System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting speech contents from a user; means for transmitting the collected speech contents as voice data; means for converting the received voice data into text; means for complementing the contents of the converted text; means for translating the complemented text into another language; means for converting the translated text into voice data; means for transmitting the converted voice data to a partner's terminal; and means for reproducing the received voice data on the partner's terminal.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In recent years, international business and academic exchanges have become increasingly active, but language barriers have become a major obstacle. In particular, we are seeing situations where talented people who are unsure of their foreign language skills are unable to fully demonstrate their abilities at international conferences and exchanges. In Japan, there is a significant lack of English proficiency, which is a major obstacle to overseas expansion and international exchange. Therefore, there is a need for technology that can overcome language barriers and enable communication in one's native language. [Means for solving the problem]

[0005] The present invention provides a system including means for collecting user utterances, means for transmitting the collected utterances as voice data, means for converting the received voice data into text, means for completing the converted text, means for translating the completed text into another language, means for converting the translated text into voice data, means for transmitting the converted voice data to a remote terminal, and means for playing the received voice data at the remote terminal. This system allows participants to feel as if they are participating in a conference in only their own language, enabling smooth communication across language barriers. By exchanging voice and text data in real time between a server and terminals via the Internet, this system allows participants from 83 countries to feel as if they are conversing in their own language.

[0006] A "user" is an individual or group that uses the system and is the entity that inputs speech into the system.

[0007] "Speech content" is voice data such as information, opinions, questions, etc. expressed in language by the user.

[0008] "Voice data" refers to data that represents the user's speech in digital form and is processed within the system.

[0009] A "terminal" is a device that allows a user to input voice and communicate with the system, such as a smartphone or computer.

[0010] A "server" is a central computer system that receives, processes, converts, and transmits audio data.

[0011] "Text" is the representation of audio data as written information, which is converted by a generative AI model.

[0012] "Content completion" refers to the use of a generative AI model to supplement missing information in the context of the converted text, resulting in natural-sounding sentences.

[0013] "Translation" is the process of converting the completed text into another language, enabling communication between different languages.

[0014] "Speech synthesis" is a technology that converts text data into voice data, and outputs text translated into other languages ​​as voice again. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0016] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0018] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0019] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0020] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0021] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0023] [First embodiment]

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

[0025] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0026] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0027] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0028] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0032] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0033] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0034] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0035] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0036] The system of the present invention translates what a user says in their own language into another language in real time and provides the translated speech to the other party. This enables smooth communication that transcends language barriers at international conferences, etc. The following describes in detail the embodiments of the present invention.

[0037] Collection of speech content

[0038] The user begins speaking in their own language. For example, the user is Japanese and says, "Hello, how is the project going?" This speech is collected by the user's device (smartphone or computer) via a microphone.

[0039] Sending audio data

[0040] The device transmits the collected voice data in real time to a server via the Internet. The voice data contains the user's speech in digital form.

[0041] Speech recognition and text conversion

[0042] The server receives voice data via the internet and inputs it into the generative AI model to convert the voice to text. At this time, speech recognition technology is used to accurately convert what the user has said into text format.

[0043] Content Complement

[0044] The server uses a generative AI model to analyze the context of the converted text and fill in any missing or ambiguous information. For example, if the generated text is "Hello, how's the project going?", it will be completed to make the meaning more specific: "Hello, regarding the project progress, what's the status?"

[0045] translation

[0046] The server translates the completed text data into the target language using a translation function. For example, the server translates the completed Japanese text into English, converting it into the English text "Hello, regarding the project's progress, how is it going?"

[0047] Speech synthesis

[0048] The server converts the translated text into speech data using speech synthesis technology, which prepares the translated content as speech output. For example, an English voice is generated based on the English text "Hello, regarding the project's progress, how is it going?"

[0049] Sending and playing audio data

[0050] The generated voice data is then sent from the server to the other party's device, which receives the voice data and plays it back through its speaker. In this way, even if the other user does not understand Japanese, they can understand what the Japanese person is saying through the English voice.

[0051] As a concrete example, consider a conversation between Japanese user A and American user B, both of whom are participating in an international conference. User A speaks in Japanese, which is played back as English audio on User B's device. User B also speaks in English, which is played back as Japanese audio on User A's device. This allows the two parties to communicate smoothly without feeling any language barrier.

[0052] The system of the present invention significantly improves international communication by processing users' speech in real time, converting it into other languages, and providing it to the other party. By exchanging voice and text data between the server and the terminals via the Internet, participants of multiple nationalities can experience a conversation as if they were speaking only in their own native language.

[0053] The processing flow will be explained below.

[0054] Step 1:

[0055] The user speaks in their own language. For example, Japanese user A says, "Hello, how is the project going?"

[0056] Step 2:

[0057] The terminal (User A's device) records the speech using a microphone and captures it as voice data, which is then temporarily stored in the terminal in digital format.

[0058] Step 3:

[0059] The device transmits the collected voice data to a server via the Internet in real time. The transmitted data includes the voice data as well as the user ID and device information.

[0060] Step 4:

[0061] The server inputs the received voice data into the generative AI model and converts it into text using speech recognition technology. This conversion process generates the text, "Hello, how is the project going?"

[0062] Step 5:

[0063] The server completes the generated text using a context analysis algorithm. For example, the text "Hello, how is the project going?" is completed as "Hello, regarding the project progress, what is the status?"

[0064] Step 6:

[0065] The server translates the completed text into the target language, for example translating the completed Japanese text into English to generate the English text "Hello, regarding the project's progress, how is it going?"

[0066] Step 7:

[0067] The server converts the translated text into speech data using speech synthesis technology, generating the English speech data "Hello, regarding the project's progress, how is it going?"

[0068] Step 8:

[0069] The server transmits the generated translated speech data via the Internet to the other party's terminal (user B's device). The transmitted data includes the translated speech data.

[0070] Step 9:

[0071] The terminal (user B's device) plays the received voice data through the device's speaker. User B can hear the English voice saying "Hello, regarding the project's progress, how is it going?"

[0072] Step 10:

[0073] User B (American) says in English, "The project is progressing well, thank you for asking."

[0074] Step 11:

[0075] The terminal (user B's device) records the speech using a microphone and captures the audio data in real time.

[0076] Step 12:

[0077] The terminal transmits the collected voice data to a server via the Internet in real time.

[0078] Step 13:

[0079] The server inputs the received voice data into the generative AI model and converts it into text using speech recognition technology. For example, the generated text might read, "The project is progressing well, thank you for asking."

[0080] Step 14:

[0081] The server uses a context analysis algorithm to complete the generated text, for example, "The project is progressing well, thank you for asking about the project's status."

[0082] Step 15:

[0083] The server translates the completed text into the target language, for example, translating the completed English text into Japanese to generate Japanese text such as "The project is progressing smoothly. Thank you for your inquiry."

[0084] Step 16:

[0085] The server converts the translated Japanese text into voice data using speech synthesis technology, generating Japanese voice data that says, "The project is progressing smoothly. Thank you for your inquiry."

[0086] Step 17:

[0087] The server transmits the generated translated speech data to the other party's terminal (user A's device) via the Internet.

[0088] Step 18:

[0089] The terminal (user A's device) plays the received voice data through the device's speaker. User A can hear the voice in his / her native language saying, "The project is progressing smoothly. Thank you for your inquiry."

[0090] Example 1

[0091] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0092] Conventional real-time translation systems have faced problems with the accuracy of translation of spoken content and a lack of context-based completion. Furthermore, when communicating between different languages, each step of speech recognition, translation, and speech synthesis is fragmented, which can impair real-time performance. There is a need to solve these problems and realize smoother, more accurate international communication.

[0093] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0094] In this invention, the server includes means for collecting user utterances, means for transmitting the collected utterances as voice data, means for converting the received voice data into text, means for completing the converted text using a generative AI model, means for translating the completed text into another language, means for converting the translated text into voice data, means for transmitting the generated voice data to a partner device, and means for playing the received voice data on the partner device. This enables the user's utterances to be translated accurately between different languages ​​in real time and for natural completion based on context.

[0095] A "user" is a person who uses the system of the present invention to input speech content.

[0096] "Speech content" refers to words and sentences spoken by the user, and is the speech data to be translated.

[0097] "Voice data" is data that expresses the contents of a user's speech in digital form.

[0098] "Text data" refers to data that has been converted from voice data into character information using voice recognition technology.

[0099] A "generative AI model" is a model that uses artificial intelligence technology to complement the content of text data, for example, natural language processing technology.

[0100] "Completion" is the process of adding contextual information to incomplete text data to make it more meaningful.

[0101] The "translation function" is a function for converting text data written in one language into another language.

[0102] "Speech synthesis technology" is a technology that generates voice data based on text data.

[0103] The "partner's device" is a receiving device used by the other party who is having a conversation with the user, and has the function of reproducing voice data.

[0104] The "collection means" is a means for acquiring the contents of a user's speech using a device such as a microphone.

[0105] The "transmission means" is a means for transmitting collected or generated voice data via the Internet or the like.

[0106] The "conversion means" is a means for converting voice data into text data and text data into voice data.

[0107] "Completion means" refers to a means of complementing the content of text data using a generative AI model.

[0108] The "playback means" is a function for playing back the audio data received by the other device from a speaker or the like.

[0109] MODE FOR CARRYING OUT THE INVENTION

[0110] The present invention provides a system that translates speech in a user's own language into another language in real time and provides the translated speech to the other party. This system enables smooth communication across language barriers in situations such as international conferences. Specific embodiments for implementing the present invention will be described below, specifying the hardware and software used.

[0111] This system consists of three main elements: the user, the terminal, and the server. The roles and specific operations of each are explained in detail below.

[0112] User Roles

[0113] Users speak in their own language. For example, a Japanese user might say, "Hello, how's the project going?" This speech is collected as voice data through a microphone installed on a device such as a smartphone or PC.

[0114] Device Role

[0115] The device sends the voice data collected from the user to a server via the Internet. The voice data is encoded in a digital format (e.g., PCM format). WebSocket or similar is often used as a communication protocol to maintain real-time performance.

[0116] Server Roles

[0117] The server performs the following sequence of operations:

[0118] 1. Speech Recognition:

[0119] The server receives the voice data and converts it into text data using a speech recognition engine (e.g., Google Speech-to-Text API). This conversion converts the user's speech into text.

[0120] 2. Content Complement:

[0121] The converted text data is input into a generative AI model (e.g., OpenAI GPT-3). The server uses this generative AI model to analyze the context of the text and fill in missing or ambiguous information. For example, for the text "Hello, how is the project progressing?", it generates a more specific completion text: "Hello, regarding the progress of the project, what is the status?"

[0122] 3. Translation:

[0123] The completed text data is sent to a translation engine (e.g., DeepL or Google Translate API). The server uses this engine to translate the text data into another language. For example, translating Japanese text into English yields the English text "Hello, regarding the project's progress, how is it going?"

[0124] 4. Speech synthesis:

[0125] The translated text is input into a speech synthesis engine (e.g., IBM Watson Text-to-Speech or Amazon Polly) and converted into audio data, which generates natural-sounding audio data based on the translated English text.

[0126] The server sends the generated voice data to the other device. The other device has the function to play the received voice data and outputs the voice from a speaker. This allows the other user to understand what the user is saying through the English voice, even if they do not understand Japanese.

[0127] Specific examples

[0128] As a concrete example, consider a conversation between Japanese user A and American user B, both of whom are attending an international conference. User A says in Japanese, "Hello, how is the project progressing?", which is played back in English on User B's device. User B also says in English, "The project is progressing well, thank you.", which is played back in Japanese on User A's device. This allows the two parties to communicate smoothly without feeling any language barrier.

[0129] Prompt Sentence Examples

[0130] "Please translate the following Japanese text into natural English and synthesize it into speech: 'Hello, how's the project going?'"

[0131] The system of the present invention combines real-time speech recognition, text completion, translation, and speech synthesis technologies over the Internet to significantly improve international communication.

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

[0133] Program processing flow

[0134] Step 1:

[0135] The user begins speaking in their own language. The speech is collected as voice data through the device's microphone. The input is the user's speech, and the output is the collected voice data. The device's specific operation is to convert the analog voice signal via the microphone into digital voice data and store it in its internal memory.

[0136] Step 2:

[0137] The device sends the collected voice data to the server via the Internet. The input is the voice data collected in step 1, and the output is the transmitted voice data. Specifically, the device prepares to transmit the voice data to the server using a protocol such as WebSocket, and then begins transmission.

[0138] Step 3:

[0139] The server calls a speech recognition engine to convert the received voice data into text data. For example, it uses the Google Speech-to-Text API. The input is the voice data sent from the device, and the output is text data. Specifically, the server sends the voice data to the API and receives the returned text data.

[0140] Step 4:

[0141] The server uses a generative AI model to complete the content of the received text data. For example, OpenAI GPT-3 is used as this model. The input is the text data converted in step 3, and the output is the completed text data. Specifically, the server inputs the text data into the generative AI model and receives the completed text.

[0142] Step 5:

[0143] The server sends the completed text data to a multilingual translation engine to translate it into other languages. For example, it uses DeepL or Google Translate API. The input is the text data completed in step 4, and the output is the translated text data. Specifically, the server sends the text data to the translation engine and receives the translated results.

[0144] Step 6:

[0145] The server inputs the translated text data into a speech synthesis engine and converts it into voice data. For example, IBM Watson Text-to-Speech or Amazon Polly can be used. The input is the text data translated in step 5, and the output is the generated voice data. Specifically, the server sends the text data to the speech synthesis engine and receives the generated voice data.

[0146] Step 7:

[0147] The server transmits the generated voice data to the other party's terminal via the Internet. The input is the voice data generated in step 6, and the output is the transmitted voice data. Specifically, the server performs the transmission process using a protocol for transmitting voice data to the other party's terminal.

[0148] Step 8:

[0149] The other device plays the received voice data. The input is the voice data sent from the server, and the output is the voice played through the speaker. Specifically, the device decodes the received voice data and plays it back through the speaker. This allows the other user to understand what the user is saying.

[0150] (Application example 1)

[0151] 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."

[0152] Establishing real-time communication between users who speak different languages ​​is a major challenge, especially in international business and tourism. There is also a demand for seamless audio and visual information delivery using devices such as smart glasses. Existing technologies struggle to fully meet these demands, particularly in terms of accuracy and speed. Therefore, a new system is needed to provide real-time translated audio and visual information between users who speak different languages.

[0153] 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.

[0154] In this invention, the server includes: means for collecting speech content from users; means for transmitting the collected speech content as voice data; means for converting the received voice data into text; means for completing the converted text; means for translating the completed text into another language; means for converting the translated text into voice data; means for transmitting the converted voice data to a remote terminal; means for playing the received voice data at the remote terminal; means for realizing real-time language translation between a service provider and a customer and providing voice and visual information through the smart glasses; and means for detecting the user's speech using a microphone and speaker of the smart glasses and providing voice feedback to the remote terminal in the translated language, thereby enabling real-time translation of voice and visual information between users speaking different languages.

[0155] "User" refers to a person who uses the system to translate spoken content.

[0156] "Speech content" refers to what the user says in natural language.

[0157] "Means for collecting" refers to the function of acquiring the user's speech content as voice data.

[0158] "Transmitting means" refers to the function of transmitting collected voice data to an external device such as a server.

[0159] "Means for converting to text" refers to a function for converting received voice data into character data.

[0160] "Means for supplementing" refers to the function of adding supplementary information based on the context of the converted text content.

[0161] "Means for translating" refers to a function for translating the completed text into a specified other language.

[0162] "Means for converting into voice data" refers to the function of converting the translated text into data for voice output.

[0163] "Means for transmitting" refers to the function of transferring voice data to the other party's terminal.

[0164] "Means for playing" refers to the function of actually outputting the audio data received on the other party's terminal as audio.

[0165] A "service provider" refers to the person who interacts with customers at a store or facility.

[0166] "Customer" means a person who uses the System for the purpose of communicating with a Service Provider.

[0167] "Smart glasses" are glasses-type devices that utilize built-in microphones and speakers to facilitate interaction between the user and the system.

[0168] "Voice feedback" refers to the function of translating a user's speech and providing it as audio.

[0169] The system of the present invention translates what a user says in their own language into another language in real time and presents the translated speech to the other party as audio, thereby realizing smooth communication between users who speak different languages. Specific embodiments for carrying out the present invention will be described below.

[0170] A means of generating a program

[0171] The programs running on the server use the following software:

[0172] speech_recognition: Collects speech from the microphone and converts it into text.

[0173] googletrans: Translate the converted text into other languages.

[0174] gtts: Converts translated text into audio data.

[0175] playsound: Plays the generated sound data.

[0176] Hardware

[0177] The system uses smart glasses as its primary device, which includes the following hardware:

[0178] Microphone: Collects user utterances.

[0179] Speaker: Plays the translated audio.

[0180] Display: Visually display the completed text information.

[0181] Data processing and calculation

[0182] The server processes the data in the following steps:

[0183] 1. Voice data collection: The user's speech is collected through the microphone of the smart glasses. The collected voice data is temporarily stored in the memory of the smart glasses.

[0184] 2. Sending voice data: The smart glasses send the collected voice data to the server.

[0185] 3. Speech recognition and text conversion: The server converts the voice data into text using "speech_recognition".

[0186] 4. Content completion: Using a "generative AI model," the context of the converted text is analyzed and information is completed as needed.

[0187] 5. Text translation: The server translates the completed text into other languages ​​using "googletrans".

[0188] 6. Speech synthesis: The translated text is converted into voice data by "gtts".

[0189] 7. Sending voice data: The converted voice data is sent to the speaker of the smart glasses.

[0190] 8. Playback of audio data: The translated audio will be played through the smart glasses' speaker.

[0191] Specific examples

[0192] For example, if a Japanese store clerk says "Here is the product," the microphone in the smart glasses collects the voice. The voice data is sent to the server and converted into Japanese text "Here is the product." The text is then translated into English, generating the English text "Here is the product." This English text is converted into speech and played back from the speaker in the customer's smart glasses as "Here is the product." In this way, the foreign customer can understand the explanation of the store clerk.

[0193] Prompt Sentence Examples

[0194] User: Says, "Hi, how's the project going?"

[0195] Server: Identify the language and provide the translated results in text format.

[0196] Output: "Hello, regarding the project's progress, how is it going?"

[0197] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0198] Step 1:

[0199] A user puts on the smart glasses and starts speaking. The microphone in the smart glasses collects the speech as voice data. The collected voice data is temporarily stored in the memory in the smart glasses.

[0200] Input: User's speech

[0201] Output: Audio data in smart glasses

[0202] Step 2:

[0203] The device (smart glasses) transmits the collected voice data to a server in real time, using an internet connection to transfer the data.

[0204] Input: Audio data in smart glasses

[0205] Output: Audio data sent to the server

[0206] Step 3:

[0207] The server converts the received voice data into text using the "speech_recognition" library. Through this conversion, the user's speech content is obtained as text data.

[0208] Input: Audio data sent to the server

[0209] Output: Converted text data

[0210] Step 4:

[0211] The server then uses a generative AI model to analyze the converted text and complete the information based on the context, thereby fleshing out any ambiguities or missing information.

[0212] Input: Converted text data

[0213] Output: Completed text data

[0214] Step 5:

[0215] The server uses "googletrans" to translate the completed text into the specified language, resulting in the user's speech being expressed in another language.

[0216] Input: Completed text data

[0217] Output: Translated text data

[0218] Step 6:

[0219] The server converts the translated text into audio data using "gtts" (Google Text-to-Speech), which then creates a playable audio format.

[0220] Input: Translated text data

[0221] Output: Generated audio data

[0222] Step 7:

[0223] The server then transmits the generated audio data to the smart glasses, again using an internet connection to transfer the data.

[0224] Input: Generated audio data

[0225] Output: Audio data sent to the smart glasses

[0226] Step 8:

[0227] The device (smart glasses) plays the received audio data through a speaker, so that the user's speech is provided as audio in the translated language.

[0228] Input: Audio data transmitted to smart glasses

[0229] Output: Playback of the translated audio

[0230] In this way, the system of the present invention translates the user's speech in real time and provides it as audio, thereby realizing smooth communication between users who speak different languages.

[0231] 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.

[0232] The system of the present invention translates what a user says in their own language into another language in real time, and also has the function of recognizing the user's emotions and reflecting them in the translation. By recognizing the user's emotions, more natural and emotional communication is realized. The following describes in detail an embodiment of the present invention.

[0233] Collection of speech content

[0234] The user speaks in their own language. For example, Japanese user A says, "Hello, how is the project going?" The content of this speech is collected by the user's device (smartphone or computer) through a microphone.

[0235] Sending audio data

[0236] The device transmits the collected voice data in real time to a server via the Internet. The voice data contains the user's speech in digital form.

[0237] Speech recognition and text conversion

[0238] The server inputs the voice data received via the internet into the generative AI model and converts the voice to text. For example, the generated text is "Hello, how is the project going?"

[0239] emotion recognition

[0240] The server uses the voice and text data to recognize the user's emotions through an emotion engine. For example, the server can recognize emotions such as "interest" and "concern" from User A's tone of voice and speaking style.

[0241] Content Complement

[0242] The server completes the generated text using a context analysis algorithm. For example, the text "Hello, how is the project going?" is completed as "Hello, regarding the project progress, what is the status?"

[0243] translation

[0244] The server translates the completed text into the target language and reflects the recognized sentiment in the translation result. For example, when translating the completed Japanese text into English and generating the English text "Hello, regarding the project's progress, how is it going?", the server reflects that User A is "interested" or "concerned."

[0245] Speech synthesis

[0246] The server converts the translated text into speech data using speech synthesis technology, generating English speech data such as "Hello, regarding the project's progress, how is it going?" with emotional intonation.

[0247] Sending and playing audio data

[0248] The generated voice data is again sent from the server to the other device (User B's device). The other device receives this voice data and plays it through the speaker. User B can hear the English voice "Hello, regarding the project's progress, how is it going?" with the emotion reflected.

[0249] As a concrete example, consider a conversation between Japanese user A and American user B, both of whom are participating in an international conference. User A speaks in Japanese, which is played back as English audio on User B's device, with User A's emotions also being conveyed. Similarly, User B speaks in English, which is played back as Japanese audio on User A's device, with User A's emotions also being reflected. This allows both parties to communicate their emotions accurately without feeling a language barrier, resulting in more natural and effective communication.

[0250] The system of the present invention recognizes the user's emotions throughout the entire process, from voice data collection, transmission, text conversion, content completion, translation, voice synthesis, and finally transmission and playback of the voice data, and reflects these emotions in the translation results, thereby significantly improving international communication.

[0251] The processing flow will be explained below.

[0252] Step 1:

[0253] The user speaks in their own language. For example, Japanese user A says, "Hello, how is the project going?"

[0254] Step 2:

[0255] The terminal (User A's device) records the speech using a microphone and captures it as voice data, which is then temporarily stored in the terminal in digital format.

[0256] Step 3:

[0257] The device transmits the collected voice data to a server via the Internet in real time. The transmitted data includes the voice data as well as the user ID and device information.

[0258] Step 4:

[0259] The server inputs the received voice data into the generative AI model and converts it into text using speech recognition technology. This conversion process generates the text, "Hello, how is the project going?"

[0260] Step 5:

[0261] The server uses the voice and text data to recognize the user's emotions through an emotion engine. For example, the server can recognize emotions such as "interest" and "concern" from User A's tone of voice and speaking style.

[0262] Step 6:

[0263] The server completes the generated text using a context analysis algorithm. For example, the text "Hello, how is the project going?" is completed as "Hello, regarding the project progress, what is the status?"

[0264] Step 7:

[0265] The server translates the completed text into the target language and reflects the recognized sentiment in the translation result. For example, when translating the completed Japanese text into English and generating the English text "Hello, regarding the project's progress, how is it going?", the server reflects that User A is "interested" or "concerned."

[0266] Step 8:

[0267] The server converts the translated text into speech data using speech synthesis technology, generating English speech data such as "Hello, regarding the project's progress, how is it going?" with emotional intonation.

[0268] Step 9:

[0269] The server transmits the generated translated speech data via the Internet to the other party's terminal (user B's device). The transmitted data includes the translated speech data.

[0270] Step 10:

[0271] The device (User B's device) plays the received voice data through the device's speaker. User B can hear the English voice "Hello, regarding the project's progress, how is it going?" with the emotion reflected.

[0272] Step 11:

[0273] User B (American) says in English, "The project is progressing well, thank you for asking."

[0274] Step 12:

[0275] The device (User B's device) records the speech with a microphone and captures the audio data in real time, which is then temporarily stored in the device in digital format.

[0276] Step 13:

[0277] The device transmits the collected voice data to a server via the Internet in real time. The transmitted data includes the voice data as well as the user ID and device information.

[0278] Step 14:

[0279] The server inputs the received voice data into the generative AI model and converts it into text using speech recognition technology. For example, the generated text might read, "The project is progressing well, thank you for asking."

[0280] Step 15:

[0281] The server uses the voice and text data to recognize the user's emotions using an emotion engine. For example, emotions such as "gratitude" and "relief" can be recognized from User B's tone of voice and speaking style.

[0282] Step 16:

[0283] The server uses a context analysis algorithm to complete the generated text. For example, the text "The project is progressing well, thank you for asking" is completed as "The project is progressing well, thank you for asking about the project's status."

[0284] Step 17:

[0285] The server translates the completed text into the target language and reflects the recognized emotions in the translation result. For example, when translating the completed English text into Japanese and generating the Japanese text "The project is progressing smoothly. Thank you for your inquiry," the "gratitude" and "relief" that User B feels are reflected.

[0286] Step 18:

[0287] The server then converts the translated Japanese text into speech data using speech synthesis technology, generating Japanese speech data that says, "The project is progressing smoothly. Thank you for your inquiry," with intonation that reflects the emotion.

[0288] Step 19:

[0289] The server transmits the generated translated speech data to the other party's terminal (user A's device) via the Internet. The transmitted data includes the translated speech data.

[0290] Step 20:

[0291] The terminal (user A's device) plays the received voice data through the device's speaker. User A can hear the voice in his / her native language saying "The project is progressing smoothly. Thank you for your inquiry." with the emotion reflected.

[0292] Example 2

[0293] 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."

[0294] In today's global society, people who speak different languages ​​need to communicate smoothly. However, while conventional translation systems can accurately translate spoken content, they struggle to convey the speaker's emotions appropriately. As a result, misunderstandings and misrecognitions are likely to occur, hindering natural communication. Furthermore, they often fail to properly complement the spoken content by taking into account the context of the spoken content, making it impossible to convey the speaker's full intention.

[0295] 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.

[0296] In this invention, the server includes a means for recognizing the user's emotions, a means for completing the content of the converted text, and a means for converting the translated text into voice data. This not only ensures accurate translation of the spoken content, but also enables the speaker's emotions to be conveyed to the other party in a manner that appropriately reflects their intention. Furthermore, context-based content completion allows the speaker to fully convey their intention, resulting in more natural and effective communication.

[0297] "User" refers to a person who uses the system to translate spoken content into another language.

[0298] The term "terminal" refers to an electronic device that collects the user's speech, transmits the voice data to a server, and plays back the voice data transmitted from the server.

[0299] "Server" refers to a central control device that processes voice data sent from a user's terminal and performs translation and voice synthesis.

[0300] "Speech content" refers to words and sentences spoken by a user in their own language.

[0301] "Voice data" refers to data that represents the contents of a user's speech in digital form.

[0302] "Text" refers to data that has been converted from audio data into character information.

[0303] "Emotion" refers to the emotional elements (e.g., interest, concern, joy, etc.) contained in the user's utterance.

[0304] "Content completion" refers to the process of analyzing the context of the spoken content and completing incomplete parts in a natural way.

[0305] "Translation" refers to the process of converting text expressed in one language into another language.

[0306] "Speech synthesis" refers to the process of converting text data into voice data.

[0307] "Emotion recognition" refers to the process of analyzing and identifying emotional elements from a user's speech.

[0308] "Real-time" means that processing occurs with little to no delay.

[0309] The system of the present invention translates what a user says in their own language into another language in real time, and further recognizes the user's emotions and reflects them in the translation, thereby realizing natural and emotional communication. A specific embodiment of this system will be described below.

[0310] Collection of speech content

[0311] The user speaks in their own language. For example, a Japanese user might say, "Hello, how's the project going?" This speech is collected by the user's device (smartphone or computer) via a microphone.

[0312] Sending audio data

[0313] The device transmits the collected voice data to a server over the Internet in real time. The voice data contains the user's speech in digital form. The device transmits the voice data using a secure communication protocol (e.g., HTTPS).

[0314] Speech recognition and text conversion

[0315] The server inputs the voice data received via the internet into a generative AI model and converts the voice into text. For example, the generative AI model used is the Google Cloud Speech-to-Text API, and generates the text "Hello, how is the project progressing?" from Japanese voice data.

[0316] emotion recognition

[0317] The server uses voice and text data to recognize the user's emotions through an emotion engine. For example, the emotion engine used is IBM Watson Tone Analyzer, which recognizes emotions such as "interest" and "concern" from the user's tone of voice and speaking style.

[0318] Content Complement

[0319] The server uses a context analysis algorithm to complete the generated text. For example, the context analysis algorithm used is OpenAI GPT-3, and the text "Hello, how is the project going?" is completed as "Hello, regarding the progress of the project, what is the status?"

[0320] translation

[0321] The server translates the completed text into the target language and reflects the recognized sentiment in the translation results. For example, the translation engine used is the Google Translate API, and the completed Japanese text is translated into English to generate the English text "Hello, regarding the project's progress, how is it going?". In this case, the user's "interest" and "concern" are reflected in the translation results.

[0322] Speech synthesis

[0323] The server converts the translated text into voice data using Amazon Polly speech synthesis technology. The generated voice data is an English voice saying "Hello, regarding the project's progress, how is it going?" with intonation that reflects the emotion.

[0324] Sending and playing audio data

[0325] The generated voice data is again sent from the server to the other device (User B's device). The other device receives this voice data and plays it through the speaker. User B can hear the English voice saying "Hello, regarding the project's progress, how is it going?" with the emotion reflected.

[0326] As a concrete example, consider a conversation between a Japanese user and an American user participating in an international conference. The Japanese user speaks in Japanese, which is played back as English speech on the American user's device, with the Japanese user's emotions also being conveyed. Similarly, the American user speaks in English, which is played back as Japanese speech on the Japanese user's device, with the Japanese user's emotions also being reflected.

[0327] Prompt Sentence Examples

[0328] "A user says, 'Hello, how's the project going?' in Japanese. Please translate this utterance into English and convert it into speech data that reflects the sentiment."

[0329] This system will significantly improve international communication, enabling more natural and emotional interactions.

[0330] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0331] Step 1:

[0332] The user speaks in their own language. The user's voice is collected by the device as input. For example, a Japanese user might say, "Hello, how's the project going?". Specifically, the device's microphone collects the voice in digital form. Digital voice data is generated as output.

[0333] Step 2:

[0334] The device transmits the collected audio data in real time to a server in uncompressed or compressed format. Digital audio data is input. Specifically, the device transmits the audio data using a secure communication protocol such as HTTPS. The audio data arrives at the server as output.

[0335] Step 3:

[0336] The server inputs the received voice data into a generative AI model and converts the voice to text. The voice data received on the server side exists as input. Specifically, it uses the Google Cloud Speech-to-Text API to convert it into text data such as "Hello, how is the project progressing?". Text data is generated as output.

[0337] Step 4:

[0338] The server uses voice and text data to recognize the user's emotions through an emotion engine. Converted text data and voice data are used as input. Specifically, the server uses IBM Watson Tone Analyzer to identify emotions such as "interest" or "concern" from the user's tone of voice and choice of words. Emotional data is generated as output.

[0339] Step 5:

[0340] The server analyzes the context of the generated text data and completes the content. Emotion data and text data are input. Specifically, OpenAI GPT-3 is used to complete the text "Hello, how is the project progressing?" to "Hello, regarding the progress of the project, what is the status?". The completed text data is generated as output.

[0341] Step 6:

[0342] The server translates the completed text into the target language and also reflects the recognized emotion in the translation result. The input is the completed text data and emotion data. Specifically, it uses the Google Translate API to translate the completed text "Hello, regarding the project's progress, how is it going?" into English and reflects the emotion to generate "Hello, regarding the project's progress, how is it going?" The translated text data is generated as the output.

[0343] Step 7:

[0344] The server converts the translated text into speech data using speech synthesis technology. The inputs are the translated text data and emotion data. Specifically, Amazon Polly is used to generate English speech data saying "Hello, regarding the project's progress, how is it going?". This speech data is given an intonation that reflects the emotion. English speech data is generated as output.

[0345] Step 8:

[0346] The server sends the generated voice data to the other party's device. The generated voice data exists as input. Specifically, the voice data is sent to User B's device via the Internet. The voice data arrives at the other party's device as output.

[0347] Step 9:

[0348] The other device plays the received audio data. The input is English audio data from the server. Specifically, the device's speaker plays the English audio, "Hello, regarding the project's progress, how is it going?". User B can hear the audio as output.

[0349] This detailed processing allows users who speak different languages ​​to communicate accurately and with rich emotion.

[0350] (Application example 2)

[0351] 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."

[0352] Currently, in order for autonomous vehicles to communicate smoothly between passengers and vehicle systems, it is necessary not only to recognize instructions but also to communicate seamlessly in various languages. It is also necessary to simultaneously recognize passenger emotions and respond appropriately to those emotions to provide a safer and more comfortable travel experience. However, current systems are unable to effectively support multiple languages ​​and recognize emotions, resulting in reduced passenger satisfaction and safety.

[0353] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting utterances from users, means for transmitting the collected utterances as voice data, means for converting the received voice data into text, means for completing the converted text, means for translating the completed text into another language, means for converting the translated text into voice data, means for transmitting the converted voice data to the other party's terminal, means for performing emotion recognition, means for generating voice data based on the recognized emotion, means for playing the received voice data at the other party's terminal, means for giving instructions to the vehicle system or setting a destination based on the passenger's utterance, and means for recognizing the emotion of the utterance and performing a corresponding action. This enables natural communication that recognizes emotions in multiple languages, thereby realizing safe and comfortable operation of autonomous vehicles.

[0354] The "means for collecting speech content from the user" refers to a device or function for recognizing the speech uttered by the user and acquiring it as digital data.

[0355] "Means for transmitting collected speech content as voice data" refers to a mechanism for sending acquired voice data to a server via a communication network such as the Internet.

[0356] The "means for converting received voice data into text" refers to a system that analyzes voice data on the server side and converts it into a corresponding text format.

[0357] "Means for completing the content of the converted text" refers to technologies and algorithms that analyze the context of the converted text, add information as needed, and arrange the sentences in a natural format.

[0358] The "means for translating the completed text into another language" is an automatic translation technology for converting the completed text into another specified language.

[0359] "Means for converting translated text into speech data" refers to technology or devices for converting translated text into synthetic speech.

[0360] The "means for transmitting the converted voice data to the other terminal" is the communication means used to retransmit the generated voice data to the target device.

[0361] "Means for emotion recognition" refers to technologies and algorithms that analyze and recognize a speaker's emotional state from audio data or text data.

[0362] The "means for generating voice data based on recognized emotions" refers to a technology that adjusts the intonation and atmosphere of the synthesized voice by reflecting the recognized emotional information.

[0363] The "means for playing back the received audio data on the other party's terminal" is a function for playing back the transmitted audio data on the device.

[0364] "Means for giving instructions to vehicle systems and setting destinations based on passenger speech" refers to technology that recognizes what passengers say and operates the vehicle's navigation system and other operations based on that speech.

[0365] "Means for recognizing emotions in speech and taking corresponding actions" refers to technologies and functions that allow the system to automatically respond or take appropriate actions based on the recognized emotional information.

[0366] DETAILED DESCRIPTION OF THE INVENTION The present invention provides a voice interface for autonomous vehicles with multilingual support and emotion recognition capabilities.

[0367] Hardware and Software Configuration

[0368] 1. Hardware:

[0369] Microphone: Used to collect passenger speech.

[0370] Speaker: Used to play back the generated audio data.

[0371] A smartphone or in-vehicle computer: Used to process the voice data.

[0372] 2. Software:

[0373] Speech Recognition Library: speech_recognition

[0374] Translation library: googletrans

[0375] Emotion Recognition Model: A sentiment analysis model using the transformers library

[0376] Text-to-speech libraries: gTTS and playsound

[0377] Data processing and calculation flow

[0378] 1. Voice input: Using the device's microphone, we collect speech in Japanese uttered by the user. For example, a passenger might say, "Please go to Shinjuku Station."

[0379] 2. Speech Recognition: Using the speech_recognition library, the device converts the collected voice data into text format.

[0380] 3. Translation: Use googletrans library to translate the converted Japanese text into English. For example, "Please take me to Shinjuku Station" is translated to "Please take me to Shinjuku Station".

[0381] 4. Emotion Recognition: Use the transformers library to recognize user emotions from translated text, for example, "tired" from the tone and content of the speech.

[0382] 5. Emotion-sensitive speech synthesis: Using the gTTS library, the translated text is converted into synthetic speech, and an intonation is added to reflect the recognized emotion, for example, "Please take me to Shinjuku Station" in a tired voice.

[0383] 6. Audio output: The generated audio data is sent to the other device and played through the speaker. Based on the passenger's request, the navigation system sets the destination and the autonomous vehicle begins operation.

[0384] Specific examples

[0385] For example, if a passenger says in Japanese, "I'm a little tired, so please go directly to the airport," the system will act as follows:

[0386] 1. The microphone collects the speech and the device performs voice recognition.

[0387] 2. Translate the recognized text "I'm a bit tired, please take me directly to the airport." using the googletrans library to get "I'm a bit tired, please take me directly to the airport."

[0388] 3. Emotion recognition is performed on the translated text to obtain the emotion label "Tired."

[0389] 4. The speech that reflects the recognized emotion "tired" is synthesized using the gTTS library to generate speech in a format that conveys the emotion in English.

[0390] 5. The generated audio is played through the speaker, the navigation system sets the destination according to the instructions, and the vehicle begins to travel along the route.

[0391] Prompt Sentence Examples

[0392] The user says in Japanese, "I'm a little tired, so please go directly to the airport." Translate this into English in real time and output the following voice output, reflecting the user's emotions.

[0393] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0394] Step 1:

[0395] The user performs voice input. The user utters "Please go to Shinjuku Station" in Japanese. The contents of this utterance are collected through the device's microphone.

[0396] Step 2:

[0397] The device converts the collected voice data into text format in real time using the speech recognition library speech_recognition. The input is voice data, and the output is text data: "Please go to Shinjuku Station."

[0398] Step 3:

[0399] After the text data is generated, the device translates this text into the target language (in this case, English) using the translation library googletrans. The input is the text data "Please take me to Shinjuku Station," and the output is the English text "Please take me to Shinjuku Station."

[0400] Step 4:

[0401] The server inputs the translated English text into an emotion recognition model (using the transformers library) to recognize the user's emotion. The input is the English text "Please take me to Shinjuku Station", and the output is an emotion label (e.g., "Tired").

[0402] Step 5:

[0403] The server uses the speech synthesis library gTTS to generate speech data reflecting the recognized emotions. The input is English text and emotion labels, and the output is speech data reflecting the emotions.

[0404] Step 6:

[0405] The device sends the generated voice data to the navigation system, which then uses it to set destinations and plan routes for the autonomous vehicle. The input is the voice data, and the output is the vehicle's instructed destination and route.

[0406] Step 7:

[0407] Finally, the terminal plays the generated voice data through a speaker to provide voice feedback to the passenger, facilitating user communication with the autonomous vehicle system.

[0408] 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.

[0409] 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.

[0410] 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.

[0411] [Second embodiment]

[0412] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0413] 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.

[0414] 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).

[0415] 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.

[0416] 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.

[0417] 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).

[0418] 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.

[0419] 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.

[0420] 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.

[0421] 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.

[0422] 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.

[0423] 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."

[0424] The system of the present invention translates what a user says in their own language into another language in real time and provides the translated speech to the other party. This enables smooth communication that transcends language barriers at international conferences, etc. The following describes in detail the embodiments of the present invention.

[0425] Collection of speech content

[0426] The user begins speaking in their own language. For example, the user is Japanese and says, "Hello, how is the project going?" This speech is collected by the user's device (smartphone or computer) via a microphone.

[0427] Sending audio data

[0428] The device transmits the collected voice data in real time to a server via the Internet. The voice data contains the user's speech in digital form.

[0429] Speech recognition and text conversion

[0430] The server receives voice data via the internet and inputs it into the generative AI model to convert the voice to text. At this time, speech recognition technology is used to accurately convert what the user has said into text format.

[0431] Content Complement

[0432] The server uses a generative AI model to analyze the context of the converted text and fill in any missing or ambiguous information. For example, if the generated text is "Hello, how's the project going?", it will be completed to make the meaning more specific: "Hello, regarding the project progress, what's the status?"

[0433] translation

[0434] The server translates the completed text data into the target language using a translation function. For example, the server translates the completed Japanese text into English, converting it into the English text "Hello, regarding the project's progress, how is it going?"

[0435] Speech synthesis

[0436] The server converts the translated text into speech data using speech synthesis technology, which prepares the translated content as speech output. For example, an English voice is generated based on the English text "Hello, regarding the project's progress, how is it going?"

[0437] Sending and playing audio data

[0438] The generated voice data is then sent from the server to the other party's device, which receives the voice data and plays it back through its speaker. In this way, even if the other user does not understand Japanese, they can understand what the Japanese person is saying through the English voice.

[0439] As a concrete example, consider a conversation between Japanese user A and American user B, both of whom are participating in an international conference. User A speaks in Japanese, which is played back as English audio on User B's device. User B also speaks in English, which is played back as Japanese audio on User A's device. This allows the two parties to communicate smoothly without feeling any language barrier.

[0440] The system of the present invention significantly improves international communication by processing users' speech in real time, converting it into other languages, and providing it to the other party. By exchanging voice and text data between the server and the terminals via the Internet, participants of multiple nationalities can experience a conversation as if they were speaking only in their own native language.

[0441] The processing flow will be explained below.

[0442] Step 1:

[0443] The user speaks in their own language. For example, Japanese user A says, "Hello, how is the project going?"

[0444] Step 2:

[0445] The terminal (User A's device) records the speech using a microphone and captures it as voice data, which is then temporarily stored in the terminal in digital format.

[0446] Step 3:

[0447] The device transmits the collected voice data to a server via the Internet in real time. The transmitted data includes the voice data as well as the user ID and device information.

[0448] Step 4:

[0449] The server inputs the received voice data into the generative AI model and converts it into text using speech recognition technology. This conversion process generates the text, "Hello, how is the project going?"

[0450] Step 5:

[0451] The server completes the generated text using a context analysis algorithm. For example, the text "Hello, how is the project going?" is completed as "Hello, regarding the project progress, what is the status?"

[0452] Step 6:

[0453] The server translates the completed text into the target language, for example translating the completed Japanese text into English to generate the English text "Hello, regarding the project's progress, how is it going?"

[0454] Step 7:

[0455] The server converts the translated text into speech data using speech synthesis technology, generating the English speech data "Hello, regarding the project's progress, how is it going?"

[0456] Step 8:

[0457] The server transmits the generated translated speech data via the Internet to the other party's terminal (user B's device). The transmitted data includes the translated speech data.

[0458] Step 9:

[0459] The terminal (user B's device) plays the received voice data through the device's speaker. User B can hear the English voice saying "Hello, regarding the project's progress, how is it going?"

[0460] Step 10:

[0461] User B (American) says in English, "The project is progressing well, thank you for asking."

[0462] Step 11:

[0463] The terminal (user B's device) records the speech using a microphone and captures the audio data in real time.

[0464] Step 12:

[0465] The terminal transmits the collected voice data to a server via the Internet in real time.

[0466] Step 13:

[0467] The server inputs the received voice data into the generative AI model and converts it into text using speech recognition technology. For example, the generated text might read, "The project is progressing well, thank you for asking."

[0468] Step 14:

[0469] The server uses a context analysis algorithm to complete the generated text, for example, "The project is progressing well, thank you for asking about the project's status."

[0470] Step 15:

[0471] The server translates the completed text into the target language, for example, translating the completed English text into Japanese to generate Japanese text such as "The project is progressing smoothly. Thank you for your inquiry."

[0472] Step 16:

[0473] The server converts the translated Japanese text into voice data using speech synthesis technology, generating Japanese voice data that says, "The project is progressing smoothly. Thank you for your inquiry."

[0474] Step 17:

[0475] The server transmits the generated translated speech data to the other party's terminal (user A's device) via the Internet.

[0476] Step 18:

[0477] The terminal (user A's device) plays the received voice data through the device's speaker. User A can hear the voice in his / her native language saying, "The project is progressing smoothly. Thank you for your inquiry."

[0478] Example 1

[0479] 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."

[0480] Conventional real-time translation systems have faced problems with the accuracy of translation of spoken content and a lack of context-based completion. Furthermore, when communicating between different languages, each step of speech recognition, translation, and speech synthesis is fragmented, which can impair real-time performance. There is a need to solve these problems and realize smoother, more accurate international communication.

[0481] 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.

[0482] In this invention, the server includes means for collecting user utterances, means for transmitting the collected utterances as voice data, means for converting the received voice data into text, means for completing the converted text using a generative AI model, means for translating the completed text into another language, means for converting the translated text into voice data, means for transmitting the generated voice data to a partner device, and means for playing the received voice data on the partner device. This enables the user's utterances to be translated accurately between different languages ​​in real time and for natural completion based on context.

[0483] A "user" is a person who uses the system of the present invention to input speech content.

[0484] "Speech content" refers to words and sentences spoken by the user, and is the speech data to be translated.

[0485] "Voice data" is data that expresses the contents of a user's speech in digital form.

[0486] "Text data" refers to data that has been converted from voice data into character information using voice recognition technology.

[0487] A "generative AI model" is a model that uses artificial intelligence technology to complement the content of text data, for example, natural language processing technology.

[0488] "Completion" is the process of adding contextual information to incomplete text data to make it more meaningful.

[0489] The "translation function" is a function for converting text data written in one language into another language.

[0490] "Speech synthesis technology" is a technology that generates voice data based on text data.

[0491] The "partner's device" is a receiving device used by the other party who is having a conversation with the user, and has the function of reproducing voice data.

[0492] The "collection means" is a means for acquiring the contents of a user's speech using a device such as a microphone.

[0493] The "transmission means" is a means for transmitting collected or generated voice data via the Internet or the like.

[0494] The "conversion means" is a means for converting voice data into text data and text data into voice data.

[0495] "Completion means" refers to a means of complementing the content of text data using a generative AI model.

[0496] The "playback means" is a function for playing back the audio data received by the other device from a speaker or the like.

[0497] MODE FOR CARRYING OUT THE INVENTION

[0498] The present invention provides a system that translates speech in a user's own language into another language in real time and provides the translated speech to the other party. This system enables smooth communication across language barriers in situations such as international conferences. Specific embodiments for implementing the present invention will be described below, specifying the hardware and software used.

[0499] This system consists of three main elements: the user, the terminal, and the server. The roles and specific operations of each are explained in detail below.

[0500] User Roles

[0501] Users speak in their own language. For example, a Japanese user might say, "Hello, how's the project going?" This speech is collected as voice data through a microphone installed on a device such as a smartphone or PC.

[0502] Device Role

[0503] The device sends the voice data collected from the user to a server via the Internet. The voice data is encoded in a digital format (e.g., PCM format). WebSocket or similar is often used as a communication protocol to maintain real-time performance.

[0504] Server Roles

[0505] The server performs the following sequence of operations:

[0506] 1. Speech Recognition:

[0507] The server receives the voice data and converts it into text data using a speech recognition engine (e.g., Google Speech-to-Text API). This conversion converts the user's speech into text.

[0508] 2. Content Complement:

[0509] The converted text data is input into a generative AI model (e.g., OpenAI GPT-3). The server uses this generative AI model to analyze the context of the text and fill in missing or ambiguous information. For example, for the text "Hello, how is the project progressing?", it generates a more specific completion text: "Hello, regarding the progress of the project, what is the status?"

[0510] 3. Translation:

[0511] The completed text data is sent to a translation engine (e.g., DeepL or Google Translate API). The server uses this engine to translate the text data into another language. For example, translating Japanese text into English yields the English text "Hello, regarding the project's progress, how is it going?"

[0512] 4. Speech synthesis:

[0513] The translated text is input into a speech synthesis engine (e.g., IBM Watson Text-to-Speech or Amazon Polly) and converted into audio data, which generates natural-sounding audio data based on the translated English text.

[0514] The server sends the generated voice data to the other device. The other device has the function to play the received voice data and outputs the voice from a speaker. This allows the other user to understand what the user is saying through the English voice, even if they do not understand Japanese.

[0515] Specific examples

[0516] As a concrete example, consider a conversation between Japanese user A and American user B, both of whom are attending an international conference. User A says in Japanese, "Hello, how is the project progressing?", which is played back in English on User B's device. User B also says in English, "The project is progressing well, thank you.", which is played back in Japanese on User A's device. This allows the two parties to communicate smoothly without feeling any language barrier.

[0517] Prompt Sentence Examples

[0518] "Please translate the following Japanese text into natural English and synthesize it into speech: 'Hello, how's the project going?'"

[0519] The system of the present invention combines real-time speech recognition, text completion, translation, and speech synthesis technologies over the Internet to significantly improve international communication.

[0520] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0521] Program processing flow

[0522] Step 1:

[0523] The user begins speaking in their own language. The speech is collected as voice data through the device's microphone. The input is the user's speech, and the output is the collected voice data. The device's specific operation is to convert the analog voice signal via the microphone into digital voice data and store it in its internal memory.

[0524] Step 2:

[0525] The device sends the collected voice data to the server via the Internet. The input is the voice data collected in step 1, and the output is the transmitted voice data. Specifically, the device prepares to transmit the voice data to the server using a protocol such as WebSocket, and then begins transmission.

[0526] Step 3:

[0527] The server calls a speech recognition engine to convert the received voice data into text data. For example, it uses the Google Speech-to-Text API. The input is the voice data sent from the device, and the output is text data. Specifically, the server sends the voice data to the API and receives the returned text data.

[0528] Step 4:

[0529] The server uses a generative AI model to complete the content of the received text data. For example, OpenAI GPT-3 is used as this model. The input is the text data converted in step 3, and the output is the completed text data. Specifically, the server inputs the text data into the generative AI model and receives the completed text.

[0530] Step 5:

[0531] The server sends the completed text data to a multilingual translation engine to translate it into other languages. For example, it uses DeepL or Google Translate API. The input is the text data completed in step 4, and the output is the translated text data. Specifically, the server sends the text data to the translation engine and receives the translated results.

[0532] Step 6:

[0533] The server inputs the translated text data into a speech synthesis engine and converts it into voice data. For example, IBM Watson Text-to-Speech or Amazon Polly can be used. The input is the text data translated in step 5, and the output is the generated voice data. Specifically, the server sends the text data to the speech synthesis engine and receives the generated voice data.

[0534] Step 7:

[0535] The server transmits the generated voice data to the other party's terminal via the Internet. The input is the voice data generated in step 6, and the output is the transmitted voice data. Specifically, the server performs the transmission process using a protocol for transmitting voice data to the other party's terminal.

[0536] Step 8:

[0537] The other device plays the received voice data. The input is the voice data sent from the server, and the output is the voice played through the speaker. Specifically, the device decodes the received voice data and plays it back through the speaker. This allows the other user to understand what the user is saying.

[0538] (Application example 1)

[0539] 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."

[0540] Establishing real-time communication between users who speak different languages ​​is a major challenge, especially in international business and tourism. There is also a demand for seamless audio and visual information delivery using devices such as smart glasses. Existing technologies struggle to fully meet these demands, particularly in terms of accuracy and speed. Therefore, a new system is needed to provide real-time translated audio and visual information between users who speak different languages.

[0541] 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.

[0542] In this invention, the server includes: means for collecting speech content from users; means for transmitting the collected speech content as voice data; means for converting the received voice data into text; means for completing the converted text; means for translating the completed text into another language; means for converting the translated text into voice data; means for transmitting the converted voice data to a remote terminal; means for playing the received voice data at the remote terminal; means for realizing real-time language translation between a service provider and a customer and providing voice and visual information through the smart glasses; and means for detecting the user's speech using a microphone and speaker of the smart glasses and providing voice feedback to the remote terminal in the translated language, thereby enabling real-time translation of voice and visual information between users speaking different languages.

[0543] "User" refers to a person who uses the system to translate spoken content.

[0544] "Speech content" refers to what the user says in natural language.

[0545] "Means for collecting" refers to the function of acquiring the user's speech content as voice data.

[0546] "Transmitting means" refers to the function of transmitting collected voice data to an external device such as a server.

[0547] "Means for converting to text" refers to a function for converting received voice data into character data.

[0548] "Means for supplementing" refers to the function of adding supplementary information based on the context of the converted text content.

[0549] "Means for translating" refers to a function for translating the completed text into a specified other language.

[0550] "Means for converting into voice data" refers to the function of converting the translated text into data for voice output.

[0551] "Means for transmitting" refers to the function of transferring voice data to the other party's terminal.

[0552] "Means for playing" refers to the function of actually outputting the audio data received on the other party's terminal as audio.

[0553] A "service provider" refers to the person who interacts with customers at a store or facility.

[0554] "Customer" means a person who uses the System for the purpose of communicating with a Service Provider.

[0555] "Smart glasses" are glasses-type devices that utilize built-in microphones and speakers to facilitate interaction between the user and the system.

[0556] "Voice feedback" refers to the function of translating a user's speech and providing it as audio.

[0557] The system of the present invention translates what a user says in their own language into another language in real time and presents the translated speech to the other party as audio, thereby realizing smooth communication between users who speak different languages. Specific embodiments for carrying out the present invention will be described below.

[0558] A means of generating a program

[0559] The programs running on the server use the following software:

[0560] speech_recognition: Collects speech from the microphone and converts it into text.

[0561] googletrans: Translate the converted text into other languages.

[0562] gtts: Converts translated text into audio data.

[0563] playsound: Plays the generated sound data.

[0564] Hardware

[0565] The system uses smart glasses as its primary device, which includes the following hardware:

[0566] Microphone: Collects user utterances.

[0567] Speaker: Plays the translated audio.

[0568] Display: Visually display the completed text information.

[0569] Data processing and calculation

[0570] The server processes the data in the following steps:

[0571] 1. Voice data collection: The user's speech is collected through the microphone of the smart glasses. The collected voice data is temporarily stored in the memory of the smart glasses.

[0572] 2. Sending voice data: The smart glasses send the collected voice data to the server.

[0573] 3. Speech recognition and text conversion: The server converts the voice data into text using "speech_recognition".

[0574] 4. Content completion: Using a "generative AI model," the context of the converted text is analyzed and information is completed as needed.

[0575] 5. Text translation: The server translates the completed text into other languages ​​using "googletrans".

[0576] 6. Speech synthesis: The translated text is converted into voice data by "gtts".

[0577] 7. Sending voice data: The converted voice data is sent to the speaker of the smart glasses.

[0578] 8. Playback of audio data: The translated audio will be played through the smart glasses' speaker.

[0579] Specific examples

[0580] For example, if a Japanese store clerk says "Here is the product," the microphone in the smart glasses collects the voice. The voice data is sent to the server and converted into Japanese text "Here is the product." The text is then translated into English, generating the English text "Here is the product." This English text is converted into speech and played back from the speaker in the customer's smart glasses as "Here is the product." In this way, the foreign customer can understand the explanation of the store clerk.

[0581] Prompt Sentence Examples

[0582] User: Says, "Hi, how's the project going?"

[0583] Server: Identify the language and provide the translated results in text format.

[0584] Output: "Hello, regarding the project's progress, how is it going?"

[0585] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0586] Step 1:

[0587] A user puts on the smart glasses and starts speaking. The microphone in the smart glasses collects the speech as voice data. The collected voice data is temporarily stored in the memory in the smart glasses.

[0588] Input: User's speech

[0589] Output: Audio data in smart glasses

[0590] Step 2:

[0591] The device (smart glasses) transmits the collected voice data to a server in real time, using an internet connection to transfer the data.

[0592] Input: Audio data in smart glasses

[0593] Output: Audio data sent to the server

[0594] Step 3:

[0595] The server converts the received voice data into text using the "speech_recognition" library. Through this conversion, the user's speech content is obtained as text data.

[0596] Input: Audio data sent to the server

[0597] Output: Converted text data

[0598] Step 4:

[0599] The server then uses a generative AI model to analyze the converted text and complete the information based on the context, thereby fleshing out any ambiguities or missing information.

[0600] Input: Converted text data

[0601] Output: Completed text data

[0602] Step 5:

[0603] The server uses "googletrans" to translate the completed text into the specified language, resulting in the user's speech being expressed in another language.

[0604] Input: Completed text data

[0605] Output: Translated text data

[0606] Step 6:

[0607] The server converts the translated text into audio data using "gtts" (Google Text-to-Speech), which then creates a playable audio format.

[0608] Input: Translated text data

[0609] Output: Generated audio data

[0610] Step 7:

[0611] The server then transmits the generated audio data to the smart glasses, again using an internet connection to transfer the data.

[0612] Input: Generated audio data

[0613] Output: Audio data sent to the smart glasses

[0614] Step 8:

[0615] The device (smart glasses) plays the received audio data through a speaker, so that the user's speech is provided as audio in the translated language.

[0616] Input: Audio data transmitted to smart glasses

[0617] Output: Playback of the translated audio

[0618] In this way, the system of the present invention translates the user's speech in real time and provides it as audio, thereby realizing smooth communication between users who speak different languages.

[0619] 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.

[0620] The system of the present invention translates what a user says in their own language into another language in real time, and also has the function of recognizing the user's emotions and reflecting them in the translation. By recognizing the user's emotions, more natural and emotional communication is realized. The following describes in detail an embodiment of the present invention.

[0621] Collection of speech content

[0622] The user speaks in their own language. For example, Japanese user A says, "Hello, how is the project going?" The content of this speech is collected by the user's device (smartphone or computer) through a microphone.

[0623] Sending audio data

[0624] The device transmits the collected voice data in real time to a server via the Internet. The voice data contains the user's speech in digital form.

[0625] Speech recognition and text conversion

[0626] The server inputs the voice data received via the internet into the generative AI model and converts the voice to text. For example, the generated text is "Hello, how is the project going?"

[0627] emotion recognition

[0628] The server uses the voice and text data to recognize the user's emotions through an emotion engine. For example, the server can recognize emotions such as "interest" and "concern" from User A's tone of voice and speaking style.

[0629] Content Complement

[0630] The server completes the generated text using a context analysis algorithm. For example, the text "Hello, how is the project going?" is completed as "Hello, regarding the project progress, what is the status?"

[0631] translation

[0632] The server translates the completed text into the target language and reflects the recognized sentiment in the translation result. For example, when translating the completed Japanese text into English and generating the English text "Hello, regarding the project's progress, how is it going?", the server reflects that User A is "interested" or "concerned."

[0633] Speech synthesis

[0634] The server converts the translated text into speech data using speech synthesis technology, generating English speech data such as "Hello, regarding the project's progress, how is it going?" with emotional intonation.

[0635] Sending and playing audio data

[0636] The generated voice data is again sent from the server to the other device (User B's device). The other device receives this voice data and plays it through the speaker. User B can hear the English voice "Hello, regarding the project's progress, how is it going?" with the emotion reflected.

[0637] As a concrete example, consider a conversation between Japanese user A and American user B, both of whom are participating in an international conference. User A speaks in Japanese, which is played back as English audio on User B's device, with User A's emotions also being conveyed. Similarly, User B speaks in English, which is played back as Japanese audio on User A's device, with User A's emotions also being reflected. This allows both parties to communicate their emotions accurately without feeling a language barrier, resulting in more natural and effective communication.

[0638] The system of the present invention recognizes the user's emotions throughout the entire process, from voice data collection, transmission, text conversion, content completion, translation, voice synthesis, and finally transmission and playback of the voice data, and reflects these emotions in the translation results, thereby significantly improving international communication.

[0639] The processing flow will be explained below.

[0640] Step 1:

[0641] The user speaks in their own language. For example, Japanese user A says, "Hello, how is the project going?"

[0642] Step 2:

[0643] The terminal (User A's device) records the speech using a microphone and captures it as voice data, which is then temporarily stored in the terminal in digital format.

[0644] Step 3:

[0645] The device transmits the collected voice data to a server via the Internet in real time. The transmitted data includes the voice data as well as the user ID and device information.

[0646] Step 4:

[0647] The server inputs the received voice data into the generative AI model and converts it into text using speech recognition technology. This conversion process generates the text, "Hello, how is the project going?"

[0648] Step 5:

[0649] The server uses the voice and text data to recognize the user's emotions through an emotion engine. For example, the server can recognize emotions such as "interest" and "concern" from User A's tone of voice and speaking style.

[0650] Step 6:

[0651] The server completes the generated text using a context analysis algorithm. For example, the text "Hello, how is the project going?" is completed as "Hello, regarding the project progress, what is the status?"

[0652] Step 7:

[0653] The server translates the completed text into the target language and reflects the recognized sentiment in the translation result. For example, when translating the completed Japanese text into English and generating the English text "Hello, regarding the project's progress, how is it going?", the server reflects that User A is "interested" or "concerned."

[0654] Step 8:

[0655] The server converts the translated text into speech data using speech synthesis technology, generating English speech data such as "Hello, regarding the project's progress, how is it going?" with emotional intonation.

[0656] Step 9:

[0657] The server transmits the generated translated speech data via the Internet to the other party's terminal (user B's device). The transmitted data includes the translated speech data.

[0658] Step 10:

[0659] The device (User B's device) plays the received voice data through the device's speaker. User B can hear the English voice "Hello, regarding the project's progress, how is it going?" with the emotion reflected.

[0660] Step 11:

[0661] User B (American) says in English, "The project is progressing well, thank you for asking."

[0662] Step 12:

[0663] The device (User B's device) records the speech with a microphone and captures the audio data in real time, which is then temporarily stored in the device in digital format.

[0664] Step 13:

[0665] The device transmits the collected voice data to a server via the Internet in real time. The transmitted data includes the voice data as well as the user ID and device information.

[0666] Step 14:

[0667] The server inputs the received voice data into the generative AI model and converts it into text using speech recognition technology. For example, the generated text might read, "The project is progressing well, thank you for asking."

[0668] Step 15:

[0669] The server uses the voice and text data to recognize the user's emotions using an emotion engine. For example, emotions such as "gratitude" and "relief" can be recognized from User B's tone of voice and speaking style.

[0670] Step 16:

[0671] The server uses a context analysis algorithm to complete the generated text. For example, the text "The project is progressing well, thank you for asking" is completed as "The project is progressing well, thank you for asking about the project's status."

[0672] Step 17:

[0673] The server translates the completed text into the target language and reflects the recognized emotions in the translation result. For example, when translating the completed English text into Japanese and generating the Japanese text "The project is progressing smoothly. Thank you for your inquiry," the "gratitude" and "relief" that User B feels are reflected.

[0674] Step 18:

[0675] The server then converts the translated Japanese text into speech data using speech synthesis technology, generating Japanese speech data that says, "The project is progressing smoothly. Thank you for your inquiry," with intonation that reflects the emotion.

[0676] Step 19:

[0677] The server transmits the generated translated speech data to the other party's terminal (user A's device) via the Internet. The transmitted data includes the translated speech data.

[0678] Step 20:

[0679] The terminal (user A's device) plays the received voice data through the device's speaker. User A can hear the voice in his / her native language saying "The project is progressing smoothly. Thank you for your inquiry." with the emotion reflected.

[0680] Example 2

[0681] 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."

[0682] In today's global society, people who speak different languages ​​need to communicate smoothly. However, while conventional translation systems can accurately translate spoken content, they struggle to convey the speaker's emotions appropriately. As a result, misunderstandings and misrecognitions are likely to occur, hindering natural communication. Furthermore, they often fail to properly complement the spoken content by taking into account the context of the spoken content, making it impossible to convey the speaker's full intention.

[0683] 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.

[0684] In this invention, the server includes a means for recognizing the user's emotions, a means for completing the content of the converted text, and a means for converting the translated text into voice data. This not only ensures accurate translation of the spoken content, but also enables the speaker's emotions to be conveyed to the other party in a manner that appropriately reflects their intention. Furthermore, context-based content completion allows the speaker to fully convey their intention, resulting in more natural and effective communication.

[0685] "User" refers to a person who uses the system to translate spoken content into another language.

[0686] The term "terminal" refers to an electronic device that collects the user's speech, transmits the voice data to a server, and plays back the voice data transmitted from the server.

[0687] "Server" refers to a central control device that processes voice data sent from a user's terminal and performs translation and voice synthesis.

[0688] "Speech content" refers to words and sentences spoken by a user in their own language.

[0689] "Voice data" refers to data that represents the contents of a user's speech in digital form.

[0690] "Text" refers to data that has been converted from audio data into character information.

[0691] "Emotion" refers to the emotional elements (e.g., interest, concern, joy, etc.) contained in the user's utterance.

[0692] "Content completion" refers to the process of analyzing the context of the spoken content and completing incomplete parts in a natural way.

[0693] "Translation" refers to the process of converting text expressed in one language into another language.

[0694] "Speech synthesis" refers to the process of converting text data into voice data.

[0695] "Emotion recognition" refers to the process of analyzing and identifying emotional elements from a user's speech.

[0696] "Real-time" means that processing occurs with little to no delay.

[0697] The system of the present invention translates what a user says in their own language into another language in real time, and further recognizes the user's emotions and reflects them in the translation, thereby realizing natural and emotional communication. A specific embodiment of this system will be described below.

[0698] Collection of speech content

[0699] The user speaks in their own language. For example, a Japanese user might say, "Hello, how's the project going?" This speech is collected by the user's device (smartphone or computer) via a microphone.

[0700] Sending audio data

[0701] The device transmits the collected voice data to a server over the Internet in real time. The voice data contains the user's speech in digital form. The device transmits the voice data using a secure communication protocol (e.g., HTTPS).

[0702] Speech recognition and text conversion

[0703] The server inputs the voice data received via the internet into a generative AI model and converts the voice into text. For example, the generative AI model used is the Google Cloud Speech-to-Text API, and generates the text "Hello, how is the project progressing?" from Japanese voice data.

[0704] emotion recognition

[0705] The server uses voice and text data to recognize the user's emotions through an emotion engine. For example, the emotion engine used is IBM Watson Tone Analyzer, which recognizes emotions such as "interest" and "concern" from the user's tone of voice and speaking style.

[0706] Content Complement

[0707] The server uses a context analysis algorithm to complete the generated text. For example, the context analysis algorithm used is OpenAI GPT-3, and the text "Hello, how is the project going?" is completed as "Hello, regarding the progress of the project, what is the status?"

[0708] translation

[0709] The server translates the completed text into the target language and reflects the recognized sentiment in the translation results. For example, the translation engine used is the Google Translate API, and the completed Japanese text is translated into English to generate the English text "Hello, regarding the project's progress, how is it going?". In this case, the user's "interest" and "concern" are reflected in the translation results.

[0710] Speech synthesis

[0711] The server converts the translated text into voice data using Amazon Polly speech synthesis technology. The generated voice data is an English voice saying "Hello, regarding the project's progress, how is it going?" with intonation that reflects the emotion.

[0712] Sending and playing audio data

[0713] The generated voice data is again sent from the server to the other device (User B's device). The other device receives this voice data and plays it through the speaker. User B can hear the English voice saying "Hello, regarding the project's progress, how is it going?" with the emotion reflected.

[0714] As a concrete example, consider a conversation between a Japanese user and an American user participating in an international conference. The Japanese user speaks in Japanese, which is played back as English speech on the American user's device, with the Japanese user's emotions also being conveyed. Similarly, the American user speaks in English, which is played back as Japanese speech on the Japanese user's device, with the Japanese user's emotions also being reflected.

[0715] Prompt Sentence Examples

[0716] "A user says, 'Hello, how's the project going?' in Japanese. Please translate this utterance into English and convert it into speech data that reflects the sentiment."

[0717] This system will significantly improve international communication, enabling more natural and emotional interactions.

[0718] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0719] Step 1:

[0720] The user speaks in their own language. The user's voice is collected by the device as input. For example, a Japanese user might say, "Hello, how's the project going?". Specifically, the device's microphone collects the voice in digital form. Digital voice data is generated as output.

[0721] Step 2:

[0722] The device transmits the collected audio data in real time to a server in uncompressed or compressed format. Digital audio data is input. Specifically, the device transmits the audio data using a secure communication protocol such as HTTPS. The audio data arrives at the server as output.

[0723] Step 3:

[0724] The server inputs the received voice data into a generative AI model and converts the voice to text. The voice data received on the server side exists as input. Specifically, it uses the Google Cloud Speech-to-Text API to convert it into text data such as "Hello, how is the project progressing?". Text data is generated as output.

[0725] Step 4:

[0726] The server uses voice and text data to recognize the user's emotions through an emotion engine. Converted text data and voice data are used as input. Specifically, the server uses IBM Watson Tone Analyzer to identify emotions such as "interest" or "concern" from the user's tone of voice and choice of words. Emotional data is generated as output.

[0727] Step 5:

[0728] The server analyzes the context of the generated text data and completes the content. Emotion data and text data are input. Specifically, OpenAI GPT-3 is used to complete the text "Hello, how is the project progressing?" to "Hello, regarding the progress of the project, what is the status?". The completed text data is generated as output.

[0729] Step 6:

[0730] The server translates the completed text into the target language and also reflects the recognized emotion in the translation result. The input is the completed text data and emotion data. Specifically, it uses the Google Translate API to translate the completed text "Hello, regarding the project's progress, how is it going?" into English and reflects the emotion to generate "Hello, regarding the project's progress, how is it going?" The translated text data is generated as the output.

[0731] Step 7:

[0732] The server converts the translated text into speech data using speech synthesis technology. The inputs are the translated text data and emotion data. Specifically, Amazon Polly is used to generate English speech data saying "Hello, regarding the project's progress, how is it going?". This speech data is given an intonation that reflects the emotion. English speech data is generated as output.

[0733] Step 8:

[0734] The server sends the generated voice data to the other party's device. The generated voice data exists as input. Specifically, the voice data is sent to User B's device via the Internet. The voice data arrives at the other party's device as output.

[0735] Step 9:

[0736] The other device plays the received audio data. The input is English audio data from the server. Specifically, the device's speaker plays the English audio, "Hello, regarding the project's progress, how is it going?". User B can hear the audio as output.

[0737] This detailed processing allows users who speak different languages ​​to communicate accurately and with rich emotion.

[0738] (Application example 2)

[0739] 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."

[0740] Currently, in order for autonomous vehicles to communicate smoothly between passengers and vehicle systems, it is necessary not only to recognize instructions but also to communicate seamlessly in various languages. It is also necessary to simultaneously recognize passenger emotions and respond appropriately to those emotions to provide a safer and more comfortable travel experience. However, current systems are unable to effectively support multiple languages ​​and recognize emotions, resulting in reduced passenger satisfaction and safety.

[0741] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting utterances from users, means for transmitting the collected utterances as voice data, means for converting the received voice data into text, means for completing the converted text, means for translating the completed text into another language, means for converting the translated text into voice data, means for transmitting the converted voice data to the other party's terminal, means for performing emotion recognition, means for generating voice data based on the recognized emotion, means for playing the received voice data at the other party's terminal, means for giving instructions to the vehicle system or setting a destination based on the passenger's utterance, and means for recognizing the emotion of the utterance and performing a corresponding action. This enables natural communication that recognizes emotions in multiple languages, thereby realizing safe and comfortable operation of autonomous vehicles.

[0742] The "means for collecting speech content from the user" refers to a device or function for recognizing the speech uttered by the user and acquiring it as digital data.

[0743] "Means for transmitting collected speech content as voice data" refers to a mechanism for sending acquired voice data to a server via a communication network such as the Internet.

[0744] The "means for converting received voice data into text" refers to a system that analyzes voice data on the server side and converts it into a corresponding text format.

[0745] "Means for completing the content of the converted text" refers to technologies and algorithms that analyze the context of the converted text, add information as needed, and arrange the sentences in a natural format.

[0746] The "means for translating the completed text into another language" is an automatic translation technology for converting the completed text into another specified language.

[0747] "Means for converting translated text into speech data" refers to technology or devices for converting translated text into synthetic speech.

[0748] The "means for transmitting the converted voice data to the other terminal" is the communication means used to retransmit the generated voice data to the target device.

[0749] "Means for emotion recognition" refers to technologies and algorithms that analyze and recognize a speaker's emotional state from audio data or text data.

[0750] The "means for generating voice data based on recognized emotions" refers to a technology that adjusts the intonation and atmosphere of the synthesized voice by reflecting the recognized emotional information.

[0751] The "means for playing back the received audio data on the other party's terminal" is a function for playing back the transmitted audio data on the device.

[0752] "Means for giving instructions to vehicle systems and setting destinations based on passenger speech" refers to technology that recognizes what passengers say and operates the vehicle's navigation system and other operations based on that speech.

[0753] "Means for recognizing emotions in speech and taking corresponding actions" refers to technologies and functions that allow the system to automatically respond or take appropriate actions based on the recognized emotional information.

[0754] DETAILED DESCRIPTION OF THE INVENTION The present invention provides a voice interface for autonomous vehicles with multilingual support and emotion recognition capabilities.

[0755] Hardware and Software Configuration

[0756] 1. Hardware:

[0757] Microphone: Used to collect passenger speech.

[0758] Speaker: Used to play back the generated audio data.

[0759] A smartphone or in-vehicle computer: Used to process the voice data.

[0760] 2. Software:

[0761] Speech Recognition Library: speech_recognition

[0762] Translation library: googletrans

[0763] Emotion Recognition Model: A sentiment analysis model using the transformers library

[0764] Text-to-speech libraries: gTTS and playsound

[0765] Data processing and calculation flow

[0766] 1. Voice input: Using the device's microphone, we collect speech in Japanese uttered by the user. For example, a passenger might say, "Please go to Shinjuku Station."

[0767] 2. Speech Recognition: Using the speech_recognition library, the device converts the collected voice data into text format.

[0768] 3. Translation: Use googletrans library to translate the converted Japanese text into English. For example, "Please take me to Shinjuku Station" is translated to "Please take me to Shinjuku Station".

[0769] 4. Emotion Recognition: Use the transformers library to recognize user emotions from translated text, for example, "tired" from the tone and content of the speech.

[0770] 5. Emotion-sensitive speech synthesis: Using the gTTS library, the translated text is converted into synthetic speech, and an intonation is added to reflect the recognized emotion, for example, "Please take me to Shinjuku Station" in a tired voice.

[0771] 6. Audio output: The generated audio data is sent to the other device and played through the speaker. Based on the passenger's request, the navigation system sets the destination and the autonomous vehicle begins operation.

[0772] Specific examples

[0773] For example, if a passenger says in Japanese, "I'm a little tired, so please go directly to the airport," the system will act as follows:

[0774] 1. The microphone collects the speech and the device performs voice recognition.

[0775] 2. Translate the recognized text "I'm a bit tired, please take me directly to the airport." using the googletrans library to get "I'm a bit tired, please take me directly to the airport."

[0776] 3. Emotion recognition is performed on the translated text to obtain the emotion label "Tired."

[0777] 4. The speech that reflects the recognized emotion "tired" is synthesized using the gTTS library to generate speech in a format that conveys the emotion in English.

[0778] 5. The generated audio is played through the speaker, the navigation system sets the destination according to the instructions, and the vehicle begins to travel along the route.

[0779] Prompt Sentence Examples

[0780] The user says in Japanese, "I'm a little tired, so please go directly to the airport." Translate this into English in real time and output the following voice output, reflecting the user's emotions.

[0781] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0782] Step 1:

[0783] The user performs voice input. The user utters "Please go to Shinjuku Station" in Japanese. The contents of this utterance are collected through the device's microphone.

[0784] Step 2:

[0785] The device converts the collected voice data into text format in real time using the speech recognition library speech_recognition. The input is voice data, and the output is text data: "Please go to Shinjuku Station."

[0786] Step 3:

[0787] After the text data is generated, the device translates this text into the target language (in this case, English) using the translation library googletrans. The input is the text data "Please take me to Shinjuku Station," and the output is the English text "Please take me to Shinjuku Station."

[0788] Step 4:

[0789] The server inputs the translated English text into an emotion recognition model (using the transformers library) to recognize the user's emotion. The input is the English text "Please take me to Shinjuku Station", and the output is an emotion label (e.g., "Tired").

[0790] Step 5:

[0791] The server uses the speech synthesis library gTTS to generate speech data reflecting the recognized emotions. The input is English text and emotion labels, and the output is speech data reflecting the emotions.

[0792] Step 6:

[0793] The device sends the generated voice data to the navigation system, which then uses it to set destinations and plan routes for the autonomous vehicle. The input is the voice data, and the output is the vehicle's instructed destination and route.

[0794] Step 7:

[0795] Finally, the terminal plays the generated voice data through a speaker to provide voice feedback to the passenger, facilitating user communication with the autonomous vehicle system.

[0796] 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.

[0797] 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.

[0798] 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.

[0799] [Third embodiment]

[0800] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0801] 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.

[0802] 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).

[0803] 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.

[0804] 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.

[0805] 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).

[0806] 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.

[0807] 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.

[0808] 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.

[0809] 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.

[0810] 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.

[0811] 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."

[0812] The system of the present invention translates what a user says in their own language into another language in real time and provides the translated speech to the other party. This enables smooth communication that transcends language barriers at international conferences, etc. The following describes in detail the embodiments of the present invention.

[0813] Collection of speech content

[0814] The user begins speaking in their own language. For example, the user is Japanese and says, "Hello, how is the project going?" This speech is collected by the user's device (smartphone or computer) via a microphone.

[0815] Sending audio data

[0816] The device transmits the collected voice data in real time to a server via the Internet. The voice data contains the user's speech in digital form.

[0817] Speech recognition and text conversion

[0818] The server receives voice data via the internet and inputs it into the generative AI model to convert the voice to text. At this time, speech recognition technology is used to accurately convert what the user has said into text format.

[0819] Content Complement

[0820] The server uses a generative AI model to analyze the context of the converted text and fill in any missing or ambiguous information. For example, if the generated text is "Hello, how's the project going?", it will be completed to make the meaning more specific: "Hello, regarding the project progress, what's the status?"

[0821] translation

[0822] The server translates the completed text data into the target language using a translation function. For example, the server translates the completed Japanese text into English, converting it into the English text "Hello, regarding the project's progress, how is it going?"

[0823] Speech synthesis

[0824] The server converts the translated text into speech data using speech synthesis technology, which prepares the translated content as speech output. For example, an English voice is generated based on the English text "Hello, regarding the project's progress, how is it going?"

[0825] Sending and playing audio data

[0826] The generated voice data is then sent from the server to the other party's device, which receives the voice data and plays it back through its speaker. In this way, even if the other user does not understand Japanese, they can understand what the Japanese person is saying through the English voice.

[0827] As a concrete example, consider a conversation between Japanese user A and American user B, both of whom are participating in an international conference. User A speaks in Japanese, which is played back as English audio on User B's device. User B also speaks in English, which is played back as Japanese audio on User A's device. This allows the two parties to communicate smoothly without feeling any language barrier.

[0828] The system of the present invention significantly improves international communication by processing users' speech in real time, converting it into other languages, and providing it to the other party. By exchanging voice and text data between the server and the terminals via the Internet, participants of multiple nationalities can experience a conversation as if they were speaking only in their own native language.

[0829] The processing flow will be explained below.

[0830] Step 1:

[0831] The user speaks in their own language. For example, Japanese user A says, "Hello, how is the project going?"

[0832] Step 2:

[0833] The terminal (User A's device) records the speech using a microphone and captures it as voice data, which is then temporarily stored in the terminal in digital format.

[0834] Step 3:

[0835] The device transmits the collected voice data to a server via the Internet in real time. The transmitted data includes the voice data as well as the user ID and device information.

[0836] Step 4:

[0837] The server inputs the received voice data into the generative AI model and converts it into text using speech recognition technology. This conversion process generates the text, "Hello, how is the project going?"

[0838] Step 5:

[0839] The server completes the generated text using a context analysis algorithm. For example, the text "Hello, how is the project going?" is completed as "Hello, regarding the project progress, what is the status?"

[0840] Step 6:

[0841] The server translates the completed text into the target language, for example translating the completed Japanese text into English to generate the English text "Hello, regarding the project's progress, how is it going?"

[0842] Step 7:

[0843] The server converts the translated text into speech data using speech synthesis technology, generating the English speech data "Hello, regarding the project's progress, how is it going?"

[0844] Step 8:

[0845] The server transmits the generated translated speech data via the Internet to the other party's terminal (user B's device). The transmitted data includes the translated speech data.

[0846] Step 9:

[0847] The terminal (user B's device) plays the received voice data through the device's speaker. User B can hear the English voice saying "Hello, regarding the project's progress, how is it going?"

[0848] Step 10:

[0849] User B (American) says in English, "The project is progressing well, thank you for asking."

[0850] Step 11:

[0851] The terminal (user B's device) records the speech using a microphone and captures the audio data in real time.

[0852] Step 12:

[0853] The terminal transmits the collected voice data to a server via the Internet in real time.

[0854] Step 13:

[0855] The server inputs the received voice data into the generative AI model and converts it into text using speech recognition technology. For example, the generated text might read, "The project is progressing well, thank you for asking."

[0856] Step 14:

[0857] The server uses a context analysis algorithm to complete the generated text, for example, "The project is progressing well, thank you for asking about the project's status."

[0858] Step 15:

[0859] The server translates the completed text into the target language, for example, translating the completed English text into Japanese to generate Japanese text such as "The project is progressing smoothly. Thank you for your inquiry."

[0860] Step 16:

[0861] The server converts the translated Japanese text into voice data using speech synthesis technology, generating Japanese voice data that says, "The project is progressing smoothly. Thank you for your inquiry."

[0862] Step 17:

[0863] The server transmits the generated translated speech data to the other party's terminal (user A's device) via the Internet.

[0864] Step 18:

[0865] The terminal (user A's device) plays the received voice data through the device's speaker. User A can hear the voice in his / her native language saying, "The project is progressing smoothly. Thank you for your inquiry."

[0866] Example 1

[0867] 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."

[0868] Conventional real-time translation systems have faced problems with the accuracy of translation of spoken content and a lack of context-based completion. Furthermore, when communicating between different languages, each step of speech recognition, translation, and speech synthesis is fragmented, which can impair real-time performance. There is a need to solve these problems and realize smoother, more accurate international communication.

[0869] 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.

[0870] In this invention, the server includes means for collecting user utterances, means for transmitting the collected utterances as voice data, means for converting the received voice data into text, means for completing the converted text using a generative AI model, means for translating the completed text into another language, means for converting the translated text into voice data, means for transmitting the generated voice data to a partner device, and means for playing the received voice data on the partner device. This enables the user's utterances to be translated accurately between different languages ​​in real time and for natural completion based on context.

[0871] A "user" is a person who uses the system of the present invention to input speech content.

[0872] "Speech content" refers to words and sentences spoken by the user, and is the speech data to be translated.

[0873] "Voice data" is data that expresses the contents of a user's speech in digital form.

[0874] "Text data" refers to data that has been converted from voice data into character information using voice recognition technology.

[0875] A "generative AI model" is a model that uses artificial intelligence technology to complement the content of text data, for example, natural language processing technology.

[0876] "Completion" is the process of adding contextual information to incomplete text data to make it more meaningful.

[0877] The "translation function" is a function for converting text data written in one language into another language.

[0878] "Speech synthesis technology" is a technology that generates voice data based on text data.

[0879] The "partner's device" is a receiving device used by the other party who is having a conversation with the user, and has the function of reproducing voice data.

[0880] The "collection means" is a means for acquiring the contents of a user's speech using a device such as a microphone.

[0881] The "transmission means" is a means for transmitting collected or generated voice data via the Internet or the like.

[0882] The "conversion means" is a means for converting voice data into text data and text data into voice data.

[0883] "Completion means" refers to a means of complementing the content of text data using a generative AI model.

[0884] The "playback means" is a function for playing back the audio data received by the other device from a speaker or the like.

[0885] MODE FOR CARRYING OUT THE INVENTION

[0886] The present invention provides a system that translates speech in a user's own language into another language in real time and provides the translated speech to the other party. This system enables smooth communication across language barriers in situations such as international conferences. Specific embodiments for implementing the present invention will be described below, specifying the hardware and software used.

[0887] This system consists of three main elements: the user, the terminal, and the server. The roles and specific operations of each are explained in detail below.

[0888] User Roles

[0889] Users speak in their own language. For example, a Japanese user might say, "Hello, how's the project going?" This speech is collected as voice data through a microphone installed on a device such as a smartphone or PC.

[0890] Device Role

[0891] The device sends the voice data collected from the user to a server via the Internet. The voice data is encoded in a digital format (e.g., PCM format). WebSocket or similar is often used as a communication protocol to maintain real-time performance.

[0892] Server Roles

[0893] The server performs the following sequence of operations:

[0894] 1. Speech Recognition:

[0895] The server receives the voice data and converts it into text data using a speech recognition engine (e.g., Google Speech-to-Text API). This conversion converts the user's speech into text.

[0896] 2. Content Complement:

[0897] The converted text data is input into a generative AI model (e.g., OpenAI GPT-3). The server uses this generative AI model to analyze the context of the text and fill in missing or ambiguous information. For example, for the text "Hello, how is the project progressing?", it generates a more specific completion text: "Hello, regarding the progress of the project, what is the status?"

[0898] 3. Translation:

[0899] The completed text data is sent to a translation engine (e.g., DeepL or Google Translate API). The server uses this engine to translate the text data into another language. For example, translating Japanese text into English yields the English text "Hello, regarding the project's progress, how is it going?"

[0900] 4. Speech synthesis:

[0901] The translated text is input into a speech synthesis engine (e.g., IBM Watson Text-to-Speech or Amazon Polly) and converted into audio data, which generates natural-sounding audio data based on the translated English text.

[0902] The server sends the generated voice data to the other device. The other device has the function to play the received voice data and outputs the voice from a speaker. This allows the other user to understand what the user is saying through the English voice, even if they do not understand Japanese.

[0903] Specific examples

[0904] As a concrete example, consider a conversation between Japanese user A and American user B, both of whom are attending an international conference. User A says in Japanese, "Hello, how is the project progressing?", which is played back in English on User B's device. User B also says in English, "The project is progressing well, thank you.", which is played back in Japanese on User A's device. This allows the two parties to communicate smoothly without feeling any language barrier.

[0905] Prompt Sentence Examples

[0906] "Please translate the following Japanese text into natural English and synthesize it into speech: 'Hello, how's the project going?'"

[0907] The system of the present invention combines real-time speech recognition, text completion, translation, and speech synthesis technologies over the Internet to significantly improve international communication.

[0908] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0909] Program processing flow

[0910] Step 1:

[0911] The user begins speaking in their own language. The speech is collected as voice data through the device's microphone. The input is the user's speech, and the output is the collected voice data. The device's specific operation is to convert the analog voice signal via the microphone into digital voice data and store it in its internal memory.

[0912] Step 2:

[0913] The device sends the collected voice data to the server via the Internet. The input is the voice data collected in step 1, and the output is the transmitted voice data. Specifically, the device prepares to transmit the voice data to the server using a protocol such as WebSocket, and then begins transmission.

[0914] Step 3:

[0915] The server calls a speech recognition engine to convert the received voice data into text data. For example, it uses the Google Speech-to-Text API. The input is the voice data sent from the device, and the output is text data. Specifically, the server sends the voice data to the API and receives the returned text data.

[0916] Step 4:

[0917] The server uses a generative AI model to complete the content of the received text data. For example, OpenAI GPT-3 is used as this model. The input is the text data converted in step 3, and the output is the completed text data. Specifically, the server inputs the text data into the generative AI model and receives the completed text.

[0918] Step 5:

[0919] The server sends the completed text data to a multilingual translation engine to translate it into other languages. For example, it uses DeepL or Google Translate API. The input is the text data completed in step 4, and the output is the translated text data. Specifically, the server sends the text data to the translation engine and receives the translated results.

[0920] Step 6:

[0921] The server inputs the translated text data into a speech synthesis engine and converts it into voice data. For example, IBM Watson Text-to-Speech or Amazon Polly can be used. The input is the text data translated in step 5, and the output is the generated voice data. Specifically, the server sends the text data to the speech synthesis engine and receives the generated voice data.

[0922] Step 7:

[0923] The server transmits the generated voice data to the other party's terminal via the Internet. The input is the voice data generated in step 6, and the output is the transmitted voice data. Specifically, the server performs the transmission process using a protocol for transmitting voice data to the other party's terminal.

[0924] Step 8:

[0925] The other device plays the received voice data. The input is the voice data sent from the server, and the output is the voice played through the speaker. Specifically, the device decodes the received voice data and plays it back through the speaker. This allows the other user to understand what the user is saying.

[0926] (Application example 1)

[0927] 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."

[0928] Establishing real-time communication between users who speak different languages ​​is a major challenge, especially in international business and tourism. There is also a demand for seamless audio and visual information delivery using devices such as smart glasses. Existing technologies struggle to fully meet these demands, particularly in terms of accuracy and speed. Therefore, a new system is needed to provide real-time translated audio and visual information between users who speak different languages.

[0929] 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.

[0930] In this invention, the server includes: means for collecting speech content from users; means for transmitting the collected speech content as voice data; means for converting the received voice data into text; means for completing the converted text; means for translating the completed text into another language; means for converting the translated text into voice data; means for transmitting the converted voice data to a remote terminal; means for playing the received voice data at the remote terminal; means for realizing real-time language translation between a service provider and a customer and providing voice and visual information through the smart glasses; and means for detecting the user's speech using a microphone and speaker of the smart glasses and providing voice feedback to the remote terminal in the translated language, thereby enabling real-time translation of voice and visual information between users speaking different languages.

[0931] "User" refers to a person who uses the system to translate spoken content.

[0932] "Speech content" refers to what the user says in natural language.

[0933] "Means for collecting" refers to the function of acquiring the user's speech content as voice data.

[0934] "Transmitting means" refers to the function of transmitting collected voice data to an external device such as a server.

[0935] "Means for converting to text" refers to a function for converting received voice data into character data.

[0936] "Means for supplementing" refers to the function of adding supplementary information based on the context of the converted text content.

[0937] "Means for translating" refers to a function for translating the completed text into a specified other language.

[0938] "Means for converting into voice data" refers to the function of converting the translated text into data for voice output.

[0939] "Means for transmitting" refers to the function of transferring voice data to the other party's terminal.

[0940] "Means for playing" refers to the function of actually outputting the audio data received on the other party's terminal as audio.

[0941] A "service provider" refers to the person who interacts with customers at a store or facility.

[0942] "Customer" means a person who uses the System for the purpose of communicating with a Service Provider.

[0943] "Smart glasses" are glasses-type devices that utilize built-in microphones and speakers to facilitate interaction between the user and the system.

[0944] "Voice feedback" refers to the function of translating a user's speech and providing it as audio.

[0945] The system of the present invention translates what a user says in their own language into another language in real time and presents the translated speech to the other party as audio, thereby realizing smooth communication between users who speak different languages. Specific embodiments for carrying out the present invention will be described below.

[0946] A means of generating a program

[0947] The programs running on the server use the following software:

[0948] speech_recognition: Collects speech from the microphone and converts it into text.

[0949] googletrans: Translate the converted text into other languages.

[0950] gtts: Converts translated text into audio data.

[0951] playsound: Plays the generated sound data.

[0952] Hardware

[0953] The system uses smart glasses as its primary device, which includes the following hardware:

[0954] Microphone: Collects user utterances.

[0955] Speaker: Plays the translated audio.

[0956] Display: Visually display the completed text information.

[0957] Data processing and calculation

[0958] The server processes the data in the following steps:

[0959] 1. Voice data collection: The user's speech is collected through the microphone of the smart glasses. The collected voice data is temporarily stored in the memory of the smart glasses.

[0960] 2. Sending voice data: The smart glasses send the collected voice data to the server.

[0961] 3. Speech recognition and text conversion: The server converts the voice data into text using "speech_recognition".

[0962] 4. Content completion: Using a "generative AI model," the context of the converted text is analyzed and information is completed as needed.

[0963] 5. Text translation: The server translates the completed text into other languages ​​using "googletrans".

[0964] 6. Speech synthesis: The translated text is converted into voice data by "gtts".

[0965] 7. Sending voice data: The converted voice data is sent to the speaker of the smart glasses.

[0966] 8. Playback of audio data: The translated audio will be played through the smart glasses' speaker.

[0967] Specific examples

[0968] For example, if a Japanese store clerk says "Here is the product," the microphone in the smart glasses collects the voice. The voice data is sent to the server and converted into Japanese text "Here is the product." The text is then translated into English, generating the English text "Here is the product." This English text is converted into speech and played back from the speaker in the customer's smart glasses as "Here is the product." In this way, the foreign customer can understand the explanation of the store clerk.

[0969] Prompt Sentence Examples

[0970] User: Says, "Hi, how's the project going?"

[0971] Server: Identify the language and provide the translated results in text format.

[0972] Output: "Hello, regarding the project's progress, how is it going?"

[0973] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0974] Step 1:

[0975] A user puts on the smart glasses and starts speaking. The microphone in the smart glasses collects the speech as voice data. The collected voice data is temporarily stored in the memory in the smart glasses.

[0976] Input: User's speech

[0977] Output: Audio data in smart glasses

[0978] Step 2:

[0979] The device (smart glasses) transmits the collected voice data to a server in real time, using an internet connection to transfer the data.

[0980] Input: Audio data in smart glasses

[0981] Output: Audio data sent to the server

[0982] Step 3:

[0983] The server converts the received voice data into text using the "speech_recognition" library. Through this conversion, the user's speech content is obtained as text data.

[0984] Input: Audio data sent to the server

[0985] Output: Converted text data

[0986] Step 4:

[0987] The server then uses a generative AI model to analyze the converted text and complete the information based on the context, thereby fleshing out any ambiguities or missing information.

[0988] Input: Converted text data

[0989] Output: Completed text data

[0990] Step 5:

[0991] The server uses "googletrans" to translate the completed text into the specified language, resulting in the user's speech being expressed in another language.

[0992] Input: Completed text data

[0993] Output: Translated text data

[0994] Step 6:

[0995] The server converts the translated text into audio data using "gtts" (Google Text-to-Speech), which then creates a playable audio format.

[0996] Input: Translated text data

[0997] Output: Generated audio data

[0998] Step 7:

[0999] The server then transmits the generated audio data to the smart glasses, again using an internet connection to transfer the data.

[1000] Input: Generated audio data

[1001] Output: Audio data sent to the smart glasses

[1002] Step 8:

[1003] The device (smart glasses) plays the received audio data through a speaker, so that the user's speech is provided as audio in the translated language.

[1004] Input: Audio data transmitted to smart glasses

[1005] Output: Playback of the translated audio

[1006] In this way, the system of the present invention translates the user's speech in real time and provides it as audio, thereby realizing smooth communication between users who speak different languages.

[1007] 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.

[1008] The system of the present invention translates what a user says in their own language into another language in real time, and also has the function of recognizing the user's emotions and reflecting them in the translation. By recognizing the user's emotions, more natural and emotional communication is realized. The following describes in detail an embodiment of the present invention.

[1009] Collection of speech content

[1010] The user speaks in their own language. For example, Japanese user A says, "Hello, how is the project going?" The content of this speech is collected by the user's device (smartphone or computer) through a microphone.

[1011] Sending audio data

[1012] The device transmits the collected voice data in real time to a server via the Internet. The voice data contains the user's speech in digital form.

[1013] Speech recognition and text conversion

[1014] The server inputs the voice data received via the internet into the generative AI model and converts the voice to text. For example, the generated text is "Hello, how is the project going?"

[1015] emotion recognition

[1016] The server uses the voice and text data to recognize the user's emotions through an emotion engine. For example, the server can recognize emotions such as "interest" and "concern" from User A's tone of voice and speaking style.

[1017] Content Complement

[1018] The server completes the generated text using a context analysis algorithm. For example, the text "Hello, how is the project going?" is completed as "Hello, regarding the project progress, what is the status?"

[1019] translation

[1020] The server translates the completed text into the target language and reflects the recognized sentiment in the translation result. For example, when translating the completed Japanese text into English and generating the English text "Hello, regarding the project's progress, how is it going?", the server reflects that User A is "interested" or "concerned."

[1021] Speech synthesis

[1022] The server converts the translated text into speech data using speech synthesis technology, generating English speech data such as "Hello, regarding the project's progress, how is it going?" with emotional intonation.

[1023] Sending and playing audio data

[1024] The generated voice data is again sent from the server to the other device (User B's device). The other device receives this voice data and plays it through the speaker. User B can hear the English voice "Hello, regarding the project's progress, how is it going?" with the emotion reflected.

[1025] As a concrete example, consider a conversation between Japanese user A and American user B, both of whom are participating in an international conference. User A speaks in Japanese, which is played back as English audio on User B's device, with User A's emotions also being conveyed. Similarly, User B speaks in English, which is played back as Japanese audio on User A's device, with User A's emotions also being reflected. This allows both parties to communicate their emotions accurately without feeling a language barrier, resulting in more natural and effective communication.

[1026] The system of the present invention recognizes the user's emotions throughout the entire process, from voice data collection, transmission, text conversion, content completion, translation, voice synthesis, and finally transmission and playback of the voice data, and reflects these emotions in the translation results, thereby significantly improving international communication.

[1027] The processing flow will be explained below.

[1028] Step 1:

[1029] The user speaks in their own language. For example, Japanese user A says, "Hello, how is the project going?"

[1030] Step 2:

[1031] The terminal (User A's device) records the speech using a microphone and captures it as voice data, which is then temporarily stored in the terminal in digital format.

[1032] Step 3:

[1033] The device transmits the collected voice data to a server via the Internet in real time. The transmitted data includes the voice data as well as the user ID and device information.

[1034] Step 4:

[1035] The server inputs the received voice data into the generative AI model and converts it into text using speech recognition technology. This conversion process generates the text, "Hello, how is the project going?"

[1036] Step 5:

[1037] The server uses the voice and text data to recognize the user's emotions through an emotion engine. For example, the server can recognize emotions such as "interest" and "concern" from User A's tone of voice and speaking style.

[1038] Step 6:

[1039] The server completes the generated text using a context analysis algorithm. For example, the text "Hello, how is the project going?" is completed as "Hello, regarding the project progress, what is the status?"

[1040] Step 7:

[1041] The server translates the completed text into the target language and reflects the recognized sentiment in the translation result. For example, when translating the completed Japanese text into English and generating the English text "Hello, regarding the project's progress, how is it going?", the server reflects that User A is "interested" or "concerned."

[1042] Step 8:

[1043] The server converts the translated text into speech data using speech synthesis technology, generating English speech data such as "Hello, regarding the project's progress, how is it going?" with emotional intonation.

[1044] Step 9:

[1045] The server transmits the generated translated speech data via the Internet to the other party's terminal (user B's device). The transmitted data includes the translated speech data.

[1046] Step 10:

[1047] The device (User B's device) plays the received voice data through the device's speaker. User B can hear the English voice "Hello, regarding the project's progress, how is it going?" with the emotion reflected.

[1048] Step 11:

[1049] User B (American) says in English, "The project is progressing well, thank you for asking."

[1050] Step 12:

[1051] The device (User B's device) records the speech with a microphone and captures the audio data in real time, which is then temporarily stored in the device in digital format.

[1052] Step 13:

[1053] The device transmits the collected voice data to a server via the Internet in real time. The transmitted data includes the voice data as well as the user ID and device information.

[1054] Step 14:

[1055] The server inputs the received voice data into the generative AI model and converts it into text using speech recognition technology. For example, the generated text might read, "The project is progressing well, thank you for asking."

[1056] Step 15:

[1057] The server uses the voice and text data to recognize the user's emotions using an emotion engine. For example, emotions such as "gratitude" and "relief" can be recognized from User B's tone of voice and speaking style.

[1058] Step 16:

[1059] The server uses a context analysis algorithm to complete the generated text. For example, the text "The project is progressing well, thank you for asking" is completed as "The project is progressing well, thank you for asking about the project's status."

[1060] Step 17:

[1061] The server translates the completed text into the target language and reflects the recognized emotions in the translation result. For example, when translating the completed English text into Japanese and generating the Japanese text "The project is progressing smoothly. Thank you for your inquiry," the "gratitude" and "relief" that User B feels are reflected.

[1062] Step 18:

[1063] The server then converts the translated Japanese text into speech data using speech synthesis technology, generating Japanese speech data that says, "The project is progressing smoothly. Thank you for your inquiry," with intonation that reflects the emotion.

[1064] Step 19:

[1065] The server transmits the generated translated speech data to the other party's terminal (user A's device) via the Internet. The transmitted data includes the translated speech data.

[1066] Step 20:

[1067] The terminal (user A's device) plays the received voice data through the device's speaker. User A can hear the voice in his / her native language saying "The project is progressing smoothly. Thank you for your inquiry." with the emotion reflected.

[1068] Example 2

[1069] 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."

[1070] In today's global society, people who speak different languages ​​need to communicate smoothly. However, while conventional translation systems can accurately translate spoken content, they struggle to convey the speaker's emotions appropriately. As a result, misunderstandings and misrecognitions are likely to occur, hindering natural communication. Furthermore, they often fail to properly complement the spoken content by taking into account the context of the spoken content, making it impossible to convey the speaker's full intention.

[1071] 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.

[1072] In this invention, the server includes a means for recognizing the user's emotions, a means for completing the content of the converted text, and a means for converting the translated text into voice data. This not only ensures accurate translation of the spoken content, but also enables the speaker's emotions to be conveyed to the other party in a manner that appropriately reflects their intention. Furthermore, context-based content completion allows the speaker to fully convey their intention, resulting in more natural and effective communication.

[1073] "User" refers to a person who uses the system to translate spoken content into another language.

[1074] The term "terminal" refers to an electronic device that collects the user's speech, transmits the voice data to a server, and plays back the voice data transmitted from the server.

[1075] "Server" refers to a central control device that processes voice data sent from a user's terminal and performs translation and voice synthesis.

[1076] "Speech content" refers to words and sentences spoken by a user in their own language.

[1077] "Voice data" refers to data that represents the contents of a user's speech in digital form.

[1078] "Text" refers to data that has been converted from audio data into character information.

[1079] "Emotion" refers to the emotional elements (e.g., interest, concern, joy, etc.) contained in the user's utterance.

[1080] "Content completion" refers to the process of analyzing the context of the spoken content and completing incomplete parts in a natural way.

[1081] "Translation" refers to the process of converting text expressed in one language into another language.

[1082] "Speech synthesis" refers to the process of converting text data into voice data.

[1083] "Emotion recognition" refers to the process of analyzing and identifying emotional elements from a user's speech.

[1084] "Real-time" means that processing occurs with little to no delay.

[1085] The system of the present invention translates what a user says in their own language into another language in real time, and further recognizes the user's emotions and reflects them in the translation, thereby realizing natural and emotional communication. A specific embodiment of this system will be described below.

[1086] Collection of speech content

[1087] The user speaks in their own language. For example, a Japanese user might say, "Hello, how's the project going?" This speech is collected by the user's device (smartphone or computer) via a microphone.

[1088] Sending audio data

[1089] The device transmits the collected voice data to a server over the Internet in real time. The voice data contains the user's speech in digital form. The device transmits the voice data using a secure communication protocol (e.g., HTTPS).

[1090] Speech recognition and text conversion

[1091] The server inputs the voice data received via the internet into a generative AI model and converts the voice into text. For example, the generative AI model used is the Google Cloud Speech-to-Text API, and generates the text "Hello, how is the project progressing?" from Japanese voice data.

[1092] emotion recognition

[1093] The server uses voice and text data to recognize the user's emotions through an emotion engine. For example, the emotion engine used is IBM Watson Tone Analyzer, which recognizes emotions such as "interest" and "concern" from the user's tone of voice and speaking style.

[1094] Content Complement

[1095] The server uses a context analysis algorithm to complete the generated text. For example, the context analysis algorithm used is OpenAI GPT-3, and the text "Hello, how is the project going?" is completed as "Hello, regarding the progress of the project, what is the status?"

[1096] translation

[1097] The server translates the completed text into the target language and reflects the recognized sentiment in the translation results. For example, the translation engine used is the Google Translate API, and the completed Japanese text is translated into English to generate the English text "Hello, regarding the project's progress, how is it going?". In this case, the user's "interest" and "concern" are reflected in the translation results.

[1098] Speech synthesis

[1099] The server converts the translated text into voice data using Amazon Polly speech synthesis technology. The generated voice data is an English voice saying "Hello, regarding the project's progress, how is it going?" with intonation that reflects the emotion.

[1100] Sending and playing audio data

[1101] The generated voice data is again sent from the server to the other device (User B's device). The other device receives this voice data and plays it through the speaker. User B can hear the English voice saying "Hello, regarding the project's progress, how is it going?" with the emotion reflected.

[1102] As a concrete example, consider a conversation between a Japanese user and an American user participating in an international conference. The Japanese user speaks in Japanese, which is played back as English speech on the American user's device, with the Japanese user's emotions also being conveyed. Similarly, the American user speaks in English, which is played back as Japanese speech on the Japanese user's device, with the Japanese user's emotions also being reflected.

[1103] Prompt Sentence Examples

[1104] "A user says, 'Hello, how's the project going?' in Japanese. Please translate this utterance into English and convert it into speech data that reflects the sentiment."

[1105] This system will significantly improve international communication, enabling more natural and emotional interactions.

[1106] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1107] Step 1:

[1108] The user speaks in their own language. The user's voice is collected by the device as input. For example, a Japanese user might say, "Hello, how's the project going?". Specifically, the device's microphone collects the voice in digital form. Digital voice data is generated as output.

[1109] Step 2:

[1110] The device transmits the collected audio data in real time to a server in uncompressed or compressed format. Digital audio data is input. Specifically, the device transmits the audio data using a secure communication protocol such as HTTPS. The audio data arrives at the server as output.

[1111] Step 3:

[1112] The server inputs the received voice data into a generative AI model and converts the voice to text. The voice data received on the server side exists as input. Specifically, it uses the Google Cloud Speech-to-Text API to convert it into text data such as "Hello, how is the project progressing?". Text data is generated as output.

[1113] Step 4:

[1114] The server uses voice and text data to recognize the user's emotions through an emotion engine. Converted text data and voice data are used as input. Specifically, the server uses IBM Watson Tone Analyzer to identify emotions such as "interest" or "concern" from the user's tone of voice and choice of words. Emotional data is generated as output.

[1115] Step 5:

[1116] The server analyzes the context of the generated text data and completes the content. Emotion data and text data are input. Specifically, OpenAI GPT-3 is used to complete the text "Hello, how is the project progressing?" to "Hello, regarding the progress of the project, what is the status?". The completed text data is generated as output.

[1117] Step 6:

[1118] The server translates the completed text into the target language and also reflects the recognized emotion in the translation result. The input is the completed text data and emotion data. Specifically, it uses the Google Translate API to translate the completed text "Hello, regarding the project's progress, how is it going?" into English and reflects the emotion to generate "Hello, regarding the project's progress, how is it going?" The translated text data is generated as the output.

[1119] Step 7:

[1120] The server converts the translated text into speech data using speech synthesis technology. The inputs are the translated text data and emotion data. Specifically, Amazon Polly is used to generate English speech data saying "Hello, regarding the project's progress, how is it going?". This speech data is given an intonation that reflects the emotion. English speech data is generated as output.

[1121] Step 8:

[1122] The server sends the generated voice data to the other party's device. The generated voice data exists as input. Specifically, the voice data is sent to User B's device via the Internet. The voice data arrives at the other party's device as output.

[1123] Step 9:

[1124] The other device plays the received audio data. The input is English audio data from the server. Specifically, the device's speaker plays the English audio, "Hello, regarding the project's progress, how is it going?". User B can hear the audio as output.

[1125] This detailed processing allows users who speak different languages ​​to communicate accurately and with rich emotion.

[1126] (Application example 2)

[1127] 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."

[1128] Currently, in order for autonomous vehicles to communicate smoothly between passengers and vehicle systems, it is necessary not only to recognize instructions but also to communicate seamlessly in various languages. It is also necessary to simultaneously recognize passenger emotions and respond appropriately to those emotions to provide a safer and more comfortable travel experience. However, current systems are unable to effectively support multiple languages ​​and recognize emotions, resulting in reduced passenger satisfaction and safety.

[1129] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting utterances from users, means for transmitting the collected utterances as voice data, means for converting the received voice data into text, means for completing the converted text, means for translating the completed text into another language, means for converting the translated text into voice data, means for transmitting the converted voice data to the other party's terminal, means for performing emotion recognition, means for generating voice data based on the recognized emotion, means for playing the received voice data at the other party's terminal, means for giving instructions to the vehicle system or setting a destination based on the passenger's utterance, and means for recognizing the emotion of the utterance and performing a corresponding action. This enables natural communication that recognizes emotions in multiple languages, thereby realizing safe and comfortable operation of autonomous vehicles.

[1130] The "means for collecting speech content from the user" refers to a device or function for recognizing the speech uttered by the user and acquiring it as digital data.

[1131] "Means for transmitting collected speech content as voice data" refers to a mechanism for sending acquired voice data to a server via a communication network such as the Internet.

[1132] The "means for converting received voice data into text" refers to a system that analyzes voice data on the server side and converts it into a corresponding text format.

[1133] "Means for completing the content of the converted text" refers to technologies and algorithms that analyze the context of the converted text, add information as needed, and arrange the sentences in a natural format.

[1134] The "means for translating the completed text into another language" is an automatic translation technology for converting the completed text into another specified language.

[1135] "Means for converting translated text into speech data" refers to technology or devices for converting translated text into synthetic speech.

[1136] The "means for transmitting the converted voice data to the other terminal" is the communication means used to retransmit the generated voice data to the target device.

[1137] "Means for emotion recognition" refers to technologies and algorithms that analyze and recognize a speaker's emotional state from audio data or text data.

[1138] The "means for generating voice data based on recognized emotions" refers to a technology that adjusts the intonation and atmosphere of the synthesized voice by reflecting the recognized emotional information.

[1139] The "means for playing back the received audio data on the other party's terminal" is a function for playing back the transmitted audio data on the device.

[1140] "Means for giving instructions to vehicle systems and setting destinations based on passenger speech" refers to technology that recognizes what passengers say and operates the vehicle's navigation system and other operations based on that speech.

[1141] "Means for recognizing emotions in speech and taking corresponding actions" refers to technologies and functions that allow the system to automatically respond or take appropriate actions based on the recognized emotional information.

[1142] DETAILED DESCRIPTION OF THE INVENTION The present invention provides a voice interface for autonomous vehicles with multilingual support and emotion recognition capabilities.

[1143] Hardware and Software Configuration

[1144] 1. Hardware:

[1145] Microphone: Used to collect passenger speech.

[1146] Speaker: Used to play back the generated audio data.

[1147] A smartphone or in-vehicle computer: Used to process the voice data.

[1148] 2. Software:

[1149] Speech Recognition Library: speech_recognition

[1150] Translation library: googletrans

[1151] Emotion Recognition Model: A sentiment analysis model using the transformers library

[1152] Text-to-speech libraries: gTTS and playsound

[1153] Data processing and calculation flow

[1154] 1. Voice input: Using the device's microphone, we collect speech in Japanese uttered by the user. For example, a passenger might say, "Please go to Shinjuku Station."

[1155] 2. Speech Recognition: Using the speech_recognition library, the device converts the collected voice data into text format.

[1156] 3. Translation: Use googletrans library to translate the converted Japanese text into English. For example, "Please take me to Shinjuku Station" is translated to "Please take me to Shinjuku Station".

[1157] 4. Emotion Recognition: Use the transformers library to recognize user emotions from translated text, for example, "tired" from the tone and content of the speech.

[1158] 5. Emotion-sensitive speech synthesis: Using the gTTS library, the translated text is converted into synthetic speech, and an intonation is added to reflect the recognized emotion, for example, "Please take me to Shinjuku Station" in a tired voice.

[1159] 6. Audio output: The generated audio data is sent to the other device and played through the speaker. Based on the passenger's request, the navigation system sets the destination and the autonomous vehicle begins operation.

[1160] Specific examples

[1161] For example, if a passenger says in Japanese, "I'm a little tired, so please go directly to the airport," the system will act as follows:

[1162] 1. The microphone collects the speech and the device performs voice recognition.

[1163] 2. Translate the recognized text "I'm a bit tired, please take me directly to the airport." using the googletrans library to get "I'm a bit tired, please take me directly to the airport."

[1164] 3. Emotion recognition is performed on the translated text to obtain the emotion label "Tired."

[1165] 4. The speech that reflects the recognized emotion "tired" is synthesized using the gTTS library to generate speech in a format that conveys the emotion in English.

[1166] 5. The generated audio is played through the speaker, the navigation system sets the destination according to the instructions, and the vehicle begins to travel along the route.

[1167] Prompt Sentence Examples

[1168] The user says in Japanese, "I'm a little tired, so please go directly to the airport." Translate this into English in real time and output the following voice output, reflecting the user's emotions.

[1169] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1170] Step 1:

[1171] The user performs voice input. The user utters "Please go to Shinjuku Station" in Japanese. The contents of this utterance are collected through the device's microphone.

[1172] Step 2:

[1173] The device converts the collected voice data into text format in real time using the speech recognition library speech_recognition. The input is voice data, and the output is text data: "Please go to Shinjuku Station."

[1174] Step 3:

[1175] After the text data is generated, the device translates this text into the target language (in this case, English) using the translation library googletrans. The input is the text data "Please take me to Shinjuku Station," and the output is the English text "Please take me to Shinjuku Station."

[1176] Step 4:

[1177] The server inputs the translated English text into an emotion recognition model (using the transformers library) to recognize the user's emotion. The input is the English text "Please take me to Shinjuku Station", and the output is an emotion label (e.g., "Tired").

[1178] Step 5:

[1179] The server uses the speech synthesis library gTTS to generate speech data reflecting the recognized emotions. The input is English text and emotion labels, and the output is speech data reflecting the emotions.

[1180] Step 6:

[1181] The device sends the generated voice data to the navigation system, which then uses it to set destinations and plan routes for the autonomous vehicle. The input is the voice data, and the output is the vehicle's instructed destination and route.

[1182] Step 7:

[1183] Finally, the terminal plays the generated voice data through a speaker to provide voice feedback to the passenger, facilitating user communication with the autonomous vehicle system.

[1184] 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.

[1185] 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.

[1186] 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.

[1187] [Fourth embodiment]

[1188] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1189] 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.

[1190] 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).

[1191] 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.

[1192] 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.

[1193] 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).

[1194] 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.

[1195] 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.

[1196] 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.

[1197] 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.

[1198] 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.

[1199] 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.

[1200] 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."

[1201] The system of the present invention translates what a user says in their own language into another language in real time and provides the translated speech to the other party. This enables smooth communication that transcends language barriers at international conferences, etc. The following describes in detail the embodiments of the present invention.

[1202] Collection of speech content

[1203] The user begins speaking in their own language. For example, the user is Japanese and says, "Hello, how is the project going?" This speech is collected by the user's device (smartphone or computer) via a microphone.

[1204] Sending audio data

[1205] The device transmits the collected voice data in real time to a server via the Internet. The voice data contains the user's speech in digital form.

[1206] Speech recognition and text conversion

[1207] The server receives voice data via the internet and inputs it into the generative AI model to convert the voice to text. At this time, speech recognition technology is used to accurately convert what the user has said into text format.

[1208] Content Complement

[1209] The server uses a generative AI model to analyze the context of the converted text and fill in any missing or ambiguous information. For example, if the generated text is "Hello, how's the project going?", it will be completed to make the meaning more specific: "Hello, regarding the project progress, what's the status?"

[1210] translation

[1211] The server translates the completed text data into the target language using a translation function. For example, the server translates the completed Japanese text into English, converting it into the English text "Hello, regarding the project's progress, how is it going?"

[1212] Speech synthesis

[1213] The server converts the translated text into speech data using speech synthesis technology, which prepares the translated content as speech output. For example, an English voice is generated based on the English text "Hello, regarding the project's progress, how is it going?"

[1214] Sending and playing audio data

[1215] The generated voice data is then sent from the server to the other party's device, which receives the voice data and plays it back through its speaker. In this way, even if the other user does not understand Japanese, they can understand what the Japanese person is saying through the English voice.

[1216] As a concrete example, consider a conversation between Japanese user A and American user B, both of whom are participating in an international conference. User A speaks in Japanese, which is played back as English audio on User B's device. User B also speaks in English, which is played back as Japanese audio on User A's device. This allows the two parties to communicate smoothly without feeling any language barrier.

[1217] The system of the present invention significantly improves international communication by processing users' speech in real time, converting it into other languages, and providing it to the other party. By exchanging voice and text data between the server and the terminals via the Internet, participants of multiple nationalities can experience a conversation as if they were speaking only in their own native language.

[1218] The processing flow will be explained below.

[1219] Step 1:

[1220] The user speaks in their own language. For example, Japanese user A says, "Hello, how is the project going?"

[1221] Step 2:

[1222] The terminal (User A's device) records the speech using a microphone and captures it as voice data, which is then temporarily stored in the terminal in digital format.

[1223] Step 3:

[1224] The device transmits the collected voice data to a server via the Internet in real time. The transmitted data includes the voice data as well as the user ID and device information.

[1225] Step 4:

[1226] The server inputs the received voice data into the generative AI model and converts it into text using speech recognition technology. This conversion process generates the text, "Hello, how is the project going?"

[1227] Step 5:

[1228] The server completes the generated text using a context analysis algorithm. For example, the text "Hello, how is the project going?" is completed as "Hello, regarding the project progress, what is the status?"

[1229] Step 6:

[1230] The server translates the completed text into the target language, for example translating the completed Japanese text into English to generate the English text "Hello, regarding the project's progress, how is it going?"

[1231] Step 7:

[1232] The server converts the translated text into speech data using speech synthesis technology, generating the English speech data "Hello, regarding the project's progress, how is it going?"

[1233] Step 8:

[1234] The server transmits the generated translated speech data via the Internet to the other party's terminal (user B's device). The transmitted data includes the translated speech data.

[1235] Step 9:

[1236] The terminal (user B's device) plays the received voice data through the device's speaker. User B can hear the English voice saying "Hello, regarding the project's progress, how is it going?"

[1237] Step 10:

[1238] User B (American) says in English, "The project is progressing well, thank you for asking."

[1239] Step 11:

[1240] The terminal (user B's device) records the speech using a microphone and captures the audio data in real time.

[1241] Step 12:

[1242] The terminal transmits the collected voice data to a server via the Internet in real time.

[1243] Step 13:

[1244] The server inputs the received voice data into the generative AI model and converts it into text using speech recognition technology. For example, the generated text might read, "The project is progressing well, thank you for asking."

[1245] Step 14:

[1246] The server uses a context analysis algorithm to complete the generated text, for example, "The project is progressing well, thank you for asking about the project's status."

[1247] Step 15:

[1248] The server translates the completed text into the target language, for example, translating the completed English text into Japanese to generate Japanese text such as "The project is progressing smoothly. Thank you for your inquiry."

[1249] Step 16:

[1250] The server converts the translated Japanese text into voice data using speech synthesis technology, generating Japanese voice data that says, "The project is progressing smoothly. Thank you for your inquiry."

[1251] Step 17:

[1252] The server transmits the generated translated speech data to the other party's terminal (user A's device) via the Internet.

[1253] Step 18:

[1254] The terminal (user A's device) plays the received voice data through the device's speaker. User A can hear the voice in his / her native language saying, "The project is progressing smoothly. Thank you for your inquiry."

[1255] Example 1

[1256] 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."

[1257] Conventional real-time translation systems have faced problems with the accuracy of translation of spoken content and a lack of context-based completion. Furthermore, when communicating between different languages, each step of speech recognition, translation, and speech synthesis is fragmented, which can impair real-time performance. There is a need to solve these problems and realize smoother, more accurate international communication.

[1258] 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.

[1259] In this invention, the server includes means for collecting user utterances, means for transmitting the collected utterances as voice data, means for converting the received voice data into text, means for completing the converted text using a generative AI model, means for translating the completed text into another language, means for converting the translated text into voice data, means for transmitting the generated voice data to a partner device, and means for playing the received voice data on the partner device. This enables the user's utterances to be translated accurately between different languages ​​in real time and for natural completion based on context.

[1260] A "user" is a person who uses the system of the present invention to input speech content.

[1261] "Speech content" refers to words and sentences spoken by the user, and is the speech data to be translated.

[1262] "Voice data" is data that expresses the contents of a user's speech in digital form.

[1263] "Text data" refers to data that has been converted from voice data into character information using voice recognition technology.

[1264] A "generative AI model" is a model that uses artificial intelligence technology to complement the content of text data, for example, natural language processing technology.

[1265] "Completion" is the process of adding contextual information to incomplete text data to make it more meaningful.

[1266] The "translation function" is a function for converting text data written in one language into another language.

[1267] "Speech synthesis technology" is a technology that generates voice data based on text data.

[1268] The "partner's device" is a receiving device used by the other party who is having a conversation with the user, and has the function of reproducing voice data.

[1269] The "collection means" is a means for acquiring the contents of a user's speech using a device such as a microphone.

[1270] The "transmission means" is a means for transmitting collected or generated voice data via the Internet or the like.

[1271] The "conversion means" is a means for converting voice data into text data and text data into voice data.

[1272] "Completion means" refers to a means of complementing the content of text data using a generative AI model.

[1273] The "playback means" is a function for playing back the audio data received by the other device from a speaker or the like.

[1274] MODE FOR CARRYING OUT THE INVENTION

[1275] The present invention provides a system that translates speech in a user's own language into another language in real time and provides the translated speech to the other party. This system enables smooth communication across language barriers in situations such as international conferences. Specific embodiments for implementing the present invention will be described below, specifying the hardware and software used.

[1276] This system consists of three main elements: the user, the terminal, and the server. The roles and specific operations of each are explained in detail below.

[1277] User Roles

[1278] Users speak in their own language. For example, a Japanese user might say, "Hello, how's the project going?" This speech is collected as voice data through a microphone installed on a device such as a smartphone or PC.

[1279] Device Role

[1280] The device sends the voice data collected from the user to a server via the Internet. The voice data is encoded in a digital format (e.g., PCM format). WebSocket or similar is often used as a communication protocol to maintain real-time performance.

[1281] Server Roles

[1282] The server performs the following sequence of operations:

[1283] 1. Speech Recognition:

[1284] The server receives the voice data and converts it into text data using a speech recognition engine (e.g., Google Speech-to-Text API). This conversion converts the user's speech into text.

[1285] 2. Content Complement:

[1286] The converted text data is input into a generative AI model (e.g., OpenAI GPT-3). The server uses this generative AI model to analyze the context of the text and fill in missing or ambiguous information. For example, for the text "Hello, how is the project progressing?", it generates a more specific completion text: "Hello, regarding the progress of the project, what is the status?"

[1287] 3. Translation:

[1288] The completed text data is sent to a translation engine (e.g., DeepL or Google Translate API). The server uses this engine to translate the text data into another language. For example, translating Japanese text into English yields the English text "Hello, regarding the project's progress, how is it going?"

[1289] 4. Speech synthesis:

[1290] The translated text is input into a speech synthesis engine (e.g., IBM Watson Text-to-Speech or Amazon Polly) and converted into audio data, which generates natural-sounding audio data based on the translated English text.

[1291] The server sends the generated voice data to the other device. The other device has the function to play the received voice data and outputs the voice from a speaker. This allows the other user to understand what the user is saying through the English voice, even if they do not understand Japanese.

[1292] Specific examples

[1293] As a concrete example, consider a conversation between Japanese user A and American user B, both of whom are attending an international conference. User A says in Japanese, "Hello, how is the project progressing?", which is played back in English on User B's device. User B also says in English, "The project is progressing well, thank you.", which is played back in Japanese on User A's device. This allows the two parties to communicate smoothly without feeling any language barrier.

[1294] Prompt Sentence Examples

[1295] "Please translate the following Japanese text into natural English and synthesize it into speech: 'Hello, how's the project going?'"

[1296] The system of the present invention combines real-time speech recognition, text completion, translation, and speech synthesis technologies over the Internet to significantly improve international communication.

[1297] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1298] Program processing flow

[1299] Step 1:

[1300] The user begins speaking in their own language. The speech is collected as voice data through the device's microphone. The input is the user's speech, and the output is the collected voice data. The device's specific operation is to convert the analog voice signal via the microphone into digital voice data and store it in its internal memory.

[1301] Step 2:

[1302] The device sends the collected voice data to the server via the Internet. The input is the voice data collected in step 1, and the output is the transmitted voice data. Specifically, the device prepares to transmit the voice data to the server using a protocol such as WebSocket, and then begins transmission.

[1303] Step 3:

[1304] The server calls a speech recognition engine to convert the received voice data into text data. For example, it uses the Google Speech-to-Text API. The input is the voice data sent from the device, and the output is text data. Specifically, the server sends the voice data to the API and receives the returned text data.

[1305] Step 4:

[1306] The server uses a generative AI model to complete the content of the received text data. For example, OpenAI GPT-3 is used as this model. The input is the text data converted in step 3, and the output is the completed text data. Specifically, the server inputs the text data into the generative AI model and receives the completed text.

[1307] Step 5:

[1308] The server sends the completed text data to a multilingual translation engine to translate it into other languages. For example, it uses DeepL or Google Translate API. The input is the text data completed in step 4, and the output is the translated text data. Specifically, the server sends the text data to the translation engine and receives the translated results.

[1309] Step 6:

[1310] The server inputs the translated text data into a speech synthesis engine and converts it into voice data. For example, IBM Watson Text-to-Speech or Amazon Polly can be used. The input is the text data translated in step 5, and the output is the generated voice data. Specifically, the server sends the text data to the speech synthesis engine and receives the generated voice data.

[1311] Step 7:

[1312] The server transmits the generated voice data to the other party's terminal via the Internet. The input is the voice data generated in step 6, and the output is the transmitted voice data. Specifically, the server performs the transmission process using a protocol for transmitting voice data to the other party's terminal.

[1313] Step 8:

[1314] The other device plays the received voice data. The input is the voice data sent from the server, and the output is the voice played through the speaker. Specifically, the device decodes the received voice data and plays it back through the speaker. This allows the other user to understand what the user is saying.

[1315] (Application example 1)

[1316] 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."

[1317] Establishing real-time communication between users who speak different languages ​​is a major challenge, especially in international business and tourism. There is also a demand for seamless audio and visual information delivery using devices such as smart glasses. Existing technologies struggle to fully meet these demands, particularly in terms of accuracy and speed. Therefore, a new system is needed to provide real-time translated audio and visual information between users who speak different languages.

[1318] 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.

[1319] In this invention, the server includes: means for collecting speech content from users; means for transmitting the collected speech content as voice data; means for converting the received voice data into text; means for completing the converted text; means for translating the completed text into another language; means for converting the translated text into voice data; means for transmitting the converted voice data to a remote terminal; means for playing the received voice data at the remote terminal; means for realizing real-time language translation between a service provider and a customer and providing voice and visual information through the smart glasses; and means for detecting the user's speech using a microphone and speaker of the smart glasses and providing voice feedback to the remote terminal in the translated language, thereby enabling real-time translation of voice and visual information between users speaking different languages.

[1320] "User" refers to a person who uses the system to translate spoken content.

[1321] "Speech content" refers to what the user says in natural language.

[1322] "Means for collecting" refers to the function of acquiring the user's speech content as voice data.

[1323] "Transmitting means" refers to the function of transmitting collected voice data to an external device such as a server.

[1324] "Means for converting to text" refers to a function for converting received voice data into character data.

[1325] "Means for supplementing" refers to the function of adding supplementary information based on the context of the converted text content.

[1326] "Means for translating" refers to a function for translating the completed text into a specified other language.

[1327] "Means for converting into voice data" refers to the function of converting the translated text into data for voice output.

[1328] "Means for transmitting" refers to the function of transferring voice data to the other party's terminal.

[1329] "Means for playing" refers to the function of actually outputting the audio data received on the other party's terminal as audio.

[1330] A "service provider" refers to the person who interacts with customers at a store or facility.

[1331] "Customer" means a person who uses the System for the purpose of communicating with a Service Provider.

[1332] "Smart glasses" are glasses-type devices that utilize built-in microphones and speakers to facilitate interaction between the user and the system.

[1333] "Voice feedback" refers to the function of translating a user's speech and providing it as audio.

[1334] The system of the present invention translates what a user says in their own language into another language in real time and presents the translated speech to the other party as audio, thereby realizing smooth communication between users who speak different languages. Specific embodiments for carrying out the present invention will be described below.

[1335] A means of generating a program

[1336] The programs running on the server use the following software:

[1337] speech_recognition: Collects speech from the microphone and converts it into text.

[1338] googletrans: Translate the converted text into other languages.

[1339] gtts: Converts translated text into audio data.

[1340] playsound: Plays the generated sound data.

[1341] Hardware

[1342] The system uses smart glasses as its primary device, which includes the following hardware:

[1343] Microphone: Collects user utterances.

[1344] Speaker: Plays the translated audio.

[1345] Display: Visually display the completed text information.

[1346] Data processing and calculation

[1347] The server processes the data in the following steps:

[1348] 1. Voice data collection: The user's speech is collected through the microphone of the smart glasses. The collected voice data is temporarily stored in the memory of the smart glasses.

[1349] 2. Sending voice data: The smart glasses send the collected voice data to the server.

[1350] 3. Speech recognition and text conversion: The server converts the voice data into text using "speech_recognition".

[1351] 4. Content completion: Using a "generative AI model," the context of the converted text is analyzed and information is completed as needed.

[1352] 5. Text translation: The server translates the completed text into other languages ​​using "googletrans".

[1353] 6. Speech synthesis: The translated text is converted into voice data by "gtts".

[1354] 7. Sending voice data: The converted voice data is sent to the speaker of the smart glasses.

[1355] 8. Playback of audio data: The translated audio will be played through the smart glasses' speaker.

[1356] Specific examples

[1357] For example, if a Japanese store clerk says "Here is the product," the microphone in the smart glasses collects the voice. The voice data is sent to the server and converted into Japanese text "Here is the product." The text is then translated into English, generating the English text "Here is the product." This English text is converted into speech and played back from the speaker in the customer's smart glasses as "Here is the product." In this way, the foreign customer can understand the explanation of the store clerk.

[1358] Prompt Sentence Examples

[1359] User: Says, "Hi, how's the project going?"

[1360] Server: Identify the language and provide the translated results in text format.

[1361] Output: "Hello, regarding the project's progress, how is it going?"

[1362] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1363] Step 1:

[1364] A user puts on the smart glasses and starts speaking. The microphone in the smart glasses collects the speech as voice data. The collected voice data is temporarily stored in the memory in the smart glasses.

[1365] Input: User's speech

[1366] Output: Audio data in smart glasses

[1367] Step 2:

[1368] The device (smart glasses) transmits the collected voice data to a server in real time, using an internet connection to transfer the data.

[1369] Input: Audio data in smart glasses

[1370] Output: Audio data sent to the server

[1371] Step 3:

[1372] The server converts the received voice data into text using the "speech_recognition" library. Through this conversion, the user's speech content is obtained as text data.

[1373] Input: Audio data sent to the server

[1374] Output: Converted text data

[1375] Step 4:

[1376] The server then uses a generative AI model to analyze the converted text and complete the information based on the context, thereby fleshing out any ambiguities or missing information.

[1377] Input: Converted text data

[1378] Output: Completed text data

[1379] Step 5:

[1380] The server uses "googletrans" to translate the completed text into the specified language, resulting in the user's speech being expressed in another language.

[1381] Input: Completed text data

[1382] Output: Translated text data

[1383] Step 6:

[1384] The server converts the translated text into audio data using "gtts" (Google Text-to-Speech), which then creates a playable audio format.

[1385] Input: Translated text data

[1386] Output: Generated audio data

[1387] Step 7:

[1388] The server then transmits the generated audio data to the smart glasses, again using an internet connection to transfer the data.

[1389] Input: Generated audio data

[1390] Output: Audio data sent to the smart glasses

[1391] Step 8:

[1392] The device (smart glasses) plays the received audio data through a speaker, so that the user's speech is provided as audio in the translated language.

[1393] Input: Audio data transmitted to smart glasses

[1394] Output: Playback of the translated audio

[1395] In this way, the system of the present invention translates the user's speech in real time and provides it as audio, thereby realizing smooth communication between users who speak different languages.

[1396] 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.

[1397] The system of the present invention translates what a user says in their own language into another language in real time, and also has the function of recognizing the user's emotions and reflecting them in the translation. By recognizing the user's emotions, more natural and emotional communication is realized. The following describes in detail an embodiment of the present invention.

[1398] Collection of speech content

[1399] The user speaks in their own language. For example, Japanese user A says, "Hello, how is the project going?" The content of this speech is collected by the user's device (smartphone or computer) through a microphone.

[1400] Sending audio data

[1401] The device transmits the collected voice data in real time to a server via the Internet. The voice data contains the user's speech in digital form.

[1402] Speech recognition and text conversion

[1403] The server inputs the voice data received via the internet into the generative AI model and converts the voice to text. For example, the generated text is "Hello, how is the project going?"

[1404] emotion recognition

[1405] The server uses the voice and text data to recognize the user's emotions through an emotion engine. For example, the server can recognize emotions such as "interest" and "concern" from User A's tone of voice and speaking style.

[1406] Content Complement

[1407] The server completes the generated text using a context analysis algorithm. For example, the text "Hello, how is the project going?" is completed as "Hello, regarding the project progress, what is the status?"

[1408] translation

[1409] The server translates the completed text into the target language and reflects the recognized sentiment in the translation result. For example, when translating the completed Japanese text into English and generating the English text "Hello, regarding the project's progress, how is it going?", the server reflects that User A is "interested" or "concerned."

[1410] Speech synthesis

[1411] The server converts the translated text into speech data using speech synthesis technology, generating English speech data such as "Hello, regarding the project's progress, how is it going?" with emotional intonation.

[1412] Sending and playing audio data

[1413] The generated voice data is again sent from the server to the other device (User B's device). The other device receives this voice data and plays it through the speaker. User B can hear the English voice "Hello, regarding the project's progress, how is it going?" with the emotion reflected.

[1414] As a concrete example, consider a conversation between Japanese user A and American user B, both of whom are participating in an international conference. User A speaks in Japanese, which is played back as English audio on User B's device, with User A's emotions also being conveyed. Similarly, User B speaks in English, which is played back as Japanese audio on User A's device, with User A's emotions also being reflected. This allows both parties to communicate their emotions accurately without feeling a language barrier, resulting in more natural and effective communication.

[1415] The system of the present invention recognizes the user's emotions throughout the entire process, from voice data collection, transmission, text conversion, content completion, translation, voice synthesis, and finally transmission and playback of the voice data, and reflects these emotions in the translation results, thereby significantly improving international communication.

[1416] The processing flow will be explained below.

[1417] Step 1:

[1418] The user speaks in their own language. For example, Japanese user A says, "Hello, how is the project going?"

[1419] Step 2:

[1420] The terminal (User A's device) records the speech using a microphone and captures it as voice data, which is then temporarily stored in the terminal in digital format.

[1421] Step 3:

[1422] The device transmits the collected voice data to a server via the Internet in real time. The transmitted data includes the voice data as well as the user ID and device information.

[1423] Step 4:

[1424] The server inputs the received voice data into the generative AI model and converts it into text using speech recognition technology. This conversion process generates the text, "Hello, how is the project going?"

[1425] Step 5:

[1426] The server uses the voice and text data to recognize the user's emotions through an emotion engine. For example, the server can recognize emotions such as "interest" and "concern" from User A's tone of voice and speaking style.

[1427] Step 6:

[1428] The server completes the generated text using a context analysis algorithm. For example, the text "Hello, how is the project going?" is completed as "Hello, regarding the project progress, what is the status?"

[1429] Step 7:

[1430] The server translates the completed text into the target language and reflects the recognized sentiment in the translation result. For example, when translating the completed Japanese text into English and generating the English text "Hello, regarding the project's progress, how is it going?", the server reflects that User A is "interested" or "concerned."

[1431] Step 8:

[1432] The server converts the translated text into speech data using speech synthesis technology, generating English speech data such as "Hello, regarding the project's progress, how is it going?" with emotional intonation.

[1433] Step 9:

[1434] The server transmits the generated translated speech data via the Internet to the other party's terminal (user B's device). The transmitted data includes the translated speech data.

[1435] Step 10:

[1436] The device (User B's device) plays the received voice data through the device's speaker. User B can hear the English voice "Hello, regarding the project's progress, how is it going?" with the emotion reflected.

[1437] Step 11:

[1438] User B (American) says in English, "The project is progressing well, thank you for asking."

[1439] Step 12:

[1440] The device (User B's device) records the speech with a microphone and captures the audio data in real time, which is then temporarily stored in the device in digital format.

[1441] Step 13:

[1442] The device transmits the collected voice data to a server via the Internet in real time. The transmitted data includes the voice data as well as the user ID and device information.

[1443] Step 14:

[1444] The server inputs the received voice data into the generative AI model and converts it into text using speech recognition technology. For example, the generated text might read, "The project is progressing well, thank you for asking."

[1445] Step 15:

[1446] The server uses the voice and text data to recognize the user's emotions using an emotion engine. For example, emotions such as "gratitude" and "relief" can be recognized from User B's tone of voice and speaking style.

[1447] Step 16:

[1448] The server uses a context analysis algorithm to complete the generated text. For example, the text "The project is progressing well, thank you for asking" is completed as "The project is progressing well, thank you for asking about the project's status."

[1449] Step 17:

[1450] The server translates the completed text into the target language and reflects the recognized emotions in the translation result. For example, when translating the completed English text into Japanese and generating the Japanese text "The project is progressing smoothly. Thank you for your inquiry," the "gratitude" and "relief" that User B feels are reflected.

[1451] Step 18:

[1452] The server then converts the translated Japanese text into speech data using speech synthesis technology, generating Japanese speech data that says, "The project is progressing smoothly. Thank you for your inquiry," with intonation that reflects the emotion.

[1453] Step 19:

[1454] The server transmits the generated translated speech data to the other party's terminal (user A's device) via the Internet. The transmitted data includes the translated speech data.

[1455] Step 20:

[1456] The terminal (user A's device) plays the received voice data through the device's speaker. User A can hear the voice in his / her native language saying "The project is progressing smoothly. Thank you for your inquiry." with the emotion reflected.

[1457] Example 2

[1458] 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."

[1459] In today's global society, people who speak different languages ​​need to communicate smoothly. However, while conventional translation systems can accurately translate spoken content, they struggle to convey the speaker's emotions appropriately. As a result, misunderstandings and misrecognitions are likely to occur, hindering natural communication. Furthermore, they often fail to properly complement the spoken content by taking into account the context of the spoken content, making it impossible to convey the speaker's full intention.

[1460] 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.

[1461] In this invention, the server includes a means for recognizing the user's emotions, a means for completing the content of the converted text, and a means for converting the translated text into voice data. This not only ensures accurate translation of the spoken content, but also enables the speaker's emotions to be conveyed to the other party in a manner that appropriately reflects their intention. Furthermore, context-based content completion allows the speaker to fully convey their intention, resulting in more natural and effective communication.

[1462] "User" refers to a person who uses the system to translate spoken content into another language.

[1463] The term "terminal" refers to an electronic device that collects the user's speech, transmits the voice data to a server, and plays back the voice data transmitted from the server.

[1464] "Server" refers to a central control device that processes voice data sent from a user's terminal and performs translation and voice synthesis.

[1465] "Speech content" refers to words and sentences spoken by a user in their own language.

[1466] "Voice data" refers to data that represents the contents of a user's speech in digital form.

[1467] "Text" refers to data that has been converted from audio data into character information.

[1468] "Emotion" refers to the emotional elements (e.g., interest, concern, joy, etc.) contained in the user's utterance.

[1469] "Content completion" refers to the process of analyzing the context of the spoken content and completing incomplete parts in a natural way.

[1470] "Translation" refers to the process of converting text expressed in one language into another language.

[1471] "Speech synthesis" refers to the process of converting text data into voice data.

[1472] "Emotion recognition" refers to the process of analyzing and identifying emotional elements from a user's speech.

[1473] "Real-time" means that processing occurs with little to no delay.

[1474] The system of the present invention translates what a user says in their own language into another language in real time, and further recognizes the user's emotions and reflects them in the translation, thereby realizing natural and emotional communication. A specific embodiment of this system will be described below.

[1475] Collection of speech content

[1476] The user speaks in their own language. For example, a Japanese user might say, "Hello, how's the project going?" This speech is collected by the user's device (smartphone or computer) via a microphone.

[1477] Sending audio data

[1478] The device transmits the collected voice data to a server over the Internet in real time. The voice data contains the user's speech in digital form. The device transmits the voice data using a secure communication protocol (e.g., HTTPS).

[1479] Speech recognition and text conversion

[1480] The server inputs the voice data received via the internet into a generative AI model and converts the voice into text. For example, the generative AI model used is the Google Cloud Speech-to-Text API, and generates the text "Hello, how is the project progressing?" from Japanese voice data.

[1481] emotion recognition

[1482] The server uses voice and text data to recognize the user's emotions through an emotion engine. For example, the emotion engine used is IBM Watson Tone Analyzer, which recognizes emotions such as "interest" and "concern" from the user's tone of voice and speaking style.

[1483] Content Complement

[1484] The server uses a context analysis algorithm to complete the generated text. For example, the context analysis algorithm used is OpenAI GPT-3, and the text "Hello, how is the project going?" is completed as "Hello, regarding the progress of the project, what is the status?"

[1485] translation

[1486] The server translates the completed text into the target language and reflects the recognized sentiment in the translation results. For example, the translation engine used is the Google Translate API, and the completed Japanese text is translated into English to generate the English text "Hello, regarding the project's progress, how is it going?". In this case, the user's "interest" and "concern" are reflected in the translation results.

[1487] Speech synthesis

[1488] The server converts the translated text into voice data using Amazon Polly speech synthesis technology. The generated voice data is an English voice saying "Hello, regarding the project's progress, how is it going?" with intonation that reflects the emotion.

[1489] Sending and playing audio data

[1490] The generated voice data is again sent from the server to the other device (User B's device). The other device receives this voice data and plays it through the speaker. User B can hear the English voice saying "Hello, regarding the project's progress, how is it going?" with the emotion reflected.

[1491] As a concrete example, consider a conversation between a Japanese user and an American user participating in an international conference. The Japanese user speaks in Japanese, which is played back as English speech on the American user's device, with the Japanese user's emotions also being conveyed. Similarly, the American user speaks in English, which is played back as Japanese speech on the Japanese user's device, with the Japanese user's emotions also being reflected.

[1492] Prompt Sentence Examples

[1493] "A user says, 'Hello, how's the project going?' in Japanese. Please translate this utterance into English and convert it into speech data that reflects the sentiment."

[1494] This system will significantly improve international communication, enabling more natural and emotional interactions.

[1495] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1496] Step 1:

[1497] The user speaks in their own language. The user's voice is collected by the device as input. For example, a Japanese user might say, "Hello, how's the project going?". Specifically, the device's microphone collects the voice in digital form. Digital voice data is generated as output.

[1498] Step 2:

[1499] The device transmits the collected audio data in real time to a server in uncompressed or compressed format. Digital audio data is input. Specifically, the device transmits the audio data using a secure communication protocol such as HTTPS. The audio data arrives at the server as output.

[1500] Step 3:

[1501] The server inputs the received voice data into a generative AI model and converts the voice to text. The voice data received on the server side exists as input. Specifically, it uses the Google Cloud Speech-to-Text API to convert it into text data such as "Hello, how is the project progressing?". Text data is generated as output.

[1502] Step 4:

[1503] The server uses voice and text data to recognize the user's emotions through an emotion engine. Converted text data and voice data are used as input. Specifically, the server uses IBM Watson Tone Analyzer to identify emotions such as "interest" or "concern" from the user's tone of voice and choice of words. Emotional data is generated as output.

[1504] Step 5:

[1505] The server analyzes the context of the generated text data and completes the content. Emotion data and text data are input. Specifically, OpenAI GPT-3 is used to complete the text "Hello, how is the project progressing?" to "Hello, regarding the progress of the project, what is the status?". The completed text data is generated as output.

[1506] Step 6:

[1507] The server translates the completed text into the target language and also reflects the recognized emotion in the translation result. The input is the completed text data and emotion data. Specifically, it uses the Google Translate API to translate the completed text "Hello, regarding the project's progress, how is it going?" into English and reflects the emotion to generate "Hello, regarding the project's progress, how is it going?" The translated text data is generated as the output.

[1508] Step 7:

[1509] The server converts the translated text into speech data using speech synthesis technology. The inputs are the translated text data and emotion data. Specifically, Amazon Polly is used to generate English speech data saying "Hello, regarding the project's progress, how is it going?". This speech data is given an intonation that reflects the emotion. English speech data is generated as output.

[1510] Step 8:

[1511] The server sends the generated voice data to the other party's device. The generated voice data exists as input. Specifically, the voice data is sent to User B's device via the Internet. The voice data arrives at the other party's device as output.

[1512] Step 9:

[1513] The other device plays the received audio data. The input is English audio data from the server. Specifically, the device's speaker plays the English audio, "Hello, regarding the project's progress, how is it going?". User B can hear the audio as output.

[1514] This detailed processing allows users who speak different languages ​​to communicate accurately and with rich emotion.

[1515] (Application example 2)

[1516] 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."

[1517] Currently, in order for autonomous vehicles to communicate smoothly between passengers and vehicle systems, it is necessary not only to recognize instructions but also to communicate seamlessly in various languages. It is also necessary to simultaneously recognize passenger emotions and respond appropriately to those emotions to provide a safer and more comfortable travel experience. However, current systems are unable to effectively support multiple languages ​​and recognize emotions, resulting in reduced passenger satisfaction and safety.

[1518] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting utterances from users, means for transmitting the collected utterances as voice data, means for converting the received voice data into text, means for completing the converted text, means for translating the completed text into another language, means for converting the translated text into voice data, means for transmitting the converted voice data to the other party's terminal, means for performing emotion recognition, means for generating voice data based on the recognized emotion, means for playing the received voice data at the other party's terminal, means for giving instructions to the vehicle system or setting a destination based on the passenger's utterance, and means for recognizing the emotion of the utterance and performing a corresponding action. This enables natural communication that recognizes emotions in multiple languages, thereby realizing safe and comfortable operation of autonomous vehicles.

[1519] The "means for collecting speech content from the user" refers to a device or function for recognizing the speech uttered by the user and acquiring it as digital data.

[1520] "Means for transmitting collected speech content as voice data" refers to a mechanism for sending acquired voice data to a server via a communication network such as the Internet.

[1521] The "means for converting received voice data into text" refers to a system that analyzes voice data on the server side and converts it into a corresponding text format.

[1522] "Means for completing the content of the converted text" refers to technologies and algorithms that analyze the context of the converted text, add information as needed, and arrange the sentences in a natural format.

[1523] The "means for translating the completed text into another language" is an automatic translation technology for converting the completed text into another specified language.

[1524] "Means for converting translated text into speech data" refers to technology or devices for converting translated text into synthetic speech.

[1525] The "means for transmitting the converted voice data to the other terminal" is the communication means used to retransmit the generated voice data to the target device.

[1526] "Means for emotion recognition" refers to technologies and algorithms that analyze and recognize a speaker's emotional state from audio data or text data.

[1527] The "means for generating voice data based on recognized emotions" refers to a technology that adjusts the intonation and atmosphere of the synthesized voice by reflecting the recognized emotional information.

[1528] The "means for playing back the received audio data on the other party's terminal" is a function for playing back the transmitted audio data on the device.

[1529] "Means for giving instructions to vehicle systems and setting destinations based on passenger speech" refers to technology that recognizes what passengers say and operates the vehicle's navigation system and other operations based on that speech.

[1530] "Means for recognizing emotions in speech and taking corresponding actions" refers to technologies and functions that allow the system to automatically respond or take appropriate actions based on the recognized emotional information.

[1531] DETAILED DESCRIPTION OF THE INVENTION The present invention provides a voice interface for autonomous vehicles with multilingual support and emotion recognition capabilities.

[1532] Hardware and Software Configuration

[1533] 1. Hardware:

[1534] Microphone: Used to collect passenger speech.

[1535] Speaker: Used to play back the generated audio data.

[1536] A smartphone or in-vehicle computer: Used to process the voice data.

[1537] 2. Software:

[1538] Speech Recognition Library: speech_recognition

[1539] Translation library: googletrans

[1540] Emotion Recognition Model: A sentiment analysis model using the transformers library

[1541] Text-to-speech libraries: gTTS and playsound

[1542] Data processing and calculation flow

[1543] 1. Voice input: Using the device's microphone, we collect speech in Japanese uttered by the user. For example, a passenger might say, "Please go to Shinjuku Station."

[1544] 2. Speech Recognition: Using the speech_recognition library, the device converts the collected voice data into text format.

[1545] 3. Translation: Use googletrans library to translate the converted Japanese text into English. For example, "Please take me to Shinjuku Station" is translated to "Please take me to Shinjuku Station".

[1546] 4. Emotion Recognition: Use the transformers library to recognize user emotions from translated text, for example, "tired" from the tone and content of the speech.

[1547] 5. Emotion-sensitive speech synthesis: Using the gTTS library, the translated text is converted into synthetic speech, and an intonation is added to reflect the recognized emotion, for example, "Please take me to Shinjuku Station" in a tired voice.

[1548] 6. Audio output: The generated audio data is sent to the other device and played through the speaker. Based on the passenger's request, the navigation system sets the destination and the autonomous vehicle begins operation.

[1549] Specific examples

[1550] For example, if a passenger says in Japanese, "I'm a little tired, so please go directly to the airport," the system will act as follows:

[1551] 1. The microphone collects the speech and the device performs voice recognition.

[1552] 2. Translate the recognized text "I'm a bit tired, please take me directly to the airport." using the googletrans library to get "I'm a bit tired, please take me directly to the airport."

[1553] 3. Emotion recognition is performed on the translated text to obtain the emotion label "Tired."

[1554] 4. The speech that reflects the recognized emotion "tired" is synthesized using the gTTS library to generate speech in a format that conveys the emotion in English.

[1555] 5. The generated audio is played through the speaker, the navigation system sets the destination according to the instructions, and the vehicle begins to travel along the route.

[1556] Prompt Sentence Examples

[1557] The user says in Japanese, "I'm a little tired, so please go directly to the airport." Translate this into English in real time and output the following voice output, reflecting the user's emotions.

[1558] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1559] Step 1:

[1560] The user performs voice input. The user utters "Please go to Shinjuku Station" in Japanese. The contents of this utterance are collected through the device's microphone.

[1561] Step 2:

[1562] The device converts the collected voice data into text format in real time using the speech recognition library speech_recognition. The input is voice data, and the output is text data: "Please go to Shinjuku Station."

[1563] Step 3:

[1564] After the text data is generated, the device translates this text into the target language (in this case, English) using the translation library googletrans. The input is the text data "Please take me to Shinjuku Station," and the output is the English text "Please take me to Shinjuku Station."

[1565] Step 4:

[1566] The server inputs the translated English text into an emotion recognition model (using the transformers library) to recognize the user's emotion. The input is the English text "Please take me to Shinjuku Station", and the output is an emotion label (e.g., "Tired").

[1567] Step 5:

[1568] The server uses the speech synthesis library gTTS to generate speech data reflecting the recognized emotions. The input is English text and emotion labels, and the output is speech data reflecting the emotions.

[1569] Step 6:

[1570] The device sends the generated voice data to the navigation system, which then uses it to set destinations and plan routes for the autonomous vehicle. The input is the voice data, and the output is the vehicle's instructed destination and route.

[1571] Step 7:

[1572] Finally, the terminal plays the generated voice data through a speaker to provide voice feedback to the passenger, facilitating user communication with the autonomous vehicle system.

[1573] 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.

[1574] 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.

[1575] 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.

[1576] 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.

[1577] 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.

[1578] 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.

[1579] 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).

[1580] 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.

[1581] 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."

[1582] 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.

[1583] 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).

[1584] 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.

[1585] 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.

[1586] 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.

[1587] 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.

[1588] 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.

[1589] 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.

[1590] 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.

[1591] 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.

[1592] 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.

[1593] 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.

[1594] The following is further disclosed regarding the above embodiment.

[1595] (Claim 1)

[1596] A means for collecting utterances from users;

[1597] A means for transmitting the collected speech content as voice data;

[1598] means for converting the received voice data into text;

[1599] a means for completing the content of the converted text;

[1600] A means of translating the completed text into other languages;

[1601] means for converting the translated text into audio data;

[1602] A means for transmitting the converted voice data to the other terminal;

[1603] A system including means for playing back the received audio data at the other party's terminal.

[1604] (Claim 2)

[1605] 2. The system according to claim 1, wherein the system translates user utterances between different languages ​​in real time.

[1606] (Claim 3)

[1607] 2. The system according to claim 1, wherein when completing the content of an utterance, information is completed based on the context.

[1608] (Claim 4)

[1609] 10. The system of claim 1, wherein the system communicates with a server via the Internet to process the voice data transmitted from the terminal.

[1610] "Example 1"

[1611] (Claim 1)

[1612] A means for collecting utterances from users;

[1613] A means for transmitting the collected speech content as voice data;

[1614] means for converting the received voice data into text;

[1615] A means for complementing the content of the converted text using a generative AI model; and

[1616] A means of translating the completed text into other languages;

[1617] means for converting the translated text into audio data;

[1618] means for transmitting the generated voice data to the other device;

[1619] A system including means for playing back the received audio data on the other device.

[1620] (Claim 2)

[1621] 2. The system according to claim 1, wherein the system translates user utterances between different languages ​​in real time.

[1622] (Claim 3)

[1623] The system of claim 1, wherein when completing the content of an utterance, the system uses a generative AI model to complete information based on context.

[1624] "Application Example 1"

[1625] (Claim 1)

[1626] A means for collecting utterances from users;

[1627] A means for transmitting the collected speech content as voice data;

[1628] means for converting the received voice data into text;

[1629] a means for completing the content of the converted text;

[1630] A means of translating the completed text into other languages;

[1631] means for converting the translated text into audio data;

[1632] A means for transmitting the converted voice data to the other terminal;

[1633] A means for playing back the received audio data at the other party's terminal;

[1634] A means for realizing real-time language translation between service providers and customers and providing audio and visual information through smart glasses;

[1635] The system includes a means for detecting a user's speech using a microphone and speaker in the smart glasses and providing audio feedback to the other party in the translated language.

[1636] (Claim 2)

[1637] 2. The system according to claim 1, wherein the system translates user utterances between different languages ​​in real time.

[1638] (Claim 3)

[1639] 2. The system according to claim 1, wherein when completing the content of an utterance, information is completed based on the context.

[1640] "Example 2: Combining Emotion Engines"

[1641] (Claim 1)

[1642] A means for collecting utterances from users;

[1643] A means for transmitting the collected speech content as voice data;

[1644] means for converting the received voice data into text;

[1645] means for recognizing a user's emotion;

[1646] a means for completing the content of the converted text;

[1647] A means of translating the completed text into other languages;

[1648] means for converting the translated text into audio data;

[1649] A means for transmitting the converted voice data to the other terminal;

[1650] A system including means for playing back the received audio data at the other party's terminal.

[1651] (Claim 2)

[1652] 2. The system according to claim 1, wherein the system translates user utterances between different languages ​​in real time.

[1653] (Claim 3)

[1654] 2. The system according to claim 1, wherein when completing the content of an utterance, information is completed based on the context.

[1655] (Claim 4)

[1656] 10. The system of claim 1, wherein the system recognizes a user's emotions and reflects the emotions in the translation and speech data.

[1657] "Application example 2 when combining emotion engines"

[1658] (Claim 1)

[1659] A means for collecting utterances from users;

[1660] A means for transmitting the collected speech content as voice data;

[1661] means for converting the received voice data into text;

[1662] a means for completing the content of the converted text;

[1663] A means of translating the completed text into other languages;

[1664] means for converting the translated text into audio data;

[1665] A means for transmitting the converted voice data to the other terminal;

[1666] a means for performing emotion recognition;

[1667] means for generating voice data based on the recognized emotion;

[1668] A system including means for playing back the received audio data at the other party's terminal.

[1669] (Claim 2)

[1670] 2. The system according to claim 1, wherein the system translates user utterances between different languages ​​in real time.

[1671] (Claim 3)

[1672] 2. The system according to claim 1, wherein when completing the content of an utterance, information is completed based on the context.

[1673] (Claim 4)

[1674] 2. The system according to claim 1, wherein instructions are given to the vehicle system and destinations are set based on passenger speech.

[1675] (Claim 5)

[1676] 10. The system of claim 1, wherein the system recognizes emotions in an utterance and performs corresponding actions. [Explanation of symbols]

[1677] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for collecting utterances from users; A means for transmitting the collected speech content as voice data; means for converting the received voice data into text; a means for completing the content of the converted text; A means of translating the completed text into other languages; means for converting the translated text into audio data; A means for transmitting the converted voice data to the other terminal; A system including means for playing back the received audio data at the other party's terminal.

2. 2. The system according to claim 1, wherein the system translates user utterances between different languages ​​in real time.

3. The system according to claim 1, wherein when completing the content of an utterance, information completion is performed based on a context.

4. 10. The system of claim 1, wherein the system communicates with a server via the Internet to process the voice data transmitted from the terminal.

Images