audio data captured by the terminal is sent to the data storage device using lossless compression technology (e.g., Zlib). The input is digital audio data, and the output is compressed audio data. This improves data transmission efficiency.

[0501] Step 3:

[0502] The server decompresses the compressed audio data it receives and converts it into text data using a speech recognition engine (e.g., Google Speech-to-Text). The input is compressed audio data and the output is text data. The speech recognition engine analyzes the audio signal frame by frame based on an acoustic model and a language model to generate a string of characters.

[0503] Step 4:

[0504] The server sends the generated text data to a language translation engine (e.g., Google Cloud Translate) to translate it into the specified target language. The input is text data, and the output is text data in the target language. The translation engine uses a neural network to perform highly accurate translation.

[0505] Step 5:

[0506] The server uses an emotion recognition engine (e.g., IBM Watson Tone Analyzer) to analyze the emotional state from the voice data. The input is the original voice data, and the output is emotional information (e.g., joy, anger, surprise, etc.). The emotion engine analyzes the tone, pitch, speed, etc. of the voice to determine the emotional state.

[0507] Step 6:

[0508] The server generates speech data using a speech synthesis engine (e.g., Amazon Polly) based on the translated text data and emotional information. The input is the translated text data and emotional information, and the output is speech data that reflects the emotion. The speech synthesis engine generates natural-sounding speech while taking into account phonology, morphology, and syntax.

[0509] Step 7:

[0510] The server sends the generated voice data and translated text data to the terminal. The input is the voice data and translated text data, and the output is the data received by the terminal. An appropriate protocol (e.g. HTTP / 2) is used for data transmission.

[0511] Step 8:

[0512] The terminal receives voice data and plays it back through the vehicle's speakers, and displays the translated text data and emotional information on the display. The input is the received data, and the output is audiovisual information presentation. This enables users to communicate in multiple languages ​​in real time.

[0513] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0514] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0515] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0516] [Third embodiment]

[0517] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0518] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0519] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0520] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0521] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0522] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0523] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0524] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0525] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0526] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0527] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0528] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0529] The system according to the present invention performs real-time translation and speech-to-text output in multiple languages. This system is implemented as follows.

[0530] First, the device captures the user's voice using a microphone. This voice data is converted into a digital signal and processed in real time. The captured data is then transmitted from the device to a server. This transmission process uses lossless compression technology to optimize data volume.

[0531] Next, the server analyzes the received voice data. It converts the voice data into text data using a voice recognition engine. The voice recognition engine divides the voice signal into frames and analyzes them based on acoustic and language models. For example, the voice data "This is a great project" is converted into text data "This is a great project."

[0532] Next, the generated text data is translated into another language specified by a language translation engine. The server uses the language translation engine via a cloud service to convert the text data into the target language. For example, "This is a wonderful project" is translated into English as "This is a wonderful project."

[0533] The translated text data is then converted into speech data using a speech synthesis engine. The server converts the text into speech, taking into account phonology, morphology, and syntax. The generated speech data is saved as an English audio file. This generates the text "This is a wonderful project" as English speech data.

[0534] The generated voice data and translated text data are then sent to the device, where the voice data is played back through the speaker and the text data is displayed on the screen, allowing the user to understand and respond to speech in another language in real time.

[0535] For example, if a user says "This is a wonderful project" in a meeting, the device's microphone captures this speech and immediately sends it to the server. The server converts this into text data, "This is a wonderful project," and then translates it into English as "This is a wonderful project." This is then converted into audio data and sent to the device. The device receives it, plays it as English speech from the speaker, and displays "This is a wonderful project" on the screen.

[0536] Through these processes, the system provides highly accurate multilingual translation in real time, facilitating meetings and face-to-face communication.

[0537] The processing flow will be explained below.

[0538] Step 1:

[0539] The device captures the user's voice using a microphone, and the voice signal is converted into digital data and processed in real time.

[0540] Step 2:

[0541] The device sends the captured audio data to the server, where lossless compression is used to optimize data size.

[0542] Step 3:

[0543] The server analyzes the received voice data using a voice recognition engine and converts the voice data into text data. The voice signal is divided into frames and analyzed based on acoustic and language models.

[0544] Step 4:

[0545] The server translates the generated text data into another language specified by the language translation engine. The translation engine is used via a cloud service to convert the text data into the target language.

[0546] Step 5:

[0547] The server converts the translated text data into speech data using a speech synthesis engine, taking into account phonology, morphology, and syntax, and stores the generated speech data.

[0548] Step 6:

[0549] The server transmits the generated voice data and the translated text data to the terminal.

[0550] Step 7:

[0551] The device receives the voice data, plays it back through the speaker, and displays the translated text data on the screen, allowing users to understand and respond to what is being said in another language in real time.

[0552] Example 1

[0553] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0554] In today's globalized society, there is a demand for real-time translation systems between multiple languages. However, existing translation systems have problems with speech recognition accuracy and translation speed and accuracy, making it difficult to communicate smoothly in real time. This poses a challenge in the smooth conduct of multilingual meetings and face-to-face communication.

[0555] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0556] In this invention, the server includes means for optimizing voice input data using lossless compression technology and transmitting the optimized data to the server, means for converting the voice input data into text data using a physical model in the server, and means for translating the converted text data into another specified language via a cloud service, thereby enabling highly accurate multilingual translation in real time.

[0557] "Voice input data" refers to data obtained by converting the user's speech into digital form.

[0558] "Lossless compression technology" is a compression method that reduces the data volume but allows the original data to be completely restored.

[0559] A "server" is a computer system that processes, stores, and provides data over a network.

[0560] A "voice recognition engine" is software or hardware for converting voice data into text data.

[0561] A "physical model" is an algorithm for analyzing a speech signal, including an acoustic model and a language model.

[0562] "Cloud services" is a general term for various computing resources and services provided via the Internet.

[0563] A "language translation engine" is software or hardware for translating text data in one language into another language.

[0564] A "speech synthesis engine" is software or hardware that converts text data into human speech.

[0565] A "terminal" is an electronic device that can be directly operated by a user and has an interface for inputting and outputting voice.

[0566] A "digital signal" is a signal that is produced by sampling continuous analog data at regular intervals and converting it into digital form.

[0567] The system according to the present invention performs real-time translation and speech-to-text output in multiple languages. This system is implemented as follows.

[0568] First, the device captures the user's voice using a microphone. This capture process uses the device's built-in microphone. For example, say someone says "This is a great project" in a meeting. This voice is picked up by the microphone as an analog signal and then immediately converted into a digital signal.

[0569] The device then converts the captured audio data into a digital signal and optimizes the data size using lossless compression techniques, such as FLAC (Free Lossless Audio Codec), and transmits the compressed audio data to the server via a secure protocol (e.g., HTTPS).

[0570] The server launches a speech recognition engine to analyze the received voice data. A typical cloud-based speech recognition service (e.g., Google Cloud Speech-to-Text or Microsoft Azure Cognitive Services) is used. The server divides the voice signal into frames and analyzes them using an acoustic model and a language model. For example, the voice data "This is a great project" is converted into text data "This is a great project."

[0571] The server sends the generated text data to a language translation engine. Specifically, a common cloud-based translation service is used (e.g., Google Translate or DeepL API). The server converts the text data into the specified target language. For example, "This is a wonderful project" is translated into English as "This is a wonderful project."

[0572] Next, the server uses a speech synthesis engine to convert the translated text data into speech data. Specifically, a common cloud-based speech synthesis service (e.g., Amazon Polly or Google Text-to-Speech) is used. The server converts the text data into English speech, taking into account phonology, morphology, and syntax. As a result, the text "This is a wonderful project" is generated as English speech data.

[0573] The server then sends the generated voice data and translated text data to the device. The device receives this, and the voice data is played from the speaker and the text data is displayed on the screen. Specifically, for example, an English voice saying "This is a wonderful project" is played from the speaker, and at the same time, the text "This is a wonderful project" is displayed on the screen.

[0574] Specific examples

[0575] During a meeting, if a user says "This is a wonderful project," the device's microphone captures this audio, compresses it in FLAC format, and sends it to the server. The server then uses a common cloud service (e.g., Google Cloud Speech-to-Text) to convert the audio data into text data saying "This is a wonderful project." The server then translates this text data into English using the Google Translate API, obtaining "This is a wonderful project." Next, Amazon Polly converts this text into audio data and sends it back to the device. Finally, the device plays the English audio "This is a wonderful project" from the speaker and displays the translated text on the screen.

[0576] Prompt Sentence Examples

[0577] Below are some example prompts to input to a generative AI model:

[0578] "Please explain the detailed processing flow of a real-time system that takes Japanese speech data as input, translates that speech into English, and plays it back."

[0579] "Please explain in detail, with concrete examples, the step-by-step operation of a system that performs multilingual translation and speech-to-text output."

[0580] Based on the above, this system provides highly accurate multilingual translation in real time, enabling smooth meetings and face-to-face communication.

[0581] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0582] Step 1:

[0583] The user provides speech input. The device captures the speech data through the microphone. The input is what the user says (e.g., "This is a great project"). In operation, the device converts the analog signal to a digital signal. The output is the digital version of the speech data.

[0584] Step 2:

[0585] The device compresses the generated digital audio data using a lossless compression technique (e.g., FLAC). The input is the digital audio data obtained in step 1. The operation is to apply a compression algorithm such as FLAC. The output is the compressed audio data.

[0586] Step 3:

[0587] The device sends the compressed audio data to the server using a secure protocol (e.g., HTTPS). The input is the compressed audio data obtained in step 2. The operation is to send the data to the server using network communication. The output is the compressed audio data received by the server.

[0588] Step 4:

[0589] The server decompresses the received compressed audio data and converts it into text data using a speech recognition engine. The input is the compressed audio data. The operation is to apply a speech recognition algorithm (e.g., a cloud service-based speech recognition engine) to analyze the audio signal. The output is to generate text data (e.g., "This is a great project").

[0590] Step 5:

[0591] The server sends the generated text data to a language translation engine to translate it into the specified target language. The input is the text data obtained in step 4. The action is to apply a language translation algorithm from a cloud service (e.g., Google Translate). The output is the translated text data (e.g., "This is a wonderful project").

[0592] Step 6:

[0593] The server sends the translated text data to a speech synthesis engine and converts it into audio data. The input is the translated text data obtained in step 5. The action is to apply a speech synthesis algorithm (e.g., Amazon Polly). The output is audio data (e.g., an English audio file).

[0594] Step 7:

[0595] The server sends the generated voice data and translated text data to the terminal. The input is the voice data and translated text data obtained in step 6. The operation is to send the data to the terminal using network communication. The output is the voice data and translated text data received by the terminal.

[0596] Step 8:

[0597] The device plays the received audio data from the speaker and displays the translated text data on the screen. The input is the audio data and translated text data obtained in step 7. The operation is to play the audio data from the speaker and display the text data on the display. The output is the audio and text that the user can hear and hear.

[0598] (Application example 1)

[0599] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0600] In manufacturing, service, and other industries where multilingual support is required, workers often communicate in different languages. In these cases, accurate translation in real time is necessary, and misunderstandings and delays in instructions due to language barriers can adversely affect productivity. This problem is particularly pronounced in situations where machine control commands must be translated into multiple languages ​​in real time. Therefore, an objective of this invention is to provide a system that enables real-time translation in multiple languages ​​and instantaneous machine control on-site.

[0601] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0602] In this invention, the server includes means for converting voice input data into text data, means for translating the converted text data into another specified language, means for converting the translated text data into voice data, and means for generating machine control commands based on the text data and controlling the corresponding machine, thereby enabling smooth communication between workers who speak different languages ​​and enabling the machine control commands to be executed in real time.

[0603] "Voice input data" refers to data obtained in digital form from the voice spoken by the user.

[0604] A "server" is a computer system that processes voice input data and provides functions such as conversion to text data, translation, and voice synthesis.

[0605] "Text data" refers to data obtained by converting voice input data into character information.

[0606] "Another language" is a language different from the original language of the speech input data.

[0607] "Translation" is the process of converting text data from one language into another.

[0608] "Audio data" refers to an audio file generated by voice synthesis of text data.

[0609] A "terminal" is a device that receives, plays back, and displays voice data and translated text data.

[0610] A "machine control command" is a command generated based on text data for operating a corresponding mechanical device.

[0611] "Machinery" refers to automated machinery and equipment used in factories, work sites, etc.

[0612] A system embodying this invention is configured as follows: First, the user's speech is captured by a microphone installed in the terminal. This speech input data is converted into a digital format and transmitted from the terminal to a server. At this time, the speech input data is losslessly compressed.

[0613] The server converts the received voice input data into text data using a speech recognition engine, which uses cloud services such as the Google Cloud Speech-to-Text API, and then translates the text data into another specified language using the Google Cloud Translate API.

[0614] The translated text data is converted to audio data using the Google Cloud Text-to-Speech API. The generated audio data is saved on the server as an audio file in the target language. This audio data and the translated text data are then sent back to the device.

[0615] The terminal plays the received voice data through a speaker and displays the translated text data on a display. It also generates machine control commands based on the text data and controls the corresponding machinery. This enables smooth communication between workers who speak different languages ​​and enables machine control commands to be executed in real time.

[0616] As a specific example, if factory worker A says "Please start line 1" in Japanese, this voice is captured by the microphone on the terminal and sent digitally to the server. The server converts the voice into text data and translates "Please start line 1" into English as "Please start line 1." This is converted back into voice data and sent back to the terminal. The terminal's speaker plays "Please start line 1" and the display shows the English text. At the same time, the corresponding machine device receives a command to start line 1.

[0617] Examples of prompts powered by generative AI models include:

[0618] "You are a smart solutions engineer developing a robot assistant for use in factories. The robot assistant needs to respond to multilingual instructions in real time. You will use Google Cloud APIs for implementation. Please generate a specific Python script based on the following requirements:

[0619] 1. Record the voice instructions of the worker.

[0620] 2. Convert the recorded voice into text.

[0621] 3. Translate the text into English.

[0622] 4. Save the translated text as an audio file.

[0623] 5. Play the generated audio file and give instructions to the robot.

[0624] This enables the system to realize real-time multilingual translation and machine control in different language environments, significantly improving work efficiency within factories.

[0625] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0626] Step 1:

[0627] The user speaks a voice command. This voice command is captured by the microphone on the device and obtained as digital voice input data. The input is the user's voice, and the output is digital voice input data.

[0628] Step 2:

[0629] The device losslessly compresses the captured audio input data, then transmits the compressed data to the server. The input is the audio input data in digital form, and the output is the losslessly compressed audio input data.

[0630] Step 3:

[0631] The server converts the received voice input data into text data using the Google Cloud Speech-to-Text API. During this process, the voice signal is divided into frames and analyzed based on acoustic and language models. The input is losslessly compressed voice input data, and the output is converted text data.

[0632] Step 4:

[0633] The server translates the converted text data into another specified language using the Google Cloud Translate API. This process involves language identification and translation of the text. The input is the text data, and the output is the translated text data.

[0634] Step 5:

[0635] The server converts the translated text data into audio data using the Google Cloud Text-to-Speech API. In this step, the text is synthesized into audio taking into account phonology, morphology, and syntax. The input is the translated text data, and the output is audio data.

[0636] Step 6:

[0637] The server sends the generated voice data and translated text data to the terminal. The input is the voice data and translated text data, and the output is the voice data and translated text data sent to the terminal.

[0638] Step 7:

[0639] The terminal plays the received voice data from a speaker, displays the translated text data on a display, and generates machine control commands based on the text data to control the corresponding machine. The inputs are the received voice data and the translated text data, and the outputs are voice playback, text display, and machine control execution.

[0640] This process flow allows smooth communication between workers who speak different languages ​​and enables machine control commands to be executed in real time.

[0641] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0642] The system of this invention performs real-time translation and speech / text output in multiple languages, and by combining it with an emotion engine that recognizes the user's emotions, it realizes more natural and rich communication that includes emotions. This system is implemented as follows.

[0643] First, the device captures the user's voice using a microphone. The voice signal is converted into digital data and processed in real time. The captured data is then sent from the device to a server. Lossless compression technology is used to optimize data volume during the transmission process.

[0644] The server then analyzes the received voice data using a speech recognition engine and converts it into text data. The voice signal is divided into frames and analyzed based on acoustic and language models to generate accurate text data.

[0645] The server translates the generated text data into another language specified by the language translation engine. The server uses the translation engine via a cloud service to convert the text data into the target language. For example, "This is a wonderful project" is translated into English as "This is a wonderful project."

[0646] The server also includes an emotion engine, which analyzes the user's emotional state from the voice input data. The emotion engine analyzes the tone, pitch, speed, rhythm, etc. of the voice to recognize the user's emotions, such as happiness, anger, or sadness. The recognized emotional information is reflected in the generated translated text and voice data.

[0647] The server then converts the translated text data into speech data using a speech synthesis engine. The text data is converted into speech while taking into account phonology, morphology, and syntax, and the generated speech data is saved. During this process, the emotion engine also applies the emotions recognized by the server, ensuring that the translated speech data reflects the emotions.

[0648] The server then sends the generated voice data and translated text data to the device. The voice data received by the device is played back through the speaker, and the text data is displayed on the screen. Emotional information is also reflected on the device's display. This allows the user to understand and respond to content spoken in another language, along with emotional information, in real time.

[0649] As a concrete example, if a user says with joy, "This is a wonderful project," in a meeting, the device's microphone captures this voice and emotion and sends it to the server. The server converts this into text data, "This is a wonderful project," and then translates it into English as "This is a wonderful project." Emotional information is also added to this translated text. This is then converted into audio data and sent to the device as audio reflecting the emotion of joy. The device receives this and plays the joyful English voice saying "This is a wonderful project" from its speaker, while the same content along with the emotional information is displayed on the screen.

[0650] Through these processes, the system provides highly accurate multilingual translation in real time, facilitating meetings and face-to-face communication while incorporating emotions.

[0651] The processing flow will be explained below.

[0652] Step 1:

[0653] The device captures the user's voice using a microphone. The voice signal is converted into digital data and processed in real time. The voice data includes the user's voice tone and pitch.

[0654] Step 2:

[0655] The device sends the captured audio data to the server, where lossless compression is used to optimize data size.

[0656] Step 3:

[0657] The server analyzes the received voice data using a voice recognition engine and converts the voice data into text data. The voice signal is divided into frames and analyzed based on acoustic and language models to generate accurate text data.

[0658] Step 4:

[0659] The server analyzes the voice input data using an emotion engine to extract the user's emotional state. This analysis uses the tone, pitch, speed, and rhythm of the voice. For example, if the voice tone is high and fast, it detects the emotion of joy.

[0660] Step 5:

[0661] The server translates the generated text data into another language specified by the language translation engine. The translation engine is used via a cloud service to convert the text data into the target language. For example, "This is a wonderful project" is translated into English as "This is a wonderful project."

[0662] Step 6:

[0663] The server adds the extracted emotion information to the translated text data, so that the emotion is reflected in the translated text.

[0664] Step 7:

[0665] The server converts the translated text data into speech data using a speech synthesis engine. The text data is converted into speech taking into account phonology, morphology, and syntax, and emotional information from the emotion engine is also reflected in the speech data. For example, the text "This is a wonderful project" is converted into speech with a joyful tone.

[0666] Step 8:

[0667] The server transmits the generated voice data and the translated text data to the terminal.

[0668] Step 9:

[0669] The device plays the received voice data through the speaker and displays the translated text data on the screen. Furthermore, emotional information is reflected in the text data, so the user can check the emotional content of the voice and text in real time.

[0670] For example, if a user says with joy, "This is a wonderful project," in a meeting, the device's microphone captures this voice and emotion and sends it to the server. The server converts this into text data, "This is a wonderful project," and then translates it into English as "This is a wonderful project." Emotional information is also added to this translated text. This is then converted into audio data and sent to the device as audio reflecting the emotion of joy. The device receives this and plays the joyful English voice saying "This is a wonderful project" from its speaker, while the same content along with the emotional information is displayed on the screen.

[0671] Through these steps, the system provides highly accurate multilingual translation in real time, facilitating meetings and face-to-face communication with full emotional impact.

[0672] Example 2

[0673] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0674] Translation systems that facilitate smooth communication between multiple languages ​​require real-time performance and the ability to accurately reflect emotions. Conventional translation systems face the challenge of simultaneously achieving both real-time performance and the ability to accurately reflect emotions. Furthermore, there are limitations to methods for efficiently transmitting speech data while maintaining its quality, and degradation can occur, particularly during data compression and decompression.

[0675] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0676] In this invention, the server includes means for transmitting voice input data to the processing device, means for converting the voice input data into text data in the processing device, means for translating the converted text data into another specified language, means for converting the translated text data into voice data, means for transmitting the voice data and the translated text data to the terminal, and means for analyzing emotional information and reflecting it in the corresponding data. This not only enables accurate multilingual translation in real time while preventing degradation of the voice data, but also enables natural communication that reflects the user's emotions.

[0677] "Voice input data" is data that represents in digital form the voice uttered by the user.

[0678] A "processing device" is a device that has the functionality to analyze, convert, and communicate voice data. It is synonymous with a server.

[0679] "Text data" is character information data generated by voice recognition.

[0680] "Translation" is the process of converting text data in one language into text data in another language.

[0681] "Audio data" is digital data in which the contents of text data are synthesized as audio.

[0682] "Lossless compression" is a technology that reduces data size without compromising data quality.

[0683] "Remote services" are services provided via the Internet using cloud computing technology.

[0684] A "voice recognition engine" is software or hardware that analyzes voice data and converts it into corresponding text data.

[0685] A "language translation engine" is software or hardware for translating text data into another language.

[0686] "Emotion information" is information that indicates the emotional state of the user analyzed from the voice data.

[0687] The system of this invention performs real-time translation and voice / text output in multiple languages, and by combining it with an emotion engine that recognizes the user's emotions, it achieves more natural and rich communication that includes emotions.

[0688] First, the device captures the user's voice using a microphone. The captured voice signal is converted from analog to digital and then transmitted to a processing unit using a lossless compression technique to optimize data volume. This compression technique uses a common lossless compression technique.

[0689] The server then analyzes the received voice data using a speech recognition engine. The voice signal is divided into frames and converted into text data based on acoustic and language models. This process can be performed using a cloud-based speech recognition service, for example.

[0690] The server translates the generated text data into another specified language using a language translation engine provided via a cloud service, for example, translating the text data "This is a wonderful project" into "This is a wonderful project."

[0691] The server is also equipped with an emotion engine that analyzes the user's emotional state from the voice input data. Based on the tone, pitch, speed, and rhythm of the voice, the engine recognizes the user's emotions, such as happiness, anger, and sadness. The analyzed emotional information is reflected in the generated text data and voice data.

[0692] Next, the server converts the translated text data into speech data using a speech synthesis engine. The speech synthesis engine converts the text data into speech while taking into account phonology, morphology, and syntax. During this process, the emotion engine also applies the emotions recognized and reflects them in the translated speech data. For example, a general-purpose speech synthesis service can be used as the speech synthesis engine.

[0693] Finally, the generated voice data and translated text data are sent from the server to the device. The device then plays the received voice data over a speaker and displays the text data on the screen. At this time, emotional information is also reflected in the display. This allows the user to understand the content of speech in another language, along with emotional information, in real time.

[0694] As a concrete example, if a user says with joy, "This is a wonderful project," in a meeting, the device's microphone captures this voice and emotion and sends it to the server. The server converts this into text data, "This is a wonderful project," and then translates it into English as "This is a wonderful project." Emotional information is also added to this translated text. This is then converted into audio data and sent to the device as audio reflecting the emotion of joy. The device receives this and plays the joyful English voice saying "This is a wonderful project" from its speaker, while the same content along with the emotional information is displayed on the screen.

[0695] An example of a prompt might be:

[0696] Please explain the process of capturing a user's speech with a device microphone, sending it to a server for translation, and generating speech and text data with added emotion. Please provide a concrete example, including the translation of "This is a great project" into English.

[0697] This system enables highly accurate real-time multilingual translation and communication that includes emotions, facilitating smoother meetings and face-to-face interactions.

[0698] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0699] The processing flow of this system's program is explained below, divided into processing steps.

[0700] Step 1:

[0701] The user inputs voice. The device's built-in microphone captures what the user is saying. The input voice signal enters the microphone as an analog signal and is then converted into a digital signal. The converted digital voice data proceeds to the next step.

[0702] Step 2:

[0703] The device sends audio data to the server. The digital audio data is compressed in a lossless compression format such as FLAC and then transmitted to the server over the network. The input is digital audio data, and the output is compressed audio data.

[0704] Step 3:

[0705] The server decompresses the received audio data and analyzes it using a speech recognition engine. The FLAC format compressed data is decompressed, and then the speech recognition engine is used to convert the audio data into text data. This engine divides the audio signal into frames, each of which is analyzed using a phonological model and a language model. The input is compressed audio data, and the output is text data.

[0706] Step 4:

[0707] The server translates the converted text data. The text data is translated into the target language using a language translation engine via a cloud service. For example, the Japanese text "This is a wonderful project" is translated into the English text "This is a wonderful project." The input is Japanese text data, and the output is English text data.

[0708] Step 5:

[0709] The server's emotion engine analyzes the user's emotions from the voice input data. It identifies the user's emotional state based on the tone, pitch, speed, rhythm, etc. of the voice. This emotional information is reflected in the translated text data and subsequent voice data. The input is voice data, and the output is emotional information.

[0710] Step 6:

[0711] The server converts the translated text data into speech data using a speech synthesis engine. The translated text is converted into speech while taking into account phonology, morphology, and syntax. At this time, emotional information obtained from the emotion engine is reflected in the speech data. For example, a speech with an emotional expression of joy is generated. The input is the translated text data and emotional information, and the output is speech data that reflects the emotion.

[0712] Step 7:

[0713] The server sends the generated voice data and translated text data to the terminal. The voice data received by the terminal is played back using a speaker, and the translated text data is displayed on the screen. At this time, emotional information is also included in the display. The input is voice data and text data, and the output is voice and text display on the terminal.

[0714] In this way, each process in the system is generated from the user's input speech, and translated speech and text containing emotions are output in real time.

[0715] (Application example 2)

[0716] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0717] In autonomous vehicles, the language barrier makes it difficult for passengers to communicate in multiple languages. Another problem is that natural and rich communication cannot be achieved because passenger emotions are not reflected. Furthermore, it is necessary to appropriately reflect passenger emotions in real-time translation and voice playback.

[0718] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for transmitting voice input data to a data storage device, means for converting the voice input data into text data in the data storage device, and means for recognizing emotional states and reflecting emotional information in the voice data and text data. This makes it possible to translate the multilingual voices of passengers in an autonomous vehicle in real time and output natural voices and text that reflect emotions.

[0719] "Voice input data" refers to data obtained by converting the user's speech into digital form.

[0720] A "data storage device" is a device that temporarily stores voice input data and text data and performs the necessary processing.

[0721] "Text data" is data that is generated by analyzing voice input data and expressing it as a character string.

[0722] "Translation" is the process of converting text data expressed in one language into text data expressed in another language.

[0723] "Audio data" refers to data obtained by converting text data back into audio format.

[0724] The "emotional state" is a state in which the user is feeling some emotion, which is analyzed from the user's voice.

[0725] "Emotional information" is data that reflects the emotional state of the user and is information that is applied to translated text data or voice data.

[0726] A "receiving terminal" is a device that receives voice data and translated text data and plays or displays them.

[0727] A "voice recognition engine" is software or hardware for converting voice input data into a string of characters.

[0728] A "language translation engine" is software or hardware for translating text data expressed in one language into another language.

[0729] "Information and communication services" refers to infrastructure for utilizing various information processing functions provided on the cloud.

[0730] MODE FOR CARRYING OUT THE INVENTION

[0731] This invention is a multilingual real-time speech translation and emotion recognition system for passengers in autonomous vehicles. To realize this system, the following specific configuration is required.

[0732] First, the user's voice data is captured by a microphone installed in the vehicle. The captured voice data is converted into digital data and then compressed using lossless compression technology (e.g., Zlib). The compressed voice data is then temporarily stored in a data storage device in the vehicle.

[0733] The server then takes this voice data and converts it into text using a speech recognition engine (e.g., Google Speech-to-Text). The converted text data is then sent to a language translation engine (e.g., Google Cloud Translate) for translation into the specified target language. This converts the text data expressed in the original language into another language.

[0734] Furthermore, the server uses an emotion recognition engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotional state from the voice data, and the analyzed emotional information is reflected in the translated text data and the voice data that is later synthesized.

[0735] Next, based on the translated text data and emotion information, a speech synthesis engine (e.g., Amazon Polly) is used to generate voice data that reflects the user's emotions.

[0736] The generated voice data and translated text data are then transmitted to a receiving terminal inside the vehicle, which then displays the text data and emotion information on a display and simultaneously plays the voice data reflecting the emotion from a speaker.

[0737] Examples:

[0738] For example, if a user says in Japanese, "Good morning, please head to the airport," this voice is captured and sent to a data storage device. The server converts this voice into text data and translates it into English as "Good morning, please head to the airport." In addition, the emotion of the user's voice is recognized as "polite and calm," and a voice that reflects this emotion is generated. Finally, the in-car speaker plays "Good morning, please head to the airport" in a polite tone, and the same message is displayed on the in-car display.

[0739] Example prompt sentence:

[0740] If the user says "This is a wonderful project!" in a joyful voice, translate that into English and generate a joyful voice saying "This is a wonderful project!"

[0741] In this way, the system of the present invention realizes natural and rich multilingual communication within an autonomous vehicle.

[0742] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0743] Step 1:

[0744] The user speaks voice commands into a microphone installed inside the vehicle. The user's voice data is captured as input. This voice data is converted from an analog signal to a digital signal, making it easier to process (e.g., PCM format).

[0745] Step 2:

[0746] The audio data captured by the terminal is sent to the data storage device using lossless compression technology (e.g., Zlib). The input is digital audio data, and the output is compressed audio data. This improves data transmission efficiency.

[0747] Step 3:

[0748] The server decompresses the compressed audio data it receives and converts it into text data using a speech recognition engine (e.g., Google Speech-to-Text). The input is compressed audio data and the output is text data. The speech recognition engine analyzes the audio signal frame by frame based on an acoustic model and a language model to generate a string of characters.

[0749] Step 4:

[0750] The server sends the generated text data to a language translation engine (e.g., Google Cloud Translate) to translate it into the specified target language. The input is text data, and the output is text data in the target language. The translation engine uses a neural network to perform highly accurate translation.

[0751] Step 5:

[0752] The server uses an emotion recognition engine (e.g., IBM Watson Tone Analyzer) to analyze the emotional state from the voice data. The input is the original voice data, and the output is emotional information (e.g., joy, anger, surprise, etc.). The emotion engine analyzes the tone, pitch, speed, etc. of the voice to determine the emotional state.

[0753] Step 6:

[0754] The server generates speech data using a speech synthesis engine (e.g., Amazon Polly) based on the translated text data and emotional information. The input is the translated text data and emotional information, and the output is speech data that reflects the emotion. The speech synthesis engine generates natural-sounding speech while taking into account phonology, morphology, and syntax.

[0755] Step 7:

[0756] The server sends the generated voice data and translated text data to the terminal. The input is the voice data and translated text data, and the output is the data received by the terminal. An appropriate protocol (e.g. HTTP / 2) is used for data transmission.

[0757] Step 8:

[0758] The terminal receives voice data and plays it back through the vehicle's speakers, and displays the translated text data and emotional information on the display. The input is the received data, and the output is audiovisual information presentation. This enables users to communicate in multiple languages ​​in real time.

[0759] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0760] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0761] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[0762] [Fourth embodiment]

[0763] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0764] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0765] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0766] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[0767] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0768] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0769] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0770] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[0771] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0772] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0773] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0774] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0775] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0776] The system according to the present invention performs real-time translation and speech-to-text output in multiple languages. This system is implemented as follows.

[0777] First, the device captures the user's voice using a microphone. This voice data is converted into a digital signal and processed in real time. The captured data is then transmitted from the device to a server. This transmission process uses lossless compression technology to optimize data volume.

[0778] Next, the server analyzes the received voice data. It converts the voice data into text data using a voice recognition engine. The voice recognition engine divides the voice signal into frames and analyzes them based on acoustic and language models. For example, the voice data "This is a great project" is converted into text data "This is a great project."

[0779] Next, the generated text data is translated into another language specified by a language translation engine. The server uses the language translation engine via a cloud service to convert the text data into the target language. For example, "This is a wonderful project" is translated into English as "This is a wonderful project."

[0780] The translated text data is then converted into speech data using a speech synthesis engine. The server converts the text into speech, taking into account phonology, morphology, and syntax. The generated speech data is saved as an English audio file. This generates the text "This is a wonderful project" as English speech data.

[0781] The generated voice data and translated text data are then sent to the device, where the voice data is played back through the speaker and the text data is displayed on the screen, allowing the user to understand and respond to speech in another language in real time.

[0782] For example, if a user says "This is a wonderful project" in a meeting, the device's microphone captures this speech and immediately sends it to the server. The server converts this into text data, "This is a wonderful project," and then translates it into English as "This is a wonderful project." This is then converted into audio data and sent to the device. The device receives it, plays it as English speech from the speaker, and displays "This is a wonderful project" on the screen.

[0783] Through these processes, the system provides highly accurate multilingual translation in real time, facilitating meetings and face-to-face communication.

[0784] The processing flow will be explained below.

[0785] Step 1:

[0786] The device captures the user's voice using a microphone, and the voice signal is converted into digital data and processed in real time.

[0787] Step 2:

[0788] The device sends the captured audio data to the server, where lossless compression is used to optimize data size.

[0789] Step 3:

[0790] The server analyzes the received voice data using a voice recognition engine and converts the voice data into text data. The voice signal is divided into frames and analyzed based on acoustic and language models.

[0791] Step 4:

[0792] The server translates the generated text data into another language specified by the language translation engine. The translation engine is used via a cloud service to convert the text data into the target language.

[0793] Step 5:

[0794] The server converts the translated text data into speech data using a speech synthesis engine, taking into account phonology, morphology, and syntax, and stores the generated speech data.

[0795] Step 6:

[0796] The server transmits the generated voice data and the translated text data to the terminal.

[0797] Step 7:

[0798] The device receives the voice data, plays it back through the speaker, and displays the translated text data on the screen, allowing users to understand and respond to what is being said in another language in real time.

[0799] Example 1

[0800] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0801] In today's globalized society, there is a demand for real-time translation systems between multiple languages. However, existing translation systems have problems with speech recognition accuracy and translation speed and accuracy, making it difficult to communicate smoothly in real time. This poses a challenge in the smooth conduct of multilingual meetings and face-to-face communication.

[0802] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0803] In this invention, the server includes means for optimizing voice input data using lossless compression technology and transmitting the optimized data to the server, means for converting the voice input data into text data using a physical model in the server, and means for translating the converted text data into another specified language via a cloud service, thereby enabling highly accurate multilingual translation in real time.

[0804] "Voice input data" refers to data obtained by converting the user's speech into digital form.

[0805] "Lossless compression technology" is a compression method that reduces the data volume but allows the original data to be completely restored.

[0806] A "server" is a computer system that processes, stores, and provides data over a network.

[0807] A "voice recognition engine" is software or hardware for converting voice data into text data.

[0808] A "physical model" is an algorithm for analyzing a speech signal, including an acoustic model and a language model.

[0809] "Cloud services" is a general term for various computing resources and services provided via the Internet.

[0810] A "language translation engine" is software or hardware for translating text data in one language into another language.

[0811] A "speech synthesis engine" is software or hardware that converts text data into human speech.

[0812] A "terminal" is an electronic device that can be directly operated by a user and has an interface for inputting and outputting voice.

[0813] A "digital signal" is a signal that is produced by sampling continuous analog data at regular intervals and converting it into digital form.

[0814] The system according to the present invention performs real-time translation and speech-to-text output in multiple languages. This system is implemented as follows.

[0815] First, the device captures the user's voice using a microphone. This capture process uses the device's built-in microphone. For example, say someone says "This is a great project" in a meeting. This voice is picked up by the microphone as an analog signal and then immediately converted into a digital signal.

[0816] The device then converts the captured audio data into a digital signal and optimizes the data size using lossless compression techniques, such as FLAC (Free Lossless Audio Codec), and transmits the compressed audio data to the server via a secure protocol (e.g., HTTPS).

[0817] The server launches a speech recognition engine to analyze the received voice data. A typical cloud-based speech recognition service (e.g., Google Cloud Speech-to-Text or Microsoft Azure Cognitive Services) is used. The server divides the voice signal into frames and analyzes them using an acoustic model and a language model. For example, the voice data "This is a great project" is converted into text data "This is a great project."

[0818] The server sends the generated text data to a language translation engine. Specifically, a common cloud-based translation service is used (e.g., Google Translate or DeepL API). The server converts the text data into the specified target language. For example, "This is a wonderful project" is translated into English as "This is a wonderful project."

[0819] Next, the server uses a speech synthesis engine to convert the translated text data into speech data. Specifically, a common cloud-based speech synthesis service (e.g., Amazon Polly or Google Text-to-Speech) is used. The server converts the text data into English speech, taking into account phonology, morphology, and syntax. As a result, the text "This is a wonderful project" is generated as English speech data.

[0820] The server then sends the generated voice data and translated text data to the device. The device receives this, and the voice data is played from the speaker and the text data is displayed on the screen. Specifically, for example, an English voice saying "This is a wonderful project" is played from the speaker, and at the same time, the text "This is a wonderful project" is displayed on the screen.

[0821] Specific examples

[0822] During a meeting, if a user says "This is a wonderful project," the device's microphone captures this audio, compresses it in FLAC format, and sends it to the server. The server then uses a common cloud service (e.g., Google Cloud Speech-to-Text) to convert the audio data into text data saying "This is a wonderful project." The server then translates this text data into English using the Google Translate API, obtaining "This is a wonderful project." Next, Amazon Polly converts this text into audio data and sends it back to the device. Finally, the device plays the English audio "This is a wonderful project" from the speaker and displays the translated text on the screen.

[0823] Prompt Sentence Examples

[0824] Below are some example prompts to input to a generative AI model:

[0825] "Please explain the detailed processing flow of a real-time system that takes Japanese speech data as input, translates that speech into English, and plays it back."

[0826] "Please explain in detail, with concrete examples, the step-by-step operation of a system that performs multilingual translation and speech-to-text output."

[0827] Based on the above, this system provides highly accurate multilingual translation in real time, enabling smooth meetings and face-to-face communication.

[0828] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0829] Step 1:

[0830] The user provides speech input. The device captures the speech data through the microphone. The input is what the user says (e.g., "This is a great project"). In operation, the device converts the analog signal to a digital signal. The output is the digital version of the speech data.

[0831] Step 2:

[0832] The device compresses the generated digital audio data using a lossless compression technique (e.g., FLAC). The input is the digital audio data obtained in step 1. The operation is to apply a compression algorithm such as FLAC. The output is the compressed audio data.

[0833] Step 3:

[0834] The device sends the compressed audio data to the server using a secure protocol (e.g., HTTPS). The input is the compressed audio data obtained in step 2. The operation is to send the data to the server using network communication. The output is the compressed audio data received by the server.

[0835] Step 4:

[0836] The server decompresses the received compressed audio data and converts it into text data using a speech recognition engine. The input is the compressed audio data. The operation is to apply a speech recognition algorithm (e.g., a cloud service-based speech recognition engine) to analyze the audio signal. The output is to generate text data (e.g., "This is a great project").

[0837] Step 5:

[0838] The server sends the generated text data to a language translation engine to translate it into the specified target language. The input is the text data obtained in step 4. The action is to apply a language translation algorithm from a cloud service (e.g., Google Translate). The output is the translated text data (e.g., "This is a wonderful project").

[0839] Step 6:

[0840] The server sends the translated text data to a speech synthesis engine and converts it into audio data. The input is the translated text data obtained in step 5. The action is to apply a speech synthesis algorithm (e.g., Amazon Polly). The output is audio data (e.g., an English audio file).

[0841] Step 7:

[0842] The server sends the generated voice data and translated text data to the terminal. The input is the voice data and translated text data obtained in step 6. The operation is to send the data to the terminal using network communication. The output is the voice data and translated text data received by the terminal.

[0843] Step 8:

[0844] The device plays the received audio data from the speaker and displays the translated text data on the screen. The input is the audio data and translated text data obtained in step 7. The operation is to play the audio data from the speaker and display the text data on the display. The output is the audio and text that the user can hear and hear.

[0845] (Application example 1)

[0846] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0847] In manufacturing, service, and other industries where multilingual support is required, workers often communicate in different languages. In these cases, accurate translation in real time is necessary, and misunderstandings and delays in instructions due to language barriers can adversely affect productivity. This problem is particularly pronounced in situations where machine control commands must be translated into multiple languages ​​in real time. Therefore, an objective of this invention is to provide a system that enables real-time translation in multiple languages ​​and instantaneous machine control on-site.

[0848] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0849] In this invention, the server includes means for converting voice input data into text data, means for translating the converted text data into another specified language, means for converting the translated text data into voice data, and means for generating machine control commands based on the text data and controlling the corresponding machine, thereby enabling smooth communication between workers who speak different languages ​​and enabling the machine control commands to be executed in real time.

[0850] "Voice input data" refers to data obtained in digital form from the voice spoken by the user.

[0851] A "server" is a computer system that processes voice input data and provides functions such as conversion to text data, translation, and voice synthesis.

[0852] "Text data" refers to data obtained by converting voice input data into character information.

[0853] "Another language" is a language different from the original language of the speech input data.

[0854] "Translation" is the process of converting text data from one language into another.

[0855] "Audio data" refers to an audio file generated by voice synthesis of text data.

[0856] A "terminal" is a device that receives, plays back, and displays voice data and translated text data.

[0857] A "machine control command" is a command generated based on text data for operating a corresponding mechanical device.

[0858] "Machinery" refers to automated machinery and equipment used in factories, work sites, etc.

[0859] A system embodying this invention is configured as follows: First, the user's speech is captured by a microphone installed in the terminal. This speech input data is converted into a digital format and transmitted from the terminal to a server. At this time, the speech input data is losslessly compressed.

[0860] The server converts the received voice input data into text data using a speech recognition engine, which uses cloud services such as the Google Cloud Speech-to-Text API, and then translates the text data into another specified language using the Google Cloud Translate API.

[0861] The translated text data is converted to audio data using the Google Cloud Text-to-Speech API. The generated audio data is saved on the server as an audio file in the target language. This audio data and the translated text data are then sent back to the device.

[0862] The terminal plays the received voice data through a speaker and displays the translated text data on a display. It also generates machine control commands based on the text data and controls the corresponding machinery. This enables smooth communication between workers who speak different languages ​​and enables machine control commands to be executed in real time.

[0863] As a specific example, if factory worker A says "Please start line 1" in Japanese, this voice is captured by the microphone on the terminal and sent digitally to the server. The server converts the voice into text data and translates "Please start line 1" into English as "Please start line 1." This is converted back into voice data and sent back to the terminal. The terminal's speaker plays "Please start line 1" and the display shows the English text. At the same time, the corresponding machine device receives a command to start line 1.

[0864] Examples of prompts powered by generative AI models include:

[0865] "You are a smart solutions engineer developing a robot assistant for use in factories. The robot assistant needs to respond to multilingual instructions in real time. You will use Google Cloud APIs for implementation. Please generate a specific Python script based on the following requirements:

[0866] 1. Record the voice instructions of the worker.

[0867] 2. Convert the recorded voice into text.

[0868] 3. Translate the text into English.

[0869] 4. Save the translated text as an audio file.

[0870] 5. Play the generated audio file and give instructions to the robot.

[0871] This enables the system to realize real-time multilingual translation and machine control in different language environments, significantly improving work efficiency within factories.

[0872] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0873] Step 1:

[0874] The user speaks a voice command. This voice command is captured by the microphone on the device and obtained as digital voice input data. The input is the user's voice, and the output is digital voice input data.

[0875] Step 2:

[0876] The device losslessly compresses the captured audio input data, then transmits the compressed data to the server. The input is the audio input data in digital form, and the output is the losslessly compressed audio input data.

[0877] Step 3:

[0878] The server converts the received voice input data into text data using the Google Cloud Speech-to-Text API. During this process, the voice signal is divided into frames and analyzed based on acoustic and language models. The input is losslessly compressed voice input data, and the output is converted text data.

[0879] Step 4:

[0880] The server translates the converted text data into another specified language using the Google Cloud Translate API. This process involves language identification and translation of the text. The input is the text data, and the output is the translated text data.

[0881] Step 5:

[0882] The server converts the translated text data into audio data using the Google Cloud Text-to-Speech API. In this step, the text is synthesized into audio taking into account phonology, morphology, and syntax. The input is the translated text data, and the output is audio data.

[0883] Step 6:

[0884] The server sends the generated voice data and translated text data to the terminal. The input is the voice data and translated text data, and the output is the voice data and translated text data sent to the terminal.

[0885] Step 7:

[0886] The terminal plays the received voice data from a speaker, displays the translated text data on a display, and generates machine control commands based on the text data to control the corresponding machine. The inputs are the received voice data and the translated text data, and the outputs are voice playback, text display, and machine control execution.

[0887] This process flow allows smooth communication between workers who speak different languages ​​and enables machine control commands to be executed in real time.

[0888] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0889] The system of this invention performs real-time translation and speech / text output in multiple languages, and by combining it with an emotion engine that recognizes the user's emotions, it realizes more natural and rich communication that includes emotions. This system is implemented as follows.

[0890] First, the device captures the user's voice using a microphone. The voice signal is converted into digital data and processed in real time. The captured data is then sent from the device to a server. Lossless compression technology is used to optimize data volume during the transmission process.

[0891] The server then analyzes the received voice data using a speech recognition engine and converts it into text data. The voice signal is divided into frames and analyzed based on acoustic and language models to generate accurate text data.

[0892] The server translates the generated text data into another language specified by the language translation engine. The server uses the translation engine via a cloud service to convert the text data into the target language. For example, "This is a wonderful project" is translated into English as "This is a wonderful project."

[0893] The server also includes an emotion engine, which analyzes the user's emotional state from the voice input data. The emotion engine analyzes the tone, pitch, speed, rhythm, etc. of the voice to recognize the user's emotions, such as happiness, anger, or sadness. The recognized emotional information is reflected in the generated translated text and voice data.

[0894] The server then converts the translated text data into speech data using a speech synthesis engine. The text data is converted into speech while taking into account phonology, morphology, and syntax, and the generated speech data is saved. During this process, the emotion engine also applies the emotions recognized by the server, ensuring that the translated speech data reflects the emotions.

[0895] The server then sends the generated voice data and translated text data to the device. The voice data received by the device is played back through the speaker, and the text data is displayed on the screen. Emotional information is also reflected on the device's display. This allows the user to understand and respond to content spoken in another language, along with emotional information, in real time.

[0896] As a concrete example, if a user says with joy, "This is a wonderful project," in a meeting, the device's microphone captures this voice and emotion and sends it to the server. The server converts this into text data, "This is a wonderful project," and then translates it into English as "This is a wonderful project." Emotional information is also added to this translated text. This is then converted into audio data and sent to the device as audio reflecting the emotion of joy. The device receives this and plays the joyful English voice saying "This is a wonderful project" from its speaker, while the same content along with the emotional information is displayed on the screen.

[0897] Through these processes, the system provides highly accurate multilingual translation in real time, facilitating meetings and face-to-face communication while incorporating emotions.

[0898] The processing flow will be explained below.

[0899] Step 1:

[0900] The device captures the user's voice using a microphone. The voice signal is converted into digital data and processed in real time. The voice data includes the user's voice tone and pitch.

[0901] Step 2:

[0902] The device sends the captured audio data to the server, where lossless compression is used to optimize data size.

[0903] Step 3:

[0904] The server analyzes the received voice data using a voice recognition engine and converts the voice data into text data. The voice signal is divided into frames and analyzed based on acoustic and language models to generate accurate text data.

[0905] Step 4:

[0906] The server analyzes the voice input data using an emotion engine to extract the user's emotional state. This analysis uses the tone, pitch, speed, and rhythm of the voice. For example, if the voice tone is high and fast, it detects the emotion of joy.

[0907] Step 5:

[0908] The server translates the generated text data into another language specified by the language translation engine. The translation engine is used via a cloud service to convert the text data into the target language. For example, "This is a wonderful project" is translated into English as "This is a wonderful project."

[0909] Step 6:

[0910] The server adds the extracted emotion information to the translated text data, so that the emotion is reflected in the translated text.

[0911] Step 7:

[0912] The server converts the translated text data into speech data using a speech synthesis engine. The text data is converted into speech taking into account phonology, morphology, and syntax, and emotional information from the emotion engine is also reflected in the speech data. For example, the text "This is a wonderful project" is converted into speech with a joyful tone.

[0913] Step 8:

[0914] The server transmits the generated voice data and the translated text data to the terminal.

[0915] Step 9:

[0916] The device plays the received voice data through the speaker and displays the translated text data on the screen. Furthermore, emotional information is reflected in the text data, so the user can check the emotional content of the voice and text in real time.

[0917] For example, if a user says with joy, "This is a wonderful project," in a meeting, the device's microphone captures this voice and emotion and sends it to the server. The server converts this into text data, "This is a wonderful project," and then translates it into English as "This is a wonderful project." Emotional information is also added to this translated text. This is then converted into audio data and sent to the device as audio reflecting the emotion of joy. The device receives this and plays the joyful English voice saying "This is a wonderful project" from its speaker, while the same content along with the emotional information is displayed on the screen.

[0918] Through these steps, the system provides highly accurate multilingual translation in real time, facilitating meetings and face-to-face communication with full emotional impact.

[0919] Example 2

[0920] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0921] Translation systems that facilitate smooth communication between multiple languages ​​require real-time performance and the ability to accurately reflect emotions. Conventional translation systems face the challenge of simultaneously achieving both real-time performance and the ability to accurately reflect emotions. Furthermore, there are limitations to methods for efficiently transmitting speech data while maintaining its quality, and degradation can occur, particularly during data compression and decompression.

[0922] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0923] In this invention, the server includes means for transmitting voice input data to the processing device, means for converting the voice input data into text data in the processing device, means for translating the converted text data into another specified language, means for converting the translated text data into voice data, means for transmitting the voice data and the translated text data to the terminal, and means for analyzing emotional information and reflecting it in the corresponding data. This not only enables accurate multilingual translation in real time while preventing degradation of the voice data, but also enables natural communication that reflects the user's emotions.

[0924] "Voice input data" is data that represents in digital form the voice uttered by the user.

[0925] A "processing device" is a device that has the functionality to analyze, convert, and communicate voice data. It is synonymous with a server.

[0926] "Text data" is character information data generated by voice recognition.

[0927] "Translation" is the process of converting text data in one language into text data in another language.

[0928] "Audio data" is digital data in which the contents of text data are synthesized as audio.

[0929] "Lossless compression" is a technology that reduces data size without compromising data quality.

[0930] "Remote services" are services provided via the Internet using cloud computing technology.

[0931] A "voice recognition engine" is software or hardware that analyzes voice data and converts it into corresponding text data.

[0932] A "language translation engine" is software or hardware for translating text data into another language.

[0933] "Emotion information" is information that indicates the emotional state of the user analyzed from the voice data.

[0934] The system of this invention performs real-time translation and voice / text output in multiple languages, and by combining it with an emotion engine that recognizes the user's emotions, it achieves more natural and rich communication that includes emotions.

[0935] First, the device captures the user's voice using a microphone. The captured voice signal is converted from analog to digital and then transmitted to a processing unit using a lossless compression technique to optimize data volume. This compression technique uses a common lossless compression technique.

[0936] The server then analyzes the received voice data using a speech recognition engine. The voice signal is divided into frames and converted into text data based on acoustic and language models. This process can be performed using a cloud-based speech recognition service, for example.

[0937] The server translates the generated text data into another specified language using a language translation engine provided via a cloud service, for example, translating the text data "This is a wonderful project" into "This is a wonderful project."

[0938] The server is also equipped with an emotion engine that analyzes the user's emotional state from the voice input data. Based on the tone, pitch, speed, and rhythm of the voice, the engine recognizes the user's emotions, such as happiness, anger, and sadness. The analyzed emotional information is reflected in the generated text data and voice data.

[0939] Next, the server converts the translated text data into speech data using a speech synthesis engine. The speech synthesis engine converts the text data into speech while taking into account phonology, morphology, and syntax. During this process, the emotion engine also applies the emotions recognized and reflects them in the translated speech data. For example, a general-purpose speech synthesis service can be used as the speech synthesis engine.

[0940] Finally, the generated voice data and translated text data are sent from the server to the device. The device then plays the received voice data over a speaker and displays the text data on the screen. At this time, emotional information is also reflected in the display. This allows the user to understand the content of speech in another language, along with emotional information, in real time.

[0941] As a concrete example, if a user says with joy, "This is a wonderful project," in a meeting, the device's microphone captures this voice and emotion and sends it to the server. The server converts this into text data, "This is a wonderful project," and then translates it into English as "This is a wonderful project." Emotional information is also added to this translated text. This is then converted into audio data and sent to the device as audio reflecting the emotion of joy. The device receives this and plays the joyful English voice saying "This is a wonderful project" from its speaker, while the same content along with the emotional information is displayed on the screen.

[0942] An example of a prompt might be:

[0943] Please explain the process of capturing a user's speech with a device microphone, sending it to a server for translation, and generating speech and text data with added emotion. Please provide a concrete example, including the translation of "This is a great project" into English.

[0944] This system enables highly accurate real-time multilingual translation and communication that includes emotions, facilitating smoother meetings and face-to-face interactions.

[0945] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0946] The processing flow of this system's program is explained below, divided into processing steps.

[0947] Step 1:

[0948] The user inputs voice. The device's built-in microphone captures what the user is saying. The input voice signal enters the microphone as an analog signal and is then converted into a digital signal. The converted digital voice data proceeds to the next step.

[0949] Step 2:

[0950] The device sends audio data to the server. The digital audio data is compressed in a lossless compression format such as FLAC and then transmitted to the server over the network. The input is digital audio data, and the output is compressed audio data.

[0951] Step 3:

[0952] The server decompresses the received audio data and analyzes it using a speech recognition engine. The FLAC format compressed data is decompressed, and then the speech recognition engine is used to convert the audio data into text data. This engine divides the audio signal into frames, each of which is analyzed using a phonological model and a language model. The input is compressed audio data, and the output is text data.

[0953] Step 4:

[0954] The server translates the converted text data. The text data is translated into the target language using a language translation engine via a cloud service. For example, the Japanese text "This is a wonderful project" is translated into the English text "This is a wonderful project." The input is Japanese text data, and the output is English text data.

[0955] Step 5:

[0956] The server's emotion engine analyzes the user's emotions from the voice input data. It identifies the user's emotional state based on the tone, pitch, speed, rhythm, etc. of the voice. This emotional information is reflected in the translated text data and subsequent voice data. The input is voice data, and the output is emotional information.

[0957] Step 6:

[0958] The server converts the translated text data into speech data using a speech synthesis engine. The translated text is converted into speech while taking into account phonology, morphology, and syntax. At this time, emotional information obtained from the emotion engine is reflected in the speech data. For example, a speech with an emotional expression of joy is generated. The input is the translated text data and emotional information, and the output is speech data that reflects the emotion.

[0959] Step 7:

[0960] The server sends the generated voice data and translated text data to the terminal. The voice data received by the terminal is played back using a speaker, and the translated text data is displayed on the screen. At this time, emotional information is also included in the display. The input is voice data and text data, and the output is voice and text display on the terminal.

[0961] In this way, each process in the system is generated from the user's input speech, and translated speech and text containing emotions are output in real time.

[0962] (Application example 2)

[0963] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0964] In autonomous vehicles, the language barrier makes it difficult for passengers to communicate in multiple languages. Another problem is that natural and rich communication cannot be achieved because passenger emotions are not reflected. Furthermore, it is necessary to appropriately reflect passenger emotions in real-time translation and voice playback.

[0965] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for transmitting voice input data to a data storage device, means for converting the voice input data into text data in the data storage device, and means for recognizing emotional states and reflecting emotional information in the voice data and text data. This makes it possible to translate the multilingual voices of passengers in an autonomous vehicle in real time and output natural voices and text that reflect emotions.

[0966] "Voice input data" refers to data obtained by converting the user's speech into digital form.

[0967] A "data storage device" is a device that temporarily stores voice input data and text data and performs the necessary processing.

[0968] "Text data" is data that is generated by analyzing voice input data and expressing it as a character string.

[0969] "Translation" is the process of converting text data expressed in one language into text data expressed in another language.

[0970] "Audio data" refers to data obtained by converting text data back into audio format.

[0971] The "emotional state" is a state in which the user is feeling some emotion, which is analyzed from the user's voice.

[0972] "Emotional information" is data that reflects the emotional state of the user and is information that is applied to translated text data or voice data.

[0973] A "receiving terminal" is a device that receives voice data and translated text data and plays or displays them.

[0974] A "voice recognition engine" is software or hardware for converting voice input data into a string of characters.

[0975] A "language translation engine" is software or hardware for translating text data expressed in one language into another language.

[0976] "Information and communication services" refers to infrastructure for utilizing various information processing functions provided on the cloud.

[0977] MODE FOR CARRYING OUT THE INVENTION

[0978] This invention is a multilingual real-time speech translation and emotion recognition system for passengers in autonomous vehicles. To realize this system, the following specific configuration is required.

[0979] First, the user's voice data is captured by a microphone installed in the vehicle. The captured voice data is converted into digital data and then compressed using lossless compression technology (e.g., Zlib). The compressed voice data is then temporarily stored in a data storage device in the vehicle.

[0980] The server then takes this voice data and converts it into text using a speech recognition engine (e.g., Google Speech-to-Text). The converted text data is then sent to a language translation engine (e.g., Google Cloud Translate) for translation into the specified target language. This converts the text data expressed in the original language into another language.

[0981] Furthermore, the server uses an emotion recognition engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotional state from the voice data, and the analyzed emotional information is reflected in the translated text data and the voice data that is later synthesized.

[0982] Next, based on the translated text data and emotion information, a speech synthesis engine (e.g., Amazon Polly) is used to generate voice data that reflects the user's emotions.

[0983] The generated voice data and translated text data are then transmitted to a receiving terminal inside the vehicle, which then displays the text data and emotion information on a display and simultaneously plays the voice data reflecting the emotion from a speaker.

[0984] Examples:

[0985] For example, if a user says in Japanese, "Good morning, please head to the airport," this voice is captured and sent to a data storage device. The server converts this voice into text data and translates it into English as "Good morning, please head to the airport." In addition, the emotion of the user's voice is recognized as "polite and calm," and a voice that reflects this emotion is generated. Finally, the in-car speaker plays "Good morning, please head to the airport" in a polite tone, and the same message is displayed on the in-car display.

[0986] Example prompt sentence:

[0987] If the user says "This is a wonderful project!" in a joyful voice, translate that into English and generate a joyful voice saying "This is a wonderful project!"

[0988] In this way, the system of the present invention realizes natural and rich multilingual communication within an autonomous vehicle.

[0989] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0990] Step 1:

[0991] The user speaks voice commands into a microphone installed inside the vehicle. The user's voice data is captured as input. This voice data is converted from an analog signal to a digital signal, making it easier to process (e.g., PCM format).

[0992] Step 2:

[0993] The audio data captured by the terminal is sent to the data storage device using lossless compression technology (e.g., Zlib). The input is digital audio data, and the output is compressed audio data. This improves data transmission efficiency.

[0994] Step 3:

[0995] The server decompresses the compressed audio data it receives and converts it into text data using a speech recognition engine (e.g., Google Speech-to-Text). The input is compressed audio data and the output is text data. The speech recognition engine analyzes the audio signal frame by frame based on an acoustic model and a language model to generate a string of characters.

[0996] Step 4:

[0997] The server sends the generated text data to a language translation engine (e.g., Google Cloud Translate) to translate it into the specified target language. The input is text data, and the output is text data in the target language. The translation engine uses a neural network to perform highly accurate translation.

[0998] Step 5:

[0999] The server uses an emotion recognition engine (e.g., IBM Watson Tone Analyzer) to analyze the emotional state from the voice data. The input is the original voice data, and the output is emotional information (e.g., joy, anger, surprise, etc.). The emotion engine analyzes the tone, pitch, speed, etc. of the voice to determine the emotional state.

[1000] Step 6:

[1001] The server generates speech data using a speech synthesis engine (e.g., Amazon Polly) based on the translated text data and emotional information. The input is the translated text data and emotional information, and the output is speech data that reflects the emotion. The speech synthesis engine generates natural-sounding speech while taking into account phonology, morphology, and syntax.

[1002] Step 7:

[1003] The server sends the generated voice data and translated text data to the terminal. The input is the voice data and translated text data, and the output is the data received by the terminal. An appropriate protocol (e.g. HTTP / 2) is used for data transmission.

[1004] Step 8:

[1005] The terminal receives voice data and plays it back through the vehicle's speakers, and displays the translated text data and emotional information on the display. The input is the received data, and the output is audiovisual information presentation. This enables users to communicate in multiple languages ​​in real time.

[1006] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1007] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1008] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1009] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1010] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1011] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1012] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1013] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1014] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1015] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1016] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1017] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1018] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1019] 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.

[1020] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1021] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1022] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1023] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1024] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1025] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1026] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1027] The following is further disclosed regarding the above embodiment.

[1028] (Claim 1)

[1029] means for obtaining voice input data;

[1030] means for transmitting voice input data to a server;

[1031] means for converting voice input data into text data in the server;

[1032] a means for translating the converted text data into another designated language;

[1033] means for converting the translated text data into audio data;

[1034] means for transmitting the voice data and the translated text data to the terminal;

[1035] means for playing the received voice data and displaying the translated text data;

[1036] A system including:

[1037] (Claim 2)

[1038] 2. The system according to claim 1, further comprising means for losslessly compressing voice input data and transmitting the data to the server.

[1039] (Claim 3)

[1040] 10. The system of claim 1, wherein the speech recognition engine and language translation engine are provided via a cloud service.

[1041] "Example 1"

[1042] (Claim 1)

[1043] means for obtaining voice input data;

[1044] a means for optimizing the acquired voice input data using lossless compression technology and transmitting the data to a server;

[1045] means for converting speech input data into text data using a physical model in the server;

[1046] a means for translating the converted text data into another specified language via a cloud service;

[1047] A means for converting the translated text data into voice data using a voice synthesis engine;

[1048] means for transmitting the voice data and the translated text data to the terminal;

[1049] means for playing the received voice data and displaying the translated text data;

[1050] A system including:

[1051] (Claim 2)

[1052] 10. The system of claim 1, wherein the speech recognition engine and language translation engine are provided via a cloud service.

[1053] (Claim 3)

[1054] 2. The system according to claim 1, further comprising a microphone used to facilitate meetings and face-to-face communication as a means for acquiring voice input data.

[1055] "Application Example 1"

[1056] (Claim 1)

[1057] means for obtaining voice input data;

[1058] means for transmitting voice input data to a server;

[1059] means for converting voice input data into text data in the server;

[1060] a means for translating the converted text data into another designated language;

[1061] means for converting the translated text data into audio data;

[1062] means for transmitting the voice data and the translated text data to the terminal;

[1063] means for playing the received voice data and displaying the translated text data;

[1064] means for generating a machine control command based on the text data and controlling a corresponding machine;

[1065] A system including:

[1066] (Claim 2)

[1067] 2. The system according to claim 1, further comprising means for losslessly compressing voice input data and transmitting the data to the server.

[1068] (Claim 3)

[1069] 10. The system of claim 1, wherein the speech recognition engine and language translation engine are provided via a cloud service.

[1070] "Example 2: Combining Emotion Engines"

[1071] (Claim 1)

[1072] means for obtaining voice input data;

[1073] means for transmitting speech input data to a processing device;

[1074] means for converting speech input data into text data in the processing device;

[1075] a means for translating the converted text data into another designated language;

[1076] means for converting the translated text data into audio data;

[1077] means for transmitting the voice data and the translated text data to the terminal;

[1078] means for playing the received voice data and displaying the translated text data;

[1079] A means for analyzing emotional information and reflecting it in corresponding data;

[1080] A system including:

[1081] (Claim 2)

[1082] 2. The system according to claim 1, further comprising means for losslessly compressing the voice input data and transmitting the compressed data to the processing device.

[1083] (Claim 3)

[1084] 10. The system of claim 1, wherein the speech recognition engine and the language translation engine are provided via a remote service.

[1085] "Application example 2 when combining emotion engines"

[1086] (Claim 1)

[1087] means for obtaining voice input data;

[1088] means for transmitting voice input data to a data storage device;

[1089] means for converting voice input data into text data in a data storage device;

[1090] a means for translating the converted text data into another designated language;

[1091] means for converting the translated text data into audio data;

[1092] means for recognizing an emotional state and reflecting the emotional information in the voice data and text data;

[1093] means for transmitting the voice data and the translated text data to a receiving terminal;

[1094] means for playing the received voice data and displaying the translated text data and emotion information;

[1095] A system including:

[1096] (Claim 2)

[1097] 2. The system of claim 1, further comprising means for losslessly compressing the audio input data before transmitting it to the data storage device.

[1098] (Claim 3)

[1099] 2. The system according to claim 1, wherein the speech recognition engine and the language translation engine are provided via an information and communication service. [Explanation of symbols]

[1100] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for obtaining voice input data; means for transmitting voice input data to a server; means for converting voice input data into text data in the server; a means for translating the converted text data into another designated language; means for converting the translated text data into audio data; means for transmitting the voice data and the translated text data to the terminal; means for playing the received voice data and displaying the translated text data; A system including:

2. 2. The system according to claim 1, further comprising means for losslessly compressing the voice input data and transmitting the compressed data to the server.

3. 10. The system of claim 1, wherein the speech recognition engine and the language translation engine are provided via a cloud service.

Images